Lean Manufacturing

Mr.
Lean Buys and

Transforms a
Manufacturing Company
The True Story of Proftably
Growing an Organization
with Lean Principles
Mr. Lean Buys and
Transforms a
Manufacturing Company
The True Story of Proftably
Growing an Organization
with Lean Principles
GreG lane
Foreword by John Shook
Productivity Press
Taylor & Francis Group
270 Madison Avenue
New York, NY 10016
2010 by Taylor and Francis Group, LLC
Productivity Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works
Printed in the United States of America on acid-free paper
10 9 8 7 6 5 4 3 2 1
International Standard Book Number: 978-1-4398-1516-8 (Hardback)
This book contains information obtained from authentic and highly regarded sources. Reasonable efforts
have been made to publish reliable data and information, but the author and publisher cannot assume
responsibility for the validity of all materials or the consequences of their use. The authors and publishers
have attempted to trace the copyright holders of all material reproduced in this publication and apologize
to copyright holders if permission to publish in this form has not been obtained. If any copyright material
has not been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, trans-
mitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter
invented, including photocopying, microfilming, and recording, or in any information storage or retrieval
system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.
com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood
Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and
registration for a variety of users. For organizations that have been granted a photocopy license by the
CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation without intent to infringe.
Library of Congress CataloginginPublication Data
Lane, Greg.
Mr. Lean buys and transforms a manufacturing company : the true story of
profitably growing an organization with lean principles / Greg Lane.
p. cm.
Includes bibliographical references and index.
ISBN 978-1-4398-1516-8
1. Semiconductor production equipment industry--Management. 2. Semiconductor
production equipment industry--Cost effectiveness. 3. Production control--Cost
effectiveness. 4. Organizational effectiveness. 5. Lane, Greg. I. Title.
HD9696.S42L36 2010
658.56--dc22 2009045800
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the Productivity Press Web site at
http://www.productivitypress.com
v
Contents
Foreword ............................................................................................. xiii
Preface ................................................................................................... xv
Prologue: The Purchase....................................................................... xix
1 Chapter The Beginning
Observation and Documentation ...................................... 1
1.1 Introduction ......................................................................1
1.2 Observing ..........................................................................2
1.3 Getting a Detailed Understanding of the
Processes I Would Be Responsible For ..........................5
1.4 A Lack of Training Can Be Costly .................................6
1.5 Training as a Machine Operator ....................................8
1.6 Discovering a Lack of Documentation ..........................8
1.7 Identifying Skill Sets Tat Needed to Be Replaced ......9
1.8 Workplace Organization (5S) Opportunities .............10
1.9 Discovering Diferent Methods Being Utilized
for the Same Task (No Standardization) .....................12
1.10 Discovering Difcult Processes to Standardize .........13
1.11 Tis Much Observation Required Me to Divide
and Conquer ..................................................................14
1.12 Summary..........................................................................16
2 Chapter Standardizing the Estimating Process ............................ 17
2.1 Hiring a New Programmer/Estimator ........................17
2.2 Writing the First Draf of Standardized Steps
for Programming and Estimating ................................18
2.3 Considering the Complications with Quoting ...........19
2.4 Minimize Training and Start-Up Time
by Utilizing Similar Equipment .................................. 20
2.5 Continuing to Determine Where Skill Sets
Would Be Lost .................................................................21
2.6 Meeting the Customers ..................................................22
vi Contents
2.7 Starting Some Measurements .......................................24
2.8 Visualizing the Measurement .......................................25
2.9 Plan, Do, Check, Act (PDCA) .......................................26
2.10 Summary......................................................................... 28
3 Chapter Learning the Office Processes ......................................... 29
3.1 Introduction ....................................................................29
3.2 Inputting Orders ............................................................ 30
3.3 Problems Working without a PDCA Cycle
in the Estimating Process ..............................................32
3.4 Repricing Discussions Are Easier if Based
on Accurate Data ............................................................33
3.5 Learning to Purchase .................................................... 34
3.6 Scheduling the Shop Based on Tribal Knowledge .....36
3.7 Job-Shop Planning ..........................................................37
3.8 Simplifed Short-Term Planning ...................................38
3.9 A More Comprehensive Work Order Package ...........39
3.10 Poor Work Order Packages Can Increase Setup
Time ..................................................................................41
3.11 Closing the Books ...........................................................41
3.12 Collecting the Money .................................................... 43
3.13 Summary......................................................................... 44
4 Chapter The First Days Alone ....................................................... 47
4.1 Introduction ....................................................................47
4.2 My First Pricing Decision Alone ................................. 48
4.3 Creating Extra Capacity with 5S ..................................49
4.4 Establishing Normal versus Abnormal .......................52
4.5 Problem Solving versus Firefghting ............................55
4.6 Identifying the Need for Shop-Floor Visuals .............55
4.7 Converting Small Productivity Improvements
into Proft .........................................................................58
4.8 Challenging the Reported Proft Margins.................. 60
4.9 Summary..........................................................................62
Contents vii
5 Chapter Creating and Selling Capacity ......................................... 63
5.1 Only Tinking in Terms of Full Absorption
Costing .............................................................................63
5.2 Overtime as a Positive ................................................... 64
5.3 Setup-Time Reductions ..................................................65
5.4 Continuously Measuring Setup Times ........................67
5.5 Creating a Team Leader .................................................70
5.6 Increased Productivity through Visual
Management ....................................................................71
5.7 PDCA of the Estimating Process ..................................73
5.8 Becoming the Estimator ................................................74
5.9 Summary..........................................................................77
6 Chapter Getting Everyone Motivated for Improving the
Machines Output ............................................................. 79
6.1 Introduction ....................................................................79
6.2 Managing Each Job Visually ........................................ 80
6.3 Te Desire for Independence from Day-to-Day
Operations .......................................................................81
6.4 Preparing for Further Setup Reductions .....................82
6.5 Continuing the Kaizen of Equipment by
Improving the Adjustment Phase of Changeovers ....83
6.6 Running a Machine while Confrming the
Quality of the First Piece .............................................. 84
6.7 Difculties in Creating a Proft-Sharing Plan ............ 86
6.8 Staying Hands-On since the Devil Is in the
Details ............................................................................ 88
6.9 Cost Justifying a Team Leader ..................................... 90
6.10 Closing the Books Becomes Anticlimactical ..............93
6.11 Summary..........................................................................94
7 Chapter Seeing the Administrative Processes from a New
Angle ................................................................................. 95
7.1 Introduction ....................................................................95
7.2 Visualization Confronts Obtaining Information
Either from Computers or in Meetings ...................... 96
viii Contents
7.3 Not Allowing Excuses for a Late Start .........................97
7.4 Small Businesses Look at the Recruiting Process
Diferently ........................................................................98
7.5 Focusing to Improve the Ofce Processes ............... 100
7.6 Purchasing Is a Critical Administrative
Function to Look toward Improving .........................102
7.7 Unsuccessful at Signifcantly Reducing the Time
Necessary for Quotations ............................................107
7.8 5S for Shared Computer Drives and Files .................109
7.9 Outside Quality Accreditation ....................................110
7.10 Summary........................................................................ 111
8 Chapter Moving the Business ...................................................... 113
8.1 Industrywide Downturn .............................................113
8.2 Background for Moving the Business at Tis
Point in Time ................................................................. 114
8.3 Creating a New Layout from a Blank Canvas ........... 116
8.4 A Nondisruptive Move ................................................. 118
8.5 Simulating the New Layout ........................................ 120
8.6 Introducing Procedural Changes during the
Move ...............................................................................121
8.7 Using a Major Event to Introduce Change in the
Ofce.............................................................................. 123
8.8 Summary........................................................................125
9 Chapter Accurate Pricing through Better Cost Allocations ...... 127
9.1 Introduction ..................................................................127
9.2 Determining More Accurate Allocations during
Tough Times ..................................................................129
9.3 Using My Modifed Activity-Based Costing .........130
9.4 Starting Simplifed Activity-Based Costing
in the Shop Using a Matrix .........................................132
9.5 How to Relate the Matrix to Cost .............................. 134
9.6 A Simple Activity-Based Costing Matrix
for the Ofce ..................................................................135
9.7 Relating the Matrices to Hourly Costs ......................136
Contents ix
9.8 How Others Ofen Allocate Costs ..............................138
9.9 Keeping the Team Leader Focused during the
Slow Times .....................................................................140
9.10 Increasing Process Capabilities to Reduce
Outsourcing ...................................................................141
9.11 Standardizing the Business Processes Allows
Time for Pursuing Other Opportunities ...................142
9.12 Summary........................................................................144
1 Chapter 0 Up-Front Delays ............................................................. 145
10.1 Te Measurements Confrm Success .........................145
10.2 Continuously Developing Protocols Allows
More Autonomy from Day-to-Day Operations ........146
10.3 Planning Only Based on Available Materials Is
Costly ..............................................................................147
10.4 Capacity Planning Based on Lead Times ..................148
10.5 Identifying Areas Where We Were Only Being
Reactive ..........................................................................150
10.6 Available Time Utilized to Document Processes .....150
10.7 Lead Time Reduced through Strategic
Inventory ........................................................................152
10.8 Prioritizing Problems during Busy Periods ............. 154
10.9 Te Lack of Hands-On Experience Is Very Costly .... 154
10.10 My Personal Experience of Generating Costly
Designs ...........................................................................157
10.11 Summary........................................................................158
1 Chapter 1 Making Money during the Good Times ....................... 161
11.1 Introduction ..................................................................161
11.2 Improving Processes during Slow Periods Pays
Of ...................................................................................161
11.3 Problem-Solving Skills Developed during the
Slow Periods Begin to Pay Of .....................................163
11.4 Te Leader Keeps in Practice ......................................164
11.5 Diversifcation Would Have Been Helpful ................166
11.6 Te Need for a Second Shif ........................................166
x Contents
11.7 Te Entrepreneurial Spirit Pays Further
Dividends during the Busy Times ..............................169
11.8 Structured Problem Solving Minimizes
Opinions ........................................................................170
11.9 Problem Solving Requires Looking into the
Process Details Using a Structure ..............................173
11.10 Equality in Enforcing Your Own Policies Is
Critical ............................................................................175
11.11 Summary........................................................................175
1 Chapter 2 Expanding into New Products ...................................... 177
12.1 Introduction ..................................................................177
12.2 Te Logic behind Expanding into Sheet Metal
Fabrications ...................................................................177
12.3 Previous Expansion Considerations Resulted
in Little Impact to the Current Layout ......................179
12.4 Filling the First Order ..................................................183
12.5 Investing Step by Step ..................................................185
12.6 OSKKK Takes to Life within Our New Processes ...187
12.7 Once Again, a PDCA Cycle in the Quoting
Process Proves Critical .................................................188
12.8 Again, Engineers Lack of Hands-On Experience
Is Costly ......................................................................... 190
12.9 Summary........................................................................192
1 Chapter 3 Business for Sale ............................................................. 193
13.1 Introduction ..................................................................193
13.2 Determining the Process Steps for Selling
a Business .......................................................................194
13.3 Determining an Asking Price .....................................195
13.4 A Marketing Package ...................................................196
13.5 Advertising ....................................................................197
13.6 Te First Nibbles ...........................................................197
13.7 Considerations of Owner Financing ..........................198
13.8 Lowering the Price ........................................................199
13.9 Tree Serious Buyers ................................................... 200
Contents xi
13.10 Evaluating Ofers ..........................................................201
13.11 Ofer Accepted ............................................................. 202
13.12 Drawing to an End ...................................................... 204
13.13 Summary....................................................................... 205
1 Chapter 4 Ref lections ...................................................................... 207
14.1 Introduction ................................................................. 207
14.2 Lessons Learned ........................................................... 207
14.3 Using the Correct Measurements Is Critical
to Achieving Results .....................................................210
14.4 Cross-Training Is a Critical Part of Success
in High-Variation Companies ....................................211
14.5 Team Leaders Increase Proft through Higher
OEE .................................................................................211
14.6 Managing Parts Diferently .........................................212
14.7 Activity-Based Costing Proves Helpful
in Securing More Orders .............................................213
14.8 A Process to Examine Your Business ........................215
Appendix A OSKKK Methodology .................................................. 219
Appendix B Documentation Techniques ......................................... 223
Appendix C My First Attempt at Standardizing the Quoting
Process .......................................................................... 227
Appendix D Example of the First Day-by-Hour Board for the
Bottleneck Machine ..................................................... 229
Appendix E Real-Time Pareto Chart ............................................... 231
Appendix F Overall Equipment Effectiveness (OEE) for
Bottlenecks .................................................................... 233
Appendix G Typical Team Leader Responsibilities ........................ 237
Appendix H Cross-Training or Skills Matrix .................................. 239
Appendix I New Layout with Some CNCs Set Up as an L Cell ... 241
Appendix J Initial ABC Costing Matrix for Shop and Office ........ 243
xii Contents
Appendix K Initial Equipment and Area Layout for Expansion
into Sheet Metal............................................................ 247
Appendix L Sequence Utilized for Learning the Business and
for Implementing Lean ................................................. 249
Appendix M Different Applications of Value Stream Mapping
and Process Mapping .................................................. 251
Index .................................................................................................... 253
About the Author ................................................................................ 269
xiii
Foreword
Tere has never been a lean consultant who didnt consider buying his
own company, to rid himself of slow-minded company managers who
dont listen, to have a chance to do it the way it should be done, and to do
it himself. Tat is, to fnally do it right.
And there has never been a business owner who hasnt wished all the
know-it-all consultants who lecture them on the obvious would put up or
shut up: Tose who can, do; those who cant either teach or consult.
Mr. Lean Buys and Transforms a Manufacturing Company is for both
camps the consultant who secretly wonders what it would be like to
actually lead the charge as CEO/owner and the CEO owner who simply
needs the performance improvement.
Greg Lane has seen lean from all sides: as a lean specialist inside large
corporations, as an independent lean consultant, and as a company owner.
Greg learned the Toyota Production System directly from the source dur-
ing his stint at New United Motor Manufacturing Inc. (NUMMI), the
joint venture between Toyota and GM.
As owner of his own machining company, purchased on a shoestring
and a prayer, Greg naturally reached deep into his bag of lean tricks. But,
his company was small with few resources. He quickly realized every
action he took had to be supremely practical. Tere could be no lean for
leans sake. No kanban where they didnt make sense. No overblown,
expensive andon systems. No extensive team of indirect employees who
add no value.
As his organizing principle, Greg landed on a lesser-known Toyota
approach to creating change in a workplace called OSKKK. OSKKK
stands for Observe deeply, then Standardize, followed by three stages of
Kaizen. Kaizen is a lesser-known Toyota approach.
Greg frst learned about OSKKK during his intense TPS Key Person
training in Toyota City, Japan in the early 1990s. One of the fortunate,
hand-picked few to undergo this specialized training, Greg learned
directly at the hands of Toyotas most experienced Toyota Production
Sensei in Toyota City. Later, as a lean consultant to his former employer
Delphi, Greg was more deeply exposed to OSKKK through the work of
xiv Foreword
former Toyota production control manager Yoshinobu Yamada, who was
introducing OSKKK throughout many of the operational divisions of
Delphi at that time.
Greg takes us blow-by-blow through the entire experience of search-
ing for an appropriate company, completing the purchase, learning a
new business, transforming it through the application of lean thinking
and achieving a two-fold increase in sales during the frst four years (an
increase in capacity and real output of 25 percent per year), including a
proft margin increase of 10 percent and fnally selling the transformed
business to a new, lucky, owner. Along the way, we learn about the applica-
tion of OSKKK, the process of learning and improving not only the plant
foor but also the ofce side of a small business. Author/business owner
Greg even introduces a novel and cleverly simple approach to practical
product costing for quicker and more accurate price quoting for new busi-
ness, a little-discussed headache for every manufacturer.
Mr. Lean Buys and Transforms a Manufacturing Company is a fun read.
If youve ever considered how you would go about transforming your own
manufacturing enterprise or if you own such an enterprise that needs
transformation this book will defnitely provide thought-provoking
ideas and some guiding principles along with specifc hints for dealing
with challenges that your company faces today.
John Shook
President, Lean Transformations Group, Inc.
Ann Arbor, MI, USA
xv
Preface
I am sure there are days when you sit at work thinking how thingsall
those things you imagine doing diferently than the way they are currently
donewould be diferent if you owned and ran the company. Maybe you
imagine being at the top of the pecking orderor at least being able to
advise your managersto create a more efcient company instead of
trudging along performing tasks that add no real value. How easily you
could implement lean ideas to better the business if you were given the
knowledge and authority. It is likely, however, that you have not been pro-
moted to a high-enough level or your personal checking account does not
contain enough money to buy the company (and if you had that kind of
cash, you would probably retire to some wonderful place instead).
I made the jump from corporate employee to business owner by com-
mitting every penny I had as well as incurring heavy debt. Buying a busi-
ness is not something I would recommend to everyone, especially those
with a weak heart or a desire for stability via a regular paycheck.* But,
you can learn a lifetime of lessons in the short term with this approach.
In my case, I was able to become the owner, the boss, and periodically just
another of the machine operators. I took the plunge, and my story follows.
It was not a perfect transition by any stretch of the imagination and many
mistakes were made, but there is a lot to be learned by these experiences.
Tis book tells the true story of how I purchased a stable, proftable
manufacturing company anddespite having no prior experience in the
industryI was able to grow it successfully and increase its proftability
just by showing up and applying a bit of common sense and the lean ideas,
strategies, and philosophies I had learned.
Tese same ideas should be

*

I would not recommend this course of action to others, especially if it involves selling of all your
personal assets and assuming a lot of debt to purchase mostly goodwill, as was my experience.
Refecting on it now that my situation has changed and I am supporting a family, I am not sure I
would risk as much the next time around.

I will not mention the name of the business because it continues to operate under the same name and
the current owners might not want this type of publicity. When I purchased the business, I chose to
keep the original name because, as I quickly found out, the purchase includes buying the goodwill,
meaning the name, the existing customers, the employees knowledge, and any proprietary technol-
ogy or processes. To change the nameespecially to incorporate ones own name into itseemed to
me to be based more on ego than on a sound business decision (unless you have a really cool name).
xvi Preface
applicable to most industries. My experience should help you to see how
lean methods can lead to proftability in almost any manufacturing or
service business, particularly a complex company with a high variation of
products or services.
One item you will not fnd discussed in great detail is management cul-
ture and that is because I did not need to convince top management to
change their culture in the direction of lean management. I came into the
business with strong beliefs and experiences in lean. Ofen the more dif-
fcult part of these transformations is getting the top managers to lead by
example; however, since these principals were already deeply ingrained in
my style, it made the transformation much easier.
My purpose was not to implement the Toyota Production System (or lean
principles) but instead to grow the business as a more proftable organiza-
tion utilizing these tools. I did not explain or sell each idea to the employees
as a lean principle; I merely implemented lean philosophies in their sim-
plest and most logical fashion. Following a simple implementation method
I began to learn while working with Toyota, a method known as OSKKK
(Observe, Standardize, Kaizen Flow & Process, Kaizen Equipment, and
Kaizen Layout), I was able to apply the steps in my transformation although
outside infuences periodically changed the order in which they were imple-
mented. Te OSKKK method was really reinforced later while I was work-
ing with Delphi Automotive during a period in which they utilized Mr.
Yamada, a former Toyota production control manager, to lead and teach
them continuous improvement.
Keep in mind that two things were taking place during this period in
the businesss history:
I was trying to learn all aspects of the business so it could be man-
aged successfully.
I was implementing a lean transformation only afer understanding
basic aspects of the business.
It is critical to keep these things in mind because in the beginning a
manager must patiently learn before he or she can successfully guide the
improvements. Many managers overlook the importance of having an
understanding of the tasks before trying to implement improvements.
OSKKK is a methodology that helps both in learning the business and in
Preface xvii
guiding a lean transformation. All the improvements really come together
as a system, and cherry-picking can lead to suboptimizationit will not
maximize profts nor create a coherent business. If you require a simple,
overriding transformation strategy, then I recommend OSKKK.
Most business owners have acquired experience in the industry prior
to becoming owners or managers. Few assume that lean alone could be
the answer to making that jump from employee to employer; however, the
following account dispels this idea by showing that lean by itself (with its
relationship to the customers desires) goes a long way toward running
a company and increasing its proftability. It also demonstrates that lean
principles are applicable in a made-to-order or job-shop business.
About the book
It has been several years since I sold the business and I tell the story as well
as my memory permits. Naturally, some experiences live in great detail in
my mind while others are a bit foggy.
Shortly afer buying the business, I was contacted about providing addi-
tional support in my former employers lean transformation and I quickly
learned that relating my own transformation experiences could beneft
others. I started keeping some specifc notes, which enhance the accuracy
of this book. Over the years, I have ofen found that relating these real-life
experiences has helped those wishing to better come to terms with mak-
ing their own improvements. Tere was nothing easy about my experi-
ences, but having a plan and sound, simple, and proven principles that I
believed in proved to be a winning combination.
Although I named the chapters according to the overriding theme dur-
ing each phase, the book is written chronologically; therefore, some points
begin in a chapter and are developed in further detail later.
To help you in relating to my planned implementation steps and the tools
I was focusing on, I have included lean principle boxes and key points
highlighted throughout the text, and lessons learned at the end of each
chapter. Te appendices also contain examples of how I worked through
some of these lean elements, and Appendix L captures what fnally ended
up being my implementation sequence.
xviii Preface
Tis is not a technical book and it does not play lean bingo by throw-
ing a lot of acronyms at you to try and impress. However, I do ofen use a
few unique terms that are worth clarifying at this point:
Tribal knowledge is when unique knowledge (e.g., the steps to com-
plete a process, or some technical knowledge) is kept and man-
aged inside someones head instead of being generally agreed-to
and documented.
CNCComputer Numerical Control is a system that reads G-code
and M-code commands and drives a machine tool (in this case, three-
and four-axis machines for cutting metal and other materials).
TPSToyota Production System is Toyotas famous production sys-
tem that I learned and used as the basis for improving my business
though I did not continuously mention it by name or utilize much of
its terminology during the transformation.
PDCAPlan, Do, Check, Act ; also known as management by fact
creates a process-centered environment that continuously follows a
loop that involves (1) studying the current situation, collecting and
analyzing the data to identify causes of problems, and then planning
for improvement (Plan); (2) implementing the plan (Do); (3) moni-
toring to determine whether its successful (Check); and (4) imple-
menting a permanent solution (Act).
Te List is a simple method to gather data and speak with the facts.
Normally, an employee keeps a simple data sheet in the work place
for a few weeks or months and records the occurrences (i.e., insert
artwork ).
As far as I know, I am the only lean practitioner who has purchased and
managed his own manufacturing company armed only with lean experi-
ence. I hope these simple fundamentals can inspire you to formulate a
method to drive continuous improvement in your business. My experi-
ences proved beyond a doubt that lean principles will lead to successful
business management even in low-volume, high-mix companies. Waste
can be found and eliminated in most processes if you have eyes for waste
and the conviction and methods to minimize it. I think my experiences
can highlight and simplify this.
xix
Prologue: The Purchase
It was 1997 when I became serious about buying my own business. I was
working for Delphi Automotive (owned by General Motors at the time,
but later separated from GM and became a publicly owned corporation).
I was Delphi Interiors manager/sensei of international lean production
implementation. I had just returned to the United States from Zurich,
Switzerland, where I had worked for about four years, directing and
implementing lean all over Europe in GMs Opel brand car and compo-
nent plants. I was in my early thirties andat GMs highest management
levelI was on the fast track to being made a director and entering the
executive rank.
I was not frustrated by my job nor did I feel like I was stuck in a dead
end; it was quite the opposite. I was convinced that once I settled into the
perks of executive life, I would likely start to thrive on the responsibility,
the respect, and those wonderful benefts that came with it. But, I had
started to get the itch to try something on my own, and advancement at
Delphi would likely lead to putting aside my entrepreneurial desires. I also
fgured that since I had arrived at this point once in my career, I could do
it againassuming that corporate America would consider a career inter-
ruption as an acceptable diversion and would provide me another chance
to rejoin the ranks if things did not work out.
Being single and having no debt, I fgured this was a great time to scratch
that entrepreneurial itch and give it a try. In retrospect, it is funny that the
only person I knew who did something similar was my father, though he
did it at a much later stage in his life. His decision carried a lot more risk,
like having the responsibility of a family and the realization that, if things
did not work out, it might be difcult to return to corporate life.
Being a company man my entire career, I was not even sure I knew how
to spell entrepreneur, let alone grasp what it took to become one. With
my life savings in hand and a very specialized lean production background
from Toyota, I started by purchasing my frst copy of Entrepreneur maga-
zine. I found that, other than articles written by successful entrepreneurs,
the magazine I purchased contained many ads to buy into franchises. Te
franchises were mostly in the service or food industry, which did ft my
xx Prologue:TePurchase
vision of owning an existing business that manufactured a product. If
asked Why not start your own business instead of buying an existing
one? my answer was simple: I had no ideas for a new business. Instead, I
had Kaizen eyes and felt very comfortable identifying waste, especially
in manufacturing environments. So, why not buy a going concern and
improve it? My undergraduate degree was in mechanical engineering, so
acquiring a company that involved a technical product was desirable.
Afer Entrepreneur magazine did not pan out, I moved into looking
through the business for sale section of local newspapers in northern
California, my target area. I also enlisted some business brokersthe
equivalent of real estate agents except they sell existing businesses.
Obviously, the biggest limitation I faced was price. My savings totaled
$200K afer selling my homeactually, my boat, which I had lived on for
four years before moving to Switzerland to work for GM. In reality, I might
have had more money to work with had I previously owned a home, which
would have appreciated in value, instead of a boat, which continuously
depreciates.* On top of my $200K, my father was also willing to invest
a bit and act as an advisor. Between us, there was about $200K to invest
directly, and another $75K in cash to operate for a month or so until some
of the receivables could be collected.
Certain banks work with the SBA (Small Business Administration) to
partially underwrite business loans if certain criteria are met. Te criteria
were not difcult and I qualifed under SBAs terms. Afer fnding a busi-
ness, I also had to qualify under the banks terms for fnancing.
I learned
most banks working with the SBA like to see an applicant put in about 30
percent as a down payment, so I fgured I could borrow between $400K
and $500K.
With my criteria for choosing a business loosely defned as an exist-
ing manufacturing business that was located somewhere in California
and did not cost more than $700K (and obviously, there were also prof-
itability and longevity conditions), I intensifed my search. Whether
searching for a business through a broker or a newspaper/magazine, the
*

Living on a boat was another experience that I had to get out of my system. I do not regret it at all;
quite the contrary, it was a great experience. However, one saying still sticks in my mind: Te two
happiest days for a boat owner are the day you buy it and the day you sell it.

By the way, what a great deal for the banks: they were charging over 11 percent interest at the time.
And, whatever risk the U.S. governments SBA program was not underwriting, the applicant was
asked to sign for personallyyou know the drill: your frstborn and anything else you might own.
Prologue:TePurchase xxi
frst steps are similar: Te potential buyer starts by signing a boilerplate
non-disclosure contract and requesting some simple fnancial informa-
tion. Te information can vary greatly but ofen the frst glance could
include a P&L (proft and loss statement) and a balance sheet going back
fve years. If a broker is involved, the buyer will receive more prepared-
in-advance fnancial information along with a write-up about the com-
pany and generally better-prepared marketing materials. At this point,
the name of the company is not disclosed to the potential buyer and
the seller is usually keeping the sale secret from customers and employ-
ees. Normally, as the buyer takes further steps, the seller discloses more
fnancial information, provides copies of tax returns, and usually allows
an afer-hours visit of the company.
Two businesses that I considered at the time were a boat manufacturer
and a cofn manufacturer. Boat manufacturing was exciting to think
about because my hobbies include sailing, motor boating, and waterski-
ing. Plus, the boat manufacturer was located in California. However,
the business had been for sale for a number of years and, at frst glance,
the fnancials looked shaky and the questions I asked about the product
and market received very vague answers. Tat was enough to scare of
someone like me because I was putting everything on the line.
Te cofn manufacturer was a little diferent. Te fnancials looked OK
but the products were sold through a network of distributors. I learned
about a problem between the company and the network that resulted in
the manufacturer being blackballed. It wasnt just luck that led me to dis-
cover this problem: I remember that a recent sales decline had caused me
to investigate further and uncover this information. Honestly, I was also
having a hard time imagining myself manufacturing cofns. In addition
to the product being a bit depressing, I imagined that, as the owner, I might
fnd myself wishing for a local war or an epidemic to help increase sales.
A last potential worth mentioning that I became serious about was a
machine tooling shop that reconditioned machine-cutting tools for air-
plane manufacturers. It was located in Southern California and, although
not my frst choice in locations, it had the type of production fow that
I liked and it seemed lean improvements would be quite applicable. Te
numbers looked good and the owner, who had started the frm some
twenty years prior, was ready to retire. Te fy in the ointment was that
one customeralthough long-standingaccounted for over 70 percent of
the business. Tis was a red fag.
xxii Prologue:TePurchase
I went through the preliminary due diligence steps that led up to my
father and I fying down to Southern California to look at the business and
get a better understanding. Prior to a site visit, a potential buyer is usually
required to provide background information and show that the fnancial
resources are in line with the required down payment. I guess I passed
that test.
Te tour of the business looked good. My Kaizen eyes identifed a host
of opportunities even though I was walking through afer hours and no
production was taking place. I had a good feeling about the owner, who
seemed honest and straightforward. At this point, I also saw tax returns
and more detailed fnancials, which I trusted at face value. Naturally,
the business broker representing the business was in attendance and he
couldnt help but point out all the positive aspects and how, with a few
changes, the new owner could really grow the business. However, I came
away with another big concern: Te owners ethnicity qualifed him as
a minority. Normally, this would not be a concern; however, upon fur-
ther research I found that airplane manufacturers were under pressure to
reach a quota for using minority-owned subcontractors and suppliers and
the businesss largest customer was borderline on reaching its quota. Not
being a member of a minority, and knowing that the business could not
survive if this customer went elsewhere, I decided this was not for me.
About the time I was reaching my decision, the owner was completing
his fnancial statements for the previous year and was so impressed by his
increased proftability that he decided on a drastic price increase. Many
methods have been developed and entire books written on how to deter-
mine a fair selling price for a business. Te three most typical methods
are valuations based on market, assets, or earnings (or some combination
of these). I had chosen to use only an earnings-based valuation, for which
there are many suggestions of multipliers to use. Afer some research and
many discussions, my father and I decided to use a multiplier of 3x EBIT
(earnings before interest and taxes). We based this on an average of the pre-
vious three to fve years of fnancial statements (i.e., fair sales price equals
three times the three-to-fve-year average EBIT). However, the owner of
the airplane tool remanufacturing business had changed the asking price
to something like 6x EBIT.
I never monitored any of the businesses I passed up to see how they
fared. In reality, the only checking up I could do would be to call and
see whether they were still in business, and I probably could have found
Prologue:TePurchase xxiii
the current owners name (or the current corporations name) in public
records. But, I was too busy with the company I purchased.
Enough about what didnt come to be.
Sometime later in 1997, afer close to a year of searching, I found the one
I went afer and fnally got. Actually, I had come across it about six months
earlier, but the price was out of sight at that time. I was reviewing so many
Internet sites and had put the word out to so many business brokers that I
dont remember whether this particular broker contacted me when the price
fell into my range, or I noticed the price change on the brokers Web site.
What I found was a machine shop that manufactured parts for the semi-
conductor equipment industry. Tis industry is extremely volatile and
cyclical not only because of the changes in demand by the end customer
but also because of spikes in demand as new technologies are developed
and once everyone has the latest, all will go quiet again. Despite the nature
of the semiconductor equipment industry, this machine shop, located in
the Bay Area of northern California, had survived many of the big down-
turns even though only a small percentage of its business volume came
from outside the industry.
A majority of development in chip-etching machine technology was
taking place in the Bay Area during this period, but because of Californias
higher manufacturing costs, there was always the threat that any signif-
cant volume would be moved elsewhere (including ofshore) and new
local suppliers established. Tere were many similar businesses in the Bay
Area because of all the development and small lot production work avail-
able. But, this one stood out because of its longevity and proftability. It
had been a going concern for about eighteen years, originally started by
two partners working from a garage. Afer a year or two, one partner had
bought out the other. Te primary customers were based in the semicon-
ductor equipment industry, which basically produces the multimillion-
dollar machines to etch silicon wafers to be utilized as computer chips for
the likes of Intel and Hewlett Packard.
Te business was small (under ten people) but looked manageable. Te
owner was basically a self-taught businessman with no formal education.
He had some machining experience prior to starting the business and
basically started with a machine or two in a garage. Although I had never
even taken a machining class in high school, I fgured it was a good ft with
my degree in mechanical engineering and an MBA to help on the busi-
ness end. Also, I am quite a handy guy and enjoy working with my hands.
xxiv Prologue:TePurchase
Topping all of that, I thought of my lean background and my training
from Toyota. Having helped others in making signifcant improvements
to their businesses, why couldnt I do it for myself?
Although most of my training and support up to that point had been
with higher-volume businesses that had a smaller variation of part num-
bers, I fgured the principles had to be applicable in a job-shop environ-
ment. Naturally, when the time came to walk the fow through the shop
and ofce processes, I noted many opportunities for improvements.
Like many businesses, everything was learned in-house and based on
experience or tribal knowledge, and without any lean experience, there
were few established processes in place and plenty of room for produc-
tivity improvements.
I wasnt completely nave to the fact that since I had no experience with
machining, and especially CNC (Computer Numerical Control) machin-
ing, I was taking a gigantic risk. But, as everybody says during an interview,
I am a quick learner, and I tried to analyze the local machining industry
and the semiconductor equipment market, and then I attempted to fgure
out how I could grow the business. Naturally, near the top of my list were
goals like diversifying my customer base away from its strong dependency
on the cyclical semiconductor equipment market. I also realized that the
competition can be stif in any industry where the cost of entry is not
exceptionally high. However, the skill/knowledge level required for CNC
machining, especially certain complex parts, is impressive, which narrowed
the feld of competition a bit. Also, some simple research made it clear that
few of these small businesses had any lean knowledge or implementation.
So, afer a lot of back and forth on price and terms, an agreement was
reached, a contract drawn up, and a closing date set. Typically, the terms
of sale are drawn up by a lawyer representing either the buyer or seller. I
am not sure which is more traditional, but in my case I hired a lawyer to
write the contract. Te seller then had a lawyer review the contract and the
other documents prior to signing.
Typically the terms of the sale included a provision for the seller to stay
on for a month or more afer the closing to help run the business and train
the new owner or employees. We set this transition for a longer perioda
total of six weeksas I had never operated a conventional metal-working
machine, let alone a CNC machining center. I had also asked to be allowed
to hang around before the closing date to learn the business, but the seller
did not want me on-site. I assumed this is frequently the case, as nothings
Prologue:TePurchase xxv
ever sure until the money is in the bank, and until things were set in
stone, my presence might only cause anxiety among the employees.
Next, it was time to secure fnancing. Afer soliciting some of the banks
that were SBA (Small Business Administration) approved lenders, I
applied at two of them. Naturally, they were not only concerned that I had
the down payment but they also wanted to see a rsum to understand
how my skill sets might apply toward running a technical business. In
many ways, you can compare some of the mechanics I was going through
to those of purchasing a home. However, you must remember that most
of the purchase price of an established business is in its goodwill, and that
can disappear mighty fast if you upset a major customer or do not have
the skill sets to drive the business forward. Whereas homes have a market
value that is normally appraised near the purchase price, the value of a
business is determined by various fnancial ratios, mainly its proftability.
As I explained, I was valuing businesses based on their earnings and was
willing to pay about three times the EBIT averaged over the previous three
to fve years. Naturally, many other factors contributed to the viability of
the business as an ongoing enterprise. But, this helped to determine a fair
price and, this time, the owner had dropped the price to 3.5x EBIT and
further negotiations got him to about 3x EBIT.
Of the two banks to which I applied and received approval, I chose
between the lesser of the evils based on their exorbitant interest rates and
terms. I went to that traditional meeting where the banker sits behind the
desk playing God and acting as if you are about to be bestowed with a great
favor. Afer convincing the banker (who knew nothing about manufactur-
ing, let alone machining) that I knew more than he did, and committing
myself to make it work or lose everything I owned (a really compelling
motivation), he graciously approved the loan on a fve-year term at just
over 11 percent interest. Tat probably makes most people feel just a bit
better about their mortgage payment. I was also required to purchase a
very expensive life insurance policy (sold through the bank, of course, at
what I assume provided them a healthy commission) to cover any out-
standing value on the loan in the event of my early demise. Tey really
know how to cover every eventuality.
A whole host of other activities must be completed before the closing,
one of which was the legal process of incorporating. At the closing, the
corporation I set up bought the business assets and rights and the business
name from the sellers corporation. Any lawyer will tell you that this at least
xxvi Prologue:TePurchase
afords a buyer some personal protection because the legal entity purchases
the business and assumes all the liabilities. In my case, the previous owners
corporate name was diferent than the name of the business he was selling
(doing business as or d/b/a). So, my recently incorporated corporation
was able to take over the existing operating name the former corporation
had used as a d/b/a for the last 18 years. Afer all, the goodwill is a big
part of what justifes paying the asking price, since only taking into account
the asset values would make many companies worth a lot less.
The actual closing is similar to that of buying a home except you
seem to sign about double the amount of documents, if you can imag-
ine that. Closing day for this purchase was set for Friday afternoon,
and I had asked the seller to plan overtime for Saturday for two rea-
sons: (1) although a seller may try to move as much volume through
the door to maximize profits before turning the business over, there
was still a backlog, and (2) as the new owner, I would need to know
some important things right away, like how to unlock the doors, turn
on the lights, and make the coffee. Sounds like I had set myself up for
a stressful first day.
Since the purpose of this book is to tell my story of a lean guy buying a
business and utilizing lean manufacturing methods to grow it proftably,
and it is not a how-to guide for purchasing a business, I think this is
enough background on the purchase. I did not highlight any of the les-
sons learned, as nothing pertained to improving an ongoing business. I
just felt it may be interesting for you to know how it all got started. I also
believe there are much more sophisticated methodologies for evaluating
and purchasing a business, as there are many books written on the sub-
ject. Terefore, I would not necessarily recommend following the way I
went through it.
1
1
The Beginning
Observation and Documentation
1.1 IntroductIon
It was Saturday morning, day one of owning and operating my frst business
and I had bought donuts to encourage a little team building. I was staying at
my brothers house while looking for my own place and, on my way out the
door that morning, my brother asked whether the donuts were in lieu of a
401K plan for the employees. I guess they were, in a manner of speaking.
I think at the time there were four or fve employees in the shop and two
in the ofce plus the owner who functioned as the manager. Te business
grew from this point although, due to the cyclical nature of the industry, it
periodically fuctuated in the number of employees.
Tose of you hoping to make analogies between this business and yours
might be thinking that because you are part of much larger organizations,
you therefore have much more complex considerations, politics, and cor-
porate cultures to get past and will not gain sufcient insights from the
experiences I am sharing. Having worked for two large companiesGen-
eral Motors and Toyotaand then owning my own small business, I can
assure you that most of the simple principles that I implemented are appli-
cable to your situation. Te question is how to cut through the red tape,
bureaucracy, fears, and personalities involved in larger organizations to
get some of these things done. In most large companies, people will ofer
their theories on why change is so difcult, why there is so much resistance
2 Mr.LeanBuysandTransformsaManufacturingCompany
to change, even when new management is brought in to drive this change.
Spending more than ffeen years helping mostly large and medium-sized
organizations with their lean transformations, I agree that the complica-
tions, politics, and culture must be addressed in the plan but the overall
steps remain the same. By reading about my experience, you are sure to
pick up on many improvement ideas along with proven methodologies to
gain buy-in, thereby convincing the organization. However, I had one very
large advantage you or your organization might lack; I am a true believer
in lean and that can be a difcult conviction to acquire.
Troughout this book, I have included highlighted text featuring the
lean principles that I focused on at the time, and I have further enhanced
the discussion with key points which are also displayed in italics. As a
business owner, my purpose was not explicitly to implement lean methods
but instead to increase the businesss proftability. Rather than turning the
employees into lean experts, I wanted to sell the ideas to them as simple
and logical steps to secure our longevity (job security). I assure you,
many of the basics I put in place in my own company are either absent
or not truly supported in other organizations, but overcoming the issues
introduced by the size, bureaucracy, and personalities in various organi-
zations is not impossible. What is required is a similar simplistic approach
and an unwavering belief in continuous improvement.
1.2 observIng
With only a six-week training period to learn an entire business of which
I knew nothing, I went in that frst Saturday morning with a plan. I was
going to follow an idea I learned at Toyota known as OSKKK (Observe,
Standardize, Kaizen Flow & Process, Kaizen Equipment, and Kaizen
Layout; a detailed breakdown of OSKKK is shown in Appendix A) to
guide my learning of the business as well as for structuring the lean
transformation. I fgured I would follow this sequence of steps as closely
as I could. And since I needed to learn the various processes, not only as
a basis for improvement but as a necessity to manage the business, this
would keep me from randomly jumping in with specifc improvement
ideas. Te beauty of OSKKK as a method of improvement is that it also
closely follows the conventional wisdom that, when taking over a new
TeBeginning 3
position, you frst observe and understand the business and its environ-
ment before implementing changes.
Te week before I took over, I had written down all the likely titles for the
administrative and shop-foor tasks in this type of business. For the ofce
side, I broke them down into sections like order entry, purchasing raw
materials, subcontracting, invoicing, closing the books, etc. For the shop, I
had grouped items like material preparation, setting up and changing over
a machine, programming, scheduling, manual machining, quality checks,
instrument calibrations, and many more. I organized a three-ring binder
according to these categories and fgured I would gather the information
as I encountered it.
Basically, I planned to learn the business from the ground up. Tis
entailed hands-on learning for each task by going to where the work was
done and the value was being added. Tis is ofen referred to as going to
Gemba. Naturally, I had put myself on the fast track with only six weeks
to accomplish this. Realistically, I knew I could not decide the timing of
learning a particular taskthat would be dictated by the demands of an
ongoing businessbut the three-ring binder would help guide me, ensur-
ing I did not miss anything that I felt was important. So, I would go with
the fow and learn quickly, observing and documenting.
Lean principle: In the lean world, going to where the work is done (or
where the value is added) to learn about the process before making
improvements or decisions is known as Gemba. At Toyota, new employ-
ees, regardless of level, normally start their employment with hands-on
learning of tasks within their areas of responsibility.
For the time being, I was planning to keep all the employees; therefore,
I did not need to immediately learn the details of every job. But, I had to
determine which things the former owner/manager did by himself and
quickly become competent in them. One advantage was that I had pur-
chased the business in the middle of the month, so the six-week training
period spanned the end of the next two months. Tis meant that I could
twice observe the end-of-month process of closing the books.
When I arrived that frst Saturday morning at 5:55 a.m., the now-former
owner, Bob, and the shop foreman, Tom, were already hard at work. Two
of the CNC (Computer Numerical Control) machines were already up and
running. We had agreed on a 6:00 a.m. start, and things already appeared
well underway. My frst observationor should I say concern?was that I
had either walked into a staged scene or this was one hard-working indus-
try. I (being a slightly later riser) had questioned Bob about the wisdom of
starting at 6:00 on a Saturday morning, and he had told me the guys liked
early starts, especially when working overtime on Saturdays because they
preferred to get it over with. Finding the machines running before the
agreed-upon start time did not jive with what I had always observed in
larger businesses, where it seems that although people arrive on time, they
do not necessarily get right to work. First, they might get cofee or have
some discussions with coworkers, and then they think about turning their
machines on and running through a warm-up procedure. A 6:00 a.m.
start time in larger companies usually meant an actual production start
closer to 6:30 a.m. When consulting, I frequently recommend an hour-by-
hour target to avoid this lost production. However, late production starts
did not appear to be a concern in my new business.
Lean principle: Managing takt time (time available/customer demand) is
usually handled in low-volume, high-variation businesses that schedule a
variety of products to run on a particular machine/process with a tool that
is ofen called a day-by-hour board. Tis board allows planning of the var-
ious customers part numbers against the processs capacity, and execution
of that plan to be visualized in real time. Tis type of tool was necessary in
my business but would be introduced during the Kaizening stages.
As I walked in, Tom, who had been with the business for more than
sixteen years, eyed me with suspicion, as anyone in his situation would.
Afer the papers were signed the previous day, Id had a short meeting
with the team during which I indicated that I had no intention of mak-
ing any changes for the time being, and they were stuck taking me at my
word. Tom, a short but solid man who looked like he could hold his own,
was also a very direct sort and had stared directly into my eyes and asked
about my machining background during that frst meeting. I frst tried to
defect the question by explaining my years of working in manufacturing
and similar businesses. But, afer talking around it for a while, I had to
TeBeginning 5
level with them: I had never managed or worked in a machine shop. All
eyes drifed toward the foor, each man now considerably more nervous
about his future. I could tell by the way Tom was looking at me that he
was probably searching for a lifeline. But, it was not the time to explain
the lean principles and how they can help to improve even in a job shop. I
guess I could have responded that Im a fast learner. At any rate, I had to
keep in mind that I was not the only one with something to lose if things
did not go right.
My fatherwho had also owned his own business although on a much
larger scalehad advised me to take the time to learn the business from
the ground up, understanding the processes and the people before even
considering any change. Tis is the wise course for anyone in a new posi-
tion, and I knew I would need it continuously reinforced. I knew it was
sound advice but, not being a patient man, I also knew it would be difcult
for me. However, in his own way, my father was describing the OSKKK
methodology I was planning to utilize to learn and transform the busi-
ness. My fathers advice proved to be valuable and very proftable.
Key point: Even if you are under pressure for quick-change, aggressive
solutions or feel sure you know what the problems and solutions are,
take the time to observe and learn the processes (go to Gemba) and
gain insights from those performing the tasks. In some cases, this might
even require understanding the customers, the market, and the people
instead of frefghting and quickly making changes without the buy-in
of those involved.
1.3 gettIng A detAIled understAndIng of the
Processes I Would be resPonsIble for
In cases where people have worked their way up in an organization, they
may intimately understand many of the processes that they have worked
their way through. However, Western companies are too frequently in a
hurry to make a new employee productive (which, by our own defni-
tion, is usually to start performing their assigned job immediately) and
tend not to allow the employee sufcient time to become familiar with the
business, its processes, and its people. Tis is especially true with either
the arrival of a seasoned manager or some university graduates; we tend
to fast-track them, skipping any structured process orientation. I am con-
vinced the reason we cannot fnd the necessary time is because we are not
able to quantify the lost revenue that this lack of training and orientation
costs our organizations.
When I was hired by Toyota (NUMMI, the ToyotaGeneral Motors
joint venture) as an engineer, I spent the frst ten weeks observing and
then learning the business in a hands-on fashion. Te frst week I was sent
to paint school, where I learned a lot of technical aspects of formulating
and matching paint, and then I spent nine weeks on the shop foor as a
team member moving through the various process steps. I worked one
week in each section of the paint shop: mixing the paint, spraying it on the
cars, and making repairs. Tis was an investment in which the payback to
my employer was at least ffy-fold. It is unfortunate that many companies
feel they cannot aford this time. Toyota understood the payback, and now
I clearly understand its importance. And so, I planned to learn all the pro-
cesses I would be responsible to oversee.
When speaking about learning many parts of the business and becom-
ing expert in the various disciplines, I realize that this would be an
impossible task in many industries. It is not humanly possible in larger
organizations today, with the ever-expanding range of disciplines and the
increased depth of knowledge that each is continuously moving toward, as
the many silos of knowledge run very deep and the number of silos are
ever-increasing. Each silo can require a lifetime of dedication to become
profcient. But in these cases, the idea is to strive instead to understand
related or connected processes. Doing this will help to better integrate and
lead to improvements between the people and related departments.
1.4 A lAck of trAInIng cAn be costly
Tis lack of time for orientation and learning was clearly displayed in
the drawings we received to quote. Te majority of engineers responsible
for designing these fabricated and machined parts had obviously never
worked in any of the industries responsible to manufacture the parts.
TeBeginning 7
And they had no practical experience in designing parts that could be
easily produced, or in the relationship between their designs and the
cost implications. Being an engineer myself, I frequently understood the
application of the parts we were building and what features were critical.
I continually found us working with difculties such as wrong tolerances
called out in drawings. Or, drawings that made the part difcult to build.
Or, there was a much easier way to build the part (i.e., making it out of
two separate pieces and welding it together). Almost always when I was
able to explain the problem to the engineer, he or she would ask what I
suggested, and follow up on my suggestion with a red line drawing,
implying they almost never took the time to change the original. Instead,
these red lines kept everyoneincluding incoming inspectorsin the
dark. Frequently, the red line drawing had to be packaged with the part
so it did not get rejected when compared to the ofcial drawing, which
had not been modifed.
What transpired more ofen when the drawing mistake only drastically
increased the cost (instead of making the part nearly impossible to build)
was to quote the part according to print and not make the efort to ques-
tion it. Tis frequently resulted in the quoted price being twice as high to
hold the tolerances called out or to manufacture the part from one solid
block instead of welding two pieces together. But, it was the easiest course
of action for the suppliers. I used to think that it would save the company
a small fortune if the engineer had just done a short stint in a machine or
fabrication shop. Tis will be discussed later because it ofen cost our cus-
tomers considerably more than necessary.
Lean principle: What our customers lacked is known as DFM (design for
manufacturing) or DFA (design for assembly). Tis is the idea or practice
of designing products that are easy and less costly to manufacture and
assemble. Tis knowledge can only be gained by spending time where the
work is being performed (Gemba). Because we were not responsible for
these design stages and were frequently a few steps removed from having
contact with the designers, we ofen had little infuence to improve the
process. Remember: this early and critical step of involving manufactur-
ing saves a lot. Typically, 75 percent or more of the cost is fxed by the
products design.
1.5 trAInIng As A MAchIne oPerAtor
On that frst Saturday, I started working as a machine operator while Bob
managed the shop. I fgured there was no use trying to manage processes
of which I did not yet understand the details. I would only be setting myself
up for failure and would lose face in front of my employees by making sug-
gestions or impossible demands. I started on what turned out to be a good
job from a learning standpoint. It involved a part that needed three sepa-
rate machining operations before being sent out for plating; then it would
come back for an assembly process. Te material had been purchased cut
to size and it had already been through one of our tumblers to remove
rough edges for better hold in the fxtures and easier handling.
Tom had already set up the machine the day before and the frst piece had
been qualifed. Qualifcation entailed the setup operator adjusting the CNC
until he felt he had a good part. Afer that, he took it to another setup man,
who would perform another 100-percent dimensional check. I had many
questions: Do we always purchase material to size? Who decides, and
how is it determined which features are machined in a particular operational
step? So, I found it difcult to pay attention to the detailed explanation I was
receiving about loading the machine: Clamp the part too tight and you might
stress the material or move the zero reference point. If not clamped tightly
enough, it might be thrown out of the fxture during cutting. Afer watching
Tom load a few pieces and press the start button, I was ready. I loaded a few
pieces wrong and caused a little scrap, but I soon had the hang of it. I was also
taught to deburr and check a few dimensions, but I was being closely watched
by Tom and Bob. Nobody wanted me blowing it the frst day.
1.6 dIscoverIng A lAck of docuMentAtIon
All the instruction on the various tasks was word-of-mouth explanations
coming from years of experience. So, when I asked whether there was any
documentation for these procedures, the response was a perplexed look,
as if I had just asked the worlds stupidest question. As is the case in most
businesses, it soon became clear that there was basically no documenta-
tion (see tips on easy ways to document a task in Appendix B) and every-
thing was tribal knowledgemeaning it was all retained in the employees
TeBeginning 9
head, and when the employee lef the company, the knowledge went, too.
ISO- (International Standard for Organization) certifed companies also
usually lack any useful work standards for their respective tasks. I knew
I had my work cut out for me, observing and learning by day and docu-
menting by night. Tis is such a valuable point, especially regarding lean
implementation and the need for standard processes (the second step of
OSKKK), that I want to highlight it as a key learning point.
Key point: Te easiest time to document previously undocumented pro-
cesses is during the observation phase or when someone new is being
trained on the particular process. Tis then becomes the basis for the
second stepstandardization.
1.7 IdentIfyIng skIll sets thAt
needed to be rePlAced
By way of background, the following list describes the three basic skill
levels in a machine shop:
Te frst level is the operator whose duties include loading and unload-
ing the CNC machine, deburring the part, and taking some simple
measurements. On a manual machine, it might entail loading the part
and running the tool to a specifc point following a prescribed method,
and then checking the part. Operators are usually not responsible for
setting up either a manual or CNC machine or for making adjustments.
Usually, a setup man is nearby, keeping a close eye on the quality.
Te next level is the setup man. In addition to being familiar with all
the operators tasks, the setup man is also responsible for setting the
tools and fxtures in the machine. In the case of a CNC, this might
include loading the program, putting the ofsets into the machines
computer, then running through and qualifying the frst piece. Also,
the setup man is responsible for making adjustments in the ofsets
and tool setups and, in some shops, making corrections and adjust-
ments in the program (changing codes and numbers).
Te programmerusually the third levelhas normally worked his
way up and is capable as a setup man and an operator. In addition,
the programmer must be able to look at a drawing, conceptualize
how to make the part, decide what machine to use and what type of
fxturing is necessary, design the necessary fxtures, and determine
the number of operations. He must be computer-literate and capable
of writing the programs and debugging the programs when prob-
lems arise. Today there is sofware that will write the individual lines
of machine code, but the programmer must put in all the applicable
information and dimensions and basically establish the priorities.
Te programmer must also draw up or specify any special tooling
required so it can be manufactured by a tool shop.
Te existing operators and setup men possessed sufcient skills for their
areas, so no worries there. But, Bob had previously done most of the pro-
gramming and he was leaving soon. Te other skill that I would not be
able to handle afer the six weeks of training was the estimating, which
had also been done solely by Bob. During the due diligence period, I had
come to the conclusion that I could survive fnancially if I hired a com-
bined programmer/estimator, and managed Bobs other tasks myself (in
addition to making the loan payments). So, my idea was to hire a pro-
grammer before Bobs exit.
Te frst problem with hiring someone from outside was the sofware
that was being used in my shop for CNC programminga DOS-based
programwas outdated, and it would be difcult to fnd a programmer
familiar with it. Can you imagine reverting back to working in DOS afer
working in Windows? Bob had purchased a newer, Windows-based pack-
age, but it was not being utilized because he had not found the time for
training. A little research revealed that this new sofware was commonly
used and it would be easier to fnd someone familiar with it than someone
familiar with the DOS-based program. Naturally, I was also planning to
learn to write programs, and there was no way I was about to revert back
to working in DOS on a daily basis.
1.8 WorkPlAce orgAnIzAtIon (5s) oPPortunItIes
Lean principle: Workplace organization, known as 5S to lean prac-
titioners, is a fve-step processSort, Straighten, Shine, Standardize,
Sustainto minimize the non-value added time caused by not having
TeBeginning 11
all necessary tools and materials organized in the most appropriate
places. Ofen during the observation step in OSKKK, waste caused by
poor workplace organization is identifed and made a priority when
starting a Kaizen.
5S (workplace organization) is important any time employees struggle to
fnd what they need, whether it is tooling on the shop foor or a fle on the
hard drive. As a consultant, I sometimes spend time helping companies
organize their computer fles, and I fnd the easiest time to do this is in
conjunction with some other change (e.g., a new regulation, new employ-
ees, new sofware, etc.). It is not always necessary to spend months going
back through the old fles. Just set the new rules: (1) every time an existing
fle is used, it should be moved to the new system; and (2) every new fle
should be created in the newly organized system. Tis also allows you to
understand which fles can be purged or archived. Files that have not been
moved to the new system afer a year or two are likely unnecessary; con-
sequently, they can remain in the old system in the rare case there is ever
a need for them in the future.
Afer buying my business, I quickly discovered that the hard-drive
directories were not set up in a simple standardized system. Tis problem
existed in both the ofce and the shop, but at this point I was addressing
only the shop. I was determined to suggest a new standardized fling sys-
tem for the hard drive. I felt it should be started for all new part numbers
that were to be programmed, and the existing fles could be cleaned up,
reprogrammed, and reorganized as we received repeat orders once the new
programmer was on board. Although I have stated that no changes should
be made until the business is truly understood, there are times when a
change can be made successfully. In this case, a new, experienced person
would be handling an aspect of the business that had formerly functioned
based on tribal knowledge. Later, I was glad I started early on with this
new structure to organize the fles, and so was my new programmer.
Key point: Another lean lesson is that 5S applies to all aspects of the
business and it is best to start its introduction in complicated and large
systems by using a signifcant event like the start of a new employee.
Begin using the new method going forward on all new work instead of
trying to motivate people to go back and clean up all tools, materials, or
data (some of which may not be represented or utilized in the future).
Although this might leave you working in two systems simultaneously
for some period, at least it gets you moving forward.
Sooner or later, you need to draw a line in the sand and make improve-
ments going forward, even if you need to operate two systems for a period
of time. Te advantage is that soon you fnd some information in the old
system that is not required; therefore, spending time and money to put
this into a new format is a waste.
1.9 dIscoverIng dIfferent Methods beIng utIlIzed
for the sAMe tAsk (no stAndArdIzAtIon)
Since the crux of improving any business lies in understanding the value-
adding processes and minimizing those that add no value, a detailed
understanding of the processes are required to know the diference. My
frst full week in the business entailed a lot of learning. Everything was
being taught as it transpired during the normal course of business; how-
ever, I concentrated on spending as much time as possible learning the
next level in the shop, which was setting up a job. Setup was the key in this
business, as the order size was small and most machines went through
three or more setups per machine in a single day.
I had already set up my frst job with Tom, and my next one was set up
with the help of one of the setup men. As you might imagine, I discovered
that things were not done the same way using similar methodologies. Each
man had his own methods and experiences. As with the other tasks I had
learned, no documentation existed for how to set up a machine. Te meth-
ods and tricks all came from personal experienceTom had learned under
Bob and the other setup man had combined what he had learned from
Tom and previous employers. As a result, the methods difered because
they used diferent sequences, had various priorities, and put their own
twist on it. Naturally, I found one method had more advantages and I doc-
umented that as the baseline. I started identifying a lot of improvements
in the setup and later, by applying some SMED (single minute exchange
of dies) methodology along with some training and small investment,
TeBeginning 13
signifcant improvements would be made. In a later chapter, I will discuss
the setup improvements as we implemented them, but for now well dis-
cuss the documentation of the current best methods, which was my focus.
I say current because, as all lean hands know, we want to continuously
improve these methods; therefore, our documentation should be easy to
change.
Lean principle: Standardization of work was the second step in my trans-
formation plan (OSKKK). Where these standards are lacking usually
becomes apparent during the observation period. It is critical because with-
out standardization, there is no baseline for improvements. Standardized
work is the current most efcient and safe working method that provides
the desired quality.
Documenting a standard method is easy and painless these days with
digital cameras, video cameras, and computers. Even if you do not have
time to properly document the activity because of its complexities, the
quick solution is to make a video of someone going through the steps
while explaining what they are doing. In the interim, you can just store the
video in a safe place, and if you need it for training, you can simply play it
back. I want to capture this as a key point because it addresses the excuse
of not having time for documentation. (Various options and methods of
documentation are shown in Appendix B.)
Key point: Tere is no excuse for not having the current best method (of
the standardized work) documented because making a video is fast and
easy. Te operator or team leader can explain what he or she is doing
while working through the process.
1.10 dIscoverIng dIffIcult Processes
to stAndArdIze
As I continued learning the various tasks, I found documenting how to
write a CNC program turned out to be signifcantly more difcult than
other processes. Bob was self-taught, so he had his own way of thinking as
well as certain idiosyncrasies, especially since he was utilizing outdated,
DOS-based sofware. I decided this would be difcult to standardize, espe-
cially since his experience was based on sofware I did not intend to learn.
Instead, I decided I would immediately start interviewing programmers.
During the interviews, I would give the applicants a copy of a drawing
and ask them to list the programming steps and describe how they would
estimate the part in order to determine whether any standard methods
appeared. I found that few of the candidates went about it in exactly the
same way but there were certain similarities, and I used these as an out-
line for the standardized work steps. Because programming frst involves
conceptualizing how to make the part and defning critical process steps,
many variations for the same part evolve, some more efcient than others.
Terefore, people in these roles need to be continuously challenged and
educated, especially if they are homegrown. I planned to use my notes of
the various methods and correlate the similarities along with refning an
ideal process fow when I had my new programmer/estimator on board.
Key point: Where processes (i.e., programming) allow diferent tools,
programming, options, or sequences that will achieve the same end
result, it is best to continuously have employees retrained and updated
on new technological developments in their felds and, where possible,
periodically challenge these outputs by comparing them to another
experienced persons results.
1.11 thIs Much observAtIon requIred
Me to dIvIde And conquer
To learn the ofce tasks, I had a slightly diferent plan in mind. To learn
everything (both shop and ofce tasks) in six weeks was a bit overwhelm-
ing so I enlisted the help of someone who I wholeheartedly trusted to
oversee fnancial matters and who was skilled in ofce administration:
my mother. She was going to help by learning the key ofce tasks (while I
was in the shop with Bob), which included focusing on all the front-ofce
activities for which Bob was not responsible (e.g., processing a purchase
TeBeginning 15
order, cutting payroll, or developing a bill of material). Tese activities
also lacked documentation so my mother would observe and document
the current processes. Naturally, these tasks were learned skills and a
small business owner may feel there is either no need to document them
or not enough time available for documentation. Obviously, there are
many opportunities to improve all these business processes but if theres
no standard method, then theres no baseline from which to make these
improvements. Again, video is an easy out.
Computer tasks are also easy to document using the print screen func-
tion along with an explanation of the steps. Most companies processes
involve many decision points; for example, do X if it is an indirect
material, but do Y if it is a direct material. For these cases, a simple
fowchart with decision boxes helps clarify things.
Step 1 of OSKKKobservationincludes the following major sub-steps:
Take the time to see what is happening in the work area by watching
multiple cycles of the same process.
Watch more than one person performing the process and note
where standardization is lacking, especially where it afects quality
or productivity.
Document by writing or videoing the individual process steps in the
sequence they occur.
Identify the origins of variation in both the fow of information and
fow of materials.
Key point: For documenting ofce processes, you can use a combination
of fowcharts, screen prints, and videos to capture the process steps for
training (examples are shown in Appendix B).
You might never have been this concerned with documentation before.
It might have been a nagging issue but there always seemed to be other
priorities. In my case, this was an easy choice for a starting point. My
lean training had taught me the need for a standard method as the basis
for making improvements, and I completely understood the danger if no
one experienced to perform these tasks was available. Te cost of training
new people is high; it is even more prohibitive if it is based on memory
and a critical step is overlooked. Most accounting systems are not able to
directly track the cost of training versus the considerable losses related
to mistakes or the low productivity and poor quality associated with a
lack of cross-training. But if they could, most managers would be able to
justify and prioritize documenting and standardizing the current best
method. Toyota had learned to justify this cost.
For the time being, my mother was using a combination of all the avail-
able methods to document what was happening in the front ofce. At a later
point, she would function in an ongoing capacity as the bookkeeper, com-
pleting the monthly statements and overseeing the accounts payable func-
tions by reviewing and signing the checks. Tat would allow me to sleep
better at night because I knew she was good at watching the pennies.
1.12 suMMAry
Step 1, the observation, had begun along with the related documen-
tation. Tis served not only as the basis for step 2 (standardization),
it also allowed me a structure in which to learn the processes so I
could lead the business and its improvements.
A lack of standardized processes was becoming apparent.
A lack of workplace organization (5S) was also identifed and its cost
implications understood.
Tribal knowledge was proving to be the predominant method of
operating.
17
2
Standardizing the Estimating Process
2.1 hIrIng A neW ProgrAMMer/estIMAtor
I had begun advertising for the programmer/estimator by the end of my
frst week as the new owner for several reasons: I wanted to have the for-
mer owner, Bob, involved in the selection process with his knowledge of
the business and his experience with the countless interviews he had con-
ducted. And I needed someone in place prior to the end of my six-week
training period because afer that point Bob would be gone and it would
all be up to me.
Although many programmers were familiar with the new sofware I
planned to use and with the controllers on our newer CNC (Computer
Numerical Control) machines, few really had much estimating experi-
ence. Many claimed to have done it, but Bob was unhappy with most of
their answers when we gave them a drawing and asked them to talk us
through their way of estimating setup and run times for each operation.
I was unhappy, too, with what I witnessed. Tere seemed to be no formal
structure in how the candidates approached the estimating process; their
thinking and their answers in terms of the time required to complete
the job varied widely. Although I was not sure what the correct answer
was, I knew it had to be determined by something more reliable than a
wild guess. Tis was driving home the point of how critical and difcult
estimating is.
I realized why so few programmers had acquired the necessary skills
when I discovered that most owners of small-to-medium-sized businesses
either did the estimating themselves or closely supervised the process and
adjusted the numbers before submitting the quotation. Tis was basically
tribal knowledge that most business owners learned through the school of
hard knocks and had never put any thought into systematizing. And this
was true in my new business, where nothing about the existing estimating
process had been documented. It was more like black magic that took place
inside Bobs head, and even he was perplexed by how to standardize the pro-
cess. You just have to know what youre doing. Every situation is diferent.
Afer two weeks of conducting interviews with about eight candidates,
Bob was just happy to fnd one who described approximately the same
operational steps for manufacturing and who fnally arrived at the setup
and run times that Bob had determined for a particular part. As for me,
my lean experience would not allow me to believe that a fowchart or a few
check sheets could not help to standardize the processes for the diferent
types of parts someone was likely to quote. As I learned how to quote, I
later put a lot of standardization behind it.
2.2 WrItIng the fIrst drAft of stAndArdIzed
stePs for ProgrAMMIng And estIMAtIng
We fnally had found our manJohn, the candidate who had come clos-
est to suggesting a sequence similar to the one Bob had used previously to
manufacture the part in question. He was also in the range of Bobs setup
and run times for each operational step. I was just happy when I asked him
how he arrived at these conclusions and he explained a clear and logical
process that he followed.
During the short shop-foor walk at the beginning of the interview,
John had assessed our various fxturing and tooling, and he was familiar
with the newer CNC equipment we had (although he had no experience
with any of our older machines). His thought process started with how
he would fxture the part for each operation with what was available, and
he thought through what features would interfere with which other fea-
tures when machining, which helped to determine the best sequence of
operations. Ten, he thought about how many tools each operation would
use to determine the setup time and how much tool change time would
be required while the program was running (e.g., he actually knew our
StandardizingtheEstimatingProcess 19
machines had a seven-second standard tool change time). Finally, based
on the types of tooling we had and the available RPMs on the machines,
he could determine the approximate cutting time for each tool. In other
words, he could sum the times for each tool and approximate the total
run time for each operation. He had quickly done his calculations on a
scrap sheet of paper while Bob and I waited; however, when he walked me
through his notes, there was a certain logic to it. I fgured that this scrap of
paper was the basis of a process that could be standardized. Obviously, he
was only quoting the labor portion of the job as he could not pull material
prices out of the air. But, he clearly knew what materials and outside oper-
ations like plating and silk-screening would be required to manufacture
the part. I had the beginnings of a standard process fow and quickly
captured it. (An example of how I frst standardized the quotation process
is shown in Appendix C.)
2.3 consIderIng the coMPlIcAtIons
WIth quotIng
Tere are many cost factors to include in the quotation, even for a simple
job shop that does not design the products. Te basics are material and
labor. Although materials would seem easy, you only need to quantify the
requirements and then get pricing from your suppliers. Even in this sim-
plicity, complications exist. In my case, for example, either I could buy
materials cut to size or I could buy some standard sizes and cut them to
size myself. So, there are decisions to be made at each juncture in the esti-
mate; therefore, the process starts to look like a fowchart with a lot of
decision boxes.
Te labor estimate should vary for the diferent processes, based on
their respective work content, and should be in line with market rates.
For example, a part that requires extra deburring time should not have
this manual task (which is a lower skill rate) charged at the same rate as
CNC machining time (unless it is very critical deburring and requires a
higher skill level). Usually, assembly rates are also lower because they do
not require capital investment, or likely require extremely high skill lev-
els (again, rates will be higher for more skilled and critical assemblies).
Te machining rates should also vary depending on the type of machines
required, their cost, and their difculty to operate. Other considerations
are the type of material being machined and the related tooling (and other
consumable) costs associated with it. Harder machining materials and/or
more expensive materials demand a higher rate because of tool wear, etc.
So, labor needs to be divided between the various tasks, and hourly rates
need to refect a competitive cost within the local market. Tings like tool-
ing costs can be established by gathering data (not taking wild guesses).
None of this had evolved into a system of various costs/rates during Bobs
tenure, and I only began to improve the system when reaching the Kaizen
Flow & Process phase of my OSKKK transformation strategy.
As John began his work, I tried to learn and document as much pos-
sible. He seemed confdent with the new sofware and appeared happy
to improve the organization of fles on the hard drive. He was familiar
with the numeric controllers on the newer CNCs in the shop, but I had
two older CNCs that were unfamiliar to almost everyone we interviewed.
Tis always helps prioritize where to document frst. Outdated and spe-
cialized equipment are by far the hardest to fnd experienced people to
operate, so this is always a better starting point for documentation and
standardized procedures.
Key point: Far fewer people in any industry are experienced with the
special requirements of older and outdated equipment and processes.
Terefore, this is a priority when documenting because it is unlikely that
you will fnd many people within your organization or outside of it who
are familiar with the specifcs of older equipment and its idiosyncrasies.
2.4 MInIMIze trAInIng And stArt-uP tIMe
by utIlIzIng sIMIlAr equIPMent
Te other thing to keep in mind when purchasing new equipment is that if you
stay with the same equipment manufacturer or at least the same type of sof-
ware or controllers, you are likely to end up with minimal training require-
ments and the programming will be the same or similar. I think if companies
could quantify the time lost training people on all the various control systems
in a plant, more efort would be put into minimizing the variations.
As my business grew, I bought the same brand of equipment with identi-
cal controllers whenever possible, and we were always able to run those at
full production the day afer the machine had been commissioned.
2.5 contInuIng to deterMIne Where
skIll sets Would be lost
Once John was hired, I went back to concentrating on learning Bobs
specifc responsibilities. Although John was handling all estimates as we
received requests for quotations, I had Bob review each one while I looked
over his shoulder. We had hired John because his estimates had come clos-
est to Bobs in the test part of the interview but I was now seeing sig-
nifcant diferences between them in estimating costs. Tese diferences
highlighted the need for standardizing the process as much as possible.
In some cases, Bob would pick up something John lef out, like a stress
relief that would help guarantee a certain fatness, or a tight tolerance on
one specifc dimension that was sure to take extra time and likely to cause
some scrap. Again, I felt that check sheets and fowcharts would be neces-
sities, especially when Bob would no longer be around and it was going to
be my responsibility to look over Johns shoulder.
Afer getting through the frst and second steps of estimatingthe
material and labor coststhe next step is to factor in the current market
conditions, the particular customer, the required delivery time, and a host
of other considerations before converting the material and labor costs into
a fnal price. It is complicated enough to determine labor accurately, but
the distribution of overhead costs like administration time, programming
time, packaging costs, etc., could further complicate each estimate.
So, Bob used the simplifed method of a standard shop rate$60 per
hourthat had all our overhead costs and proft margin rolled in. Tere
were some exceptions, but that was the norm. Bob felt it was representative
of the area and the market in which we competed. However, he also felt
this number should be adjusted in some cases depending on the customer,
how desperate they were, the likelihood of the part receiving a repeat order,
etc. Nothing as easy as $50 of material plus two hours at a rate of $60 per
hour for a fnal price of $170. Tis sales/marketing twist of adjusting the
price required knowledge of the customer, the current market conditions,
identifying likely competitors for the quotation, and a few other consid-
erations. And, of course, all this had been previously considered and ana-
lyzed in Bobs head based on his experience. To me, utilizing one fat rate
with some vague rules for adjustment at the end of the process did not
appear to be a reliable system. So, to my way of thinking, standardization
would be a necessity.
Most sales people say the market, not the supplier, sets the price unless
there is some type of proprietary product or special patent. Terefore,
there is always a requirement for knowledge of the market. And there are
various viewpoints on whether to use standard proft margins, cost plus
pricing, or simply an adjustment made by the sales person based on mar-
ket conditions at that time. In our case there was not a market price
since these were customer designed parts; instead establishing a price was
a responsibility shared between the estimator and salesperson, as is the
case in most make to order businesses. Terefore, my focus was to quickly
understand the customers and their habits, and start applying a markup
or markdown as required. At a later point, I would revisit this to see what
type of standardization could be introduced.
2.6 MeetIng the custoMers
During my second week, Bob had started taking me out to meet the cus-
tomers and explain the ownership change. I had seven major customers
and others that made less-frequent purchases. Bob called the customers
frst and asked for a bit of their time, and he mentioned that I was coming
along to be introduced and he had an announcement to make. Naturally,
customers become nervous anytime there is an ownership or management
change because dealing with the unknown afects their comfort level.
As I met the customers, I assured them that there were no major changes
planned and tried to discuss any current concerns or issues they had. Some
had a few comments on price but, overall, everyone was very satisfed with
the quality, customer service, and on-time delivery. Naturally, I assured
them that this would not change.
I had noted during the due diligence process that although the company
I purchased had a nice sales distribution among its larger customers, most
of the customers were concentrated in the design and manufacturing of
semiconductor etching equipment, which as I previously mentioned, was
a volatile market. During these meetings with the customers, I continued
to think about my diversifcation plans. When frst reviewing the com-
pany, I had questioned Bob about why he had not diversifed into other
industries as his equipment was very fexible, and he indicated it was pos-
sible but he had not put in a lot of efort. Tis is a typical response during
the sale of a business as the seller obviously wants to dangle the carrot for
the prospective owner. But I was pushing Bob to get his specifc ideas, even
though I frst needed my own experience observing and leading the busi-
ness as it currently operated.
Lean principle: Customer focus is a lean principle that might appear
intuitive but is ofen not well-executed. In product development, it
entails understanding the customers needs and the market. With regard
to quality, it requires understanding what quality level the customer
desires and is willing to pay for, and then ensuring those needs are met.
With regard to delivery, it is necessary to understand the customers
expectations and what others in the market can provide. Finally, apply-
ing the focus to cost, its imperative to understand how cost-sensitive the
customers and the market are.
As Bob explained the peculiarities of each customer, I found I needed
to further subdivide and categorize the habits of each individual buyer
with whom we dealt (ofen the customers had more than one buyer). Tey
all had their unique personalities and methods, which dictated how they
needed to be handled. It quickly became my responsibility to take the esti-
mate, which assumed a standard hourly rate, and adjust that according to
the customer.
Tis was further complicated by how the request for quotation was
received. A rush job would ofen support a higher margin (or rate). But,
a standardized quote sheet that appeared to be sent out to a few other
companies deserved a sharpening of the pencil, usually equating to a
lower margin. I also had to factor in how busy the shop was and how des-
perately I needed the work. Naturally, this always posed a bit of danger
because it was hard to know which parts were candidates for repeat orders
and how busy the shop would be when that order came in. So, a low price
during a slow time could come back to haunt me as a repeat order during
a busy period. Although I have said that I did not know which parts would
repeat, there were a few tell-tale signs that separated one-time orders from
repeaters. Te type of part, the revision level on the drawing, the type of
customer, the end product with which the part was associatedall helped
in determining whether the job was prototype or production work and
where it was in its life cycle. Previous experience with the customers as to
how they liked to work and which types of parts repeated most ofen was
also invaluable.
2.7 stArtIng soMe MeAsureMents
Regarding the quotations at this point in my deployment, the underlining
assumptions I was forced to accept were that the labor had been correctly
calculated, the necessary materials had all been accounted for in their cor-
rect quantities, and all necessary outsourced processes had been included.
We all know the danger of assumptions, but what choice did I have? At
that point, I did not have enough experience to question my estimator. But
then, the simple solution hit me: measure the actual against the estimates.
Tis would be straightforward to do for the time required inside our shop.
It could also be done with the materials by measuring what was actually
used and comparing it to the estimate. But my priorityand overriding
suspicionwas that the greatest diferences would likely lie in the esti-
mated versus actual labor times. It was not going to be complicated to
make a comparison since the estimator now followed the standardized
quotation form. All I needed to do was add columnsto be flled in by
the employeesfor the actual times (see the example in Appendix C). I
also added a column where any problems that were encountered could
be noted, providing a place to explain why the expectation was not met. I
would then circulate the quotation sheet with the documentation package
that followed all work orders throughout the shop.
Lean principle: Visual management, which ofen entails monitoring sta-
tus compared to the expectation (the actual vs. the goal), will only yield
results if the leader frequently monitors the visual (instead of gathering
this information from the computer, through meetings, or in discus-
sions), observing and questioning the deviations, and working with the
team to resolve the issues. Visuals should begin slowly as leaders need to
change their style and the culture as to how they obtain information. In
addition, it takes a lot of support to make improvements, and most orga-
nizations can only support a few areas simultaneously. Te team using
the visual must also understand its purpose and will only see its value if
management supports this by helping to resolve problems.
2.8 vIsuAlIzIng the MeAsureMent
It is always important to show respect and involve people in process
changes so they understand the overall contextthe signifcancewithin
the company. I explained to the entire team the importance of recording
their actual times: Tese hours were multiplied by a fgure that covered
all costs related to the business. If we hit or improved on the target, we
would be on time to the customer and within the budget, which would
allow us to make sufcient money to pay wages and secure jobs. I also
explained that if a job took more time than what was estimated, it was OK,
but we needed to understand why and learn from this. Everyone seemed
to understand, though I think John felt the most pressure since we were
measuring against his estimates. Tis pressure to learn and improve is
another advantage of visualization.
I noticed estimates were now taking a little longer, and I assumed John
was taking more time to review his numbers. He and I had agreed we
would go over the feedback together on Friday afernoonsnot to pick
through it, looking for mistakes, but to see whether we could continue to
improve the estimating process.
So, as each work packet was introduced to the shop foor, I stapled the
newly formatted estimate sheet to the front. John had already flled in his
estimated times on the sheet, and next to the actual hours column was a
column where problems could be noted.
As I began running around, anxiously checking how these sheets were
getting flled in and how the numbers compared, everyone complied and
completed them to the best of their abilities. We had essentially estab-
lished a feedback loop for our estimates.
Key point: If a new measurement is introduced (which I did, by record-
ing actual vs. estimated hours) and it is not important enough in the
leaders mind to drive him or her to continuously be seen monitoring it,
it is guaranteed to fail.
A manager needs not only to monitor any new visual measurement but
also to understand the root cause of the problems that keep a goal from
being achieved. Ofen, how the data is input and the way in which the
measurement is calculated will likely require further refnements.
2.9 PlAn, do, check, Act (PdcA)
Six years afer selling the business, I continue to be amazed as a consul-
tant to repeatedly fnd few estimating processes with feedback systems. It
seems most managers have never worked in this end of the business and
do not realize how many assumptions are used, how much pure guesswork
goes in, and the opportunities they are missing in terms of continuous
improvement to the process without some feedback system. Tey seem to
be content with explanations of complicated quoting methodologies and
studying the big spreadsheets they are shown, along with a belief that the
responsible people have enough experience to determine the correct costs.
If managers had some feedback and better accounting methods, I think
they would really be surprised at the error rate. I encourage many of my
clients to diligently track actual costs incurred for just one or two difcult
jobs and compare them to their estimates. Tose who have done this with
any kind of resolve have realized there are a host of opportunities.
Another method I encourage as a reality check is to choose an estimate
that includes a lot of labor hours and ask the estimator (usually during
a review with the estimating team) to look over the breakdown of labor
hours. Ten, target an area that has a lot of labor hours, one in which
you feel there is likely to be inaccuracies, and suggest that the estimated
labor hours should be reduced by 15 percent, for example, and see if you
invoke discussion or disagreement. If you fnd no strong disagreement
and are not challenged, you likely have a lot of guesswork going on and
should question the process being used to determine the numbers. Always
remember not to make it personal but bring the questioning and improve-
ments back to the process utilized to establish the numbers. Always center
on improving the process.
Lean principle: A fundamental principle within Toyotas management
system and necessary in all lean transformations is to instill PDCA as
a necessary way to behave for all levels in the organization. Tis very
simple principle becomes only a theoretical idea within frefghting
organizations that are always moving full speed into the next problem.
It is not only a way to solve problems; it should also be part of all actions
and processes we support.
I had identifed a process in desperate need of utilizing a PDCA loop
(see Figure 2.1) to drive improvement. Te estimating process only
employed the Plan and Do parts by prioritizing estimates and then
going through the process to arrive at a cost. But, there was no Check
for the accuracy of the estimate versus the actual costs incurred and
nobody Acted to learn and standardize improvements as part of the
process before utilizing the process for the next quotation.
Up to this point, I had only implemented a method to verify the accuracy
of the direct labor hours and the direct materials. I will come back to this
because I found I was only scratching the surface of how complicated this
process really is, even for a small company. Tese estimates are vital since
they are so closely linked with proft margins and how well you compre-
hend where you are making money versus breaking even or even losing
money. My remaining time with Bob was ticking away and I had only a
few more weeks to complete my education (pulling the tribal knowledge
Plan Do
Check Act
Figure 2.1
out of his head) before he was of fshing and manufacturing fshing reels,
which was his real desire afer he retired. It seemed his way to combine his
work experience of running a machine shop and his hobby of fshing was
to design and manufacture reels. Who was I to question this? I guess once
machining is in your blood you never want to be too far from it.
Had I not later learned the details of each process step for which a time
estimate was required, I do not think I could have ever have arrived at
realistic estimates. A good basis for estimating depends on having some-
one with an intimate understanding of the processes for which he or she is
responsible to quote. Aferwards, estimators need a feedback system with
specifc data to reinforce where they were correct or incorrect so they can
continue to improve their skills.
Key point: Te quotation process is key in every business and every step
of the process should be standardized and continuously refned. To have
a good system, it is necessary to continuously provide feedback on how
quotations compare to actual costs and learn from this to improve the
accuracy (a PDCA loop for making continuous improvement). Tis also
has a strong link to activity-based costing, which will be discussed later.
2.10 suMMAry
Prioritization of processes to document and standardize should be
based on the tasks for which you have only one or two qualifed
team members trained within the organization. In my case, it was
easy because I was aware of Bobs imminent departure (so program-
ming and estimating were at the top of my list), though ofen we do
not know who will leave the organization and are frequently caught
of guard.
Complicated processes are ofen a high priority for instilling a PDCA
loop. In my case, the estimation task was one of those.
Measurements are unlikely to improve if they are not important
enough for frequent follow-up by managers.
29
3
Learning the Offce Processes
3.1 IntroductIon
With only a few weeks of training lef, it was time for me to learn the ofce
processes that the former owner, Bob, had always handled and my mother
had not begun learning or documenting. I had been focusing on what I
considered critical skills in the shop that had previously been handled
exclusively by Bob, and now I needed to shif my focus to the front ofce.
(I continue to call it the front ofce because we also had a small ofce
in the shop where the programming was done along with other tasks
attendance records, completing critical measurements, flling in quality
sheetsthat were performed by the foreman.) I knew that before I had
enough background to make any decisions or improvements, I needed to
understand the processes in sufcient detail to ask reasonable questions
and make educated choices.
In addition to certain purchasing, planning, and accounting tasks that
Bob regularly reviewed and made decisions about, the front ofce also
did some part number marking, label adhering, packaging of the smaller
parts, preparing shipping labels, scheduling the shipping, and invoicing.
Because nearly half the orders were for new part numbers (parts we had
not previously manufactured), there always appeared to be a lot to set up
in the computer systems and a lot of materials that required specifying
and ordering. I knew there were bound to be a lot of non-value added
steps being performed within these processes, and my mother had already
indicated that both documentation and standardization were lacking.
Before we could begin making improvements and changes, we would have
to apply a simple OSKKK strategy (Observe, Standardize, Kaizen Flow &
Process, Kaizen Equipment, and Kaizen Layout).
3.2 InPuttIng orders
Although it was not one of Bobs responsibilities, inputting an order into
the computer was my frst task of the morning. My mother had already
documented the task and written a standardized procedure, but I would
also need to understand it for overseeing the day-to-day operations. As I
said, I was not learning how to input orders because I might be expected to
do it; rather, I needed to understand where the pitfalls were in the process
so I could later make improvements and handle the task if necessary. Using
Kaizen eyes means understanding the detailed steps and questioning
whether or not each step adds value. Toyota managers usually know how
to perform the majority of tasks performed by their direct reports (natu-
rally, this does not apply to highly technical/complicated disciplines). I
doubt that can be said across the board for the majority of managers I have
come across. Tey usually feel they control too diverse or large an area
for this and their function is to get employees to do things by motivation
and organization, forgetting they need to bring out the operators natural
problem solving abilities. I would have preferred that all my employees
had Kaizen eyes and would fnd the waste on their own, but I knew they
had not been taught or given the responsibility to suggest and implement
changes. Tat would come. In the meantime, it would be my responsibility
to identify the opportunities and teach through example.
Our version of an order entry and MRP (material requirements plan-
ning) system was an old DOS-based program in which all details of the
ordersexcept for the BOM (bill of material)were entered. (Just to be
clear, I did say DOS, that terrible old operating system of the past.) Te
program had been locally developed for job shops, and my business had
been one of the test sites. Almost eighteen years of data had been entered
into it. Unlike most MRP systems, the program only included a blank
screen for inputting and tracking of materials in a free format, and it was
not capable of monitoring incoming materials and linking their arrival so
that the corresponding job could be scheduled. Terefore, I was inclined
LearningtheOfceProcesses 31
to wonder how helpful it was as an MRP system (although MRP is not as
critical to the made-to-order functionality of a business like mine as it
might be elsewhere). It simply scheduled orders based on due dates with-
out taking material or manufacturing lead time into account. It might
have had additional functionalitybut if it did, nobody was aware of it.
A new Windows-based version of the sofware was available for more
than $15,000; however, money was a little tight at this point, as you can
imagine. And because I needed to better understand the business before
making any investment decisions, the old program would have to do for
the time being. Although my comments in Chapter 2 indicated I did not
want to return to using DOS, this program seemed friendly enough at frst
glance and it served the purpose of tracking customer orders, invoicing,
and providing materials and parts histories. Actually, inputting data into
the system was fairly simple and it seemed you could pull out most neces-
sary reports, which was ironic, in my opinion, for a DOS-based system.
Lean principle: MRP systems were not a fundamental principle within
the Toyota Production System. Instead, the tendency was to use Kanban
systems where appropriate for material replenishment, which is linked
directly to the customers actual consumption patterns and helps in visu-
alizing material levels allowing everyone to participate in a PDCA (Plan,
Do, Check, Act) cycle and constantly refne the min/max levels. Job shops
like minewhere the product is made to ordercould only order most
materials afer receiving a customer order, so MRP systems are less appli-
cable. I put in place some Kanban systems for standard materials (those
that we could stock with low risk) and for consumable items.
One small change I implemented early on was that all new part numbers
had their estimated setup times and run times entered into the program.
Naturally, these became the basis for costing and for shop-foor planning.
Since the frst established times (in hours and minutes) were only esti-
mates, we required that part of our PDCA cycle also included entering
the actual times into the program. Once the actual times were entered,
we could decide which timeestimated or actualwas to be used as the
standard basis for costing and planning on any future orders.
3.3 ProbleMs WorkIng WIthout A PdcA
cycle In the estIMAtIng Process
Tree problems I found without any actual feedback on our estimates were
(1) Bobs estimates did not have any structured feedback on the accuracy
of his estimated labor hours (as discussed in the previous chapter); (2) if
the hours were not correct, we could not accurately plan and determine
available capacity; and (3) if the hours were not correct, then the proft
margins were also not correct and all business decisions were likely based
on poor costing data. If we got this wrong too ofen, we would randomly
be overcharging for some parts while undercharging for others.
Although some might think this balances itself out, I had previously
witnessed managers in other companies not accepting new orders for
products they felt had too low a margin or pushing products they believed
had a high margin. Tat might work if the data is correct. But, I found the
labor times in some cases to be far enough of that some managers were not
making the margin they thought on those cash cows and other products
were really making a higher margin than shown on paper. Tis is further
obscured by standard costing and allocations, subjects that deserve a lot
of attention and will be discussed throughout the book. For now, I will
summarize this as a key point.
Key point: If you do not have some type of feedback or verifcation sys-
tem for labor hours and are not continuously learning about inaccura-
cies and adjusting them, you likely have many errors and a poor basis
for costing and planning. It is also possible you do not have a clear pic-
ture of your proft margin for each part number or family of parts.
I found a number of occasions where I improved the accuracy of this
data and feel it helped me make good decisions; other times, I suspected
we made rather poor decisions based on the lack of accurate data. Proft
margins are further and sometimes more strongly infuenced by account-
ing methodsbut that discovery was to come later.
3.4 rePrIcIng dIscussIons Are eAsIer
If bAsed on AccurAte dAtA
Later, when I found customers pushing for price reductions (or asking
us to match a target price) for certain parts that had started to receive
more regular orders, armed with more accurate labor and materiaI costs
I now had a basis from which to make a good decision. Periodically, the
customers asked for price reductions of anywhere from 10 percent to
25 percent, or threatened to try to fnd another supplier to meet their
target price. As our feedback system for quotations continuously pro-
gressed, we were better able to understand our true labor and overhead
costs as they applied to each product and part number. Combined with
accurate material costs, this put us in a much better position to analyze
and respond to these requests. With sufcient confdence, we could now
understand exactly what proft would remain if we agreed to these new
prices. In almost 50 percent of the cases, we were able to come close to
the target price (and maintain enough proft to justify the order); in the
other cases, we confdently stuck with our price and took our chances
by not ofering the customers a new price, as we knew our hourly rate or
margin would be unacceptable.
Lean principle: Kaizen eyes are something developed over the long
term. This involves the ability to recognize/identify all types of non-
value added activities through structured observation. One basic
part of having Kaizen eyes is being skilled with motion Kaizen
where you learn to focus on the employees hands, feet, and eyes
as they are performing a shop-f loor or administrative activityto
identify the traditional seven wastes within the tasks work elements.
This includes identifying the wastes of overproduction, correction,
motion, material movement, waiting, inventory, overprocessing, and
the recently added waste of not involving/engaging the mind of the
operator. To complete my training with Toyota, I was continuously
tested by being asked to observe a task and in a few minutes identify
ten wastes (or provide ten ideas for improvement).
3.5 leArnIng to PurchAse
Te ofce was managed by Sandra, who had been with the company for
more than ten years. She seemed the picture of efciency as she bustled
around the small ofce, working with her assistant who did a lot of the
repetitive tasks and the grunt work. It was Sandra who taught me the next
task: how to purchase material. What I found pleasant in the ofce, and
I later learned some of the customers also enjoyed, was Sandras accent.
She was British and although she had lived in the United States for many
years she still spoke the Queens English. Whatever it takes to keep the
customers happy, I just hoped that when I answered the phone with my
plain voice and simple Midwestern drawl, I did not send anyone running
to the competition.
Sandra would periodically wait for a day or two before ordering to see
whether we received multiple POs (purchase orders) requiring similar
materials. Purchasing materials in large quantities qualifed us for volume
discounts as our suppliers generally priced by the total weight of the steel
or aluminum we were buying. Sandra felt the savings frequently justifed
the delay. Naturally, if the PO had a short lead time, the order was placed
before the end of the day, and this was ofen the case, as we worked in an
industry that demanded short lead times. Being a lean practitioner, I was
constantly thinking about lead time, so I wanted to better understand the
cost versus savings related to this delaybut that would come later.
A simple PO format was used. Each blank form was flled out by hand
and had a unique PO number on it. (It was usually not necessary to frst
complete a purchase requisition and then have it approved. As a made-
to-order business, we were mostly purchasing direct materials for a frm,
short-term customer order.) For each diferent supplier involved, one PO
form was completed with information like material type, quantity, job
number, due date, etc. Only the suppliers name was omitted since deter-
mining this was the function of purchasing (Sandra or her assistant). Te
completed forms were fled according to the promise date on which the
material was due to arrive; however, they were put into a fle drawerout
of sightand only referenced as material arrived.
My idea was to use a simple visual system as a start. I asked Sandra to
keep the completed POs in our receiving area, out in the open in a simple
fle that had a slot for each day of the month. Each PO would be fled in the
slot that corresponded to the promised delivery date. When the material
arrived for that order and all was accounted for, the form was to be moved
to another fle to be reconciled with the invoice prior to payment.
Moving the POs from the receiving area fle to the reconciliation fle gave
us an easy, visual method of identifying outstanding POslate orders. We
could now follow up on these late materials easily, and I did not have to
check with Sandra each day about them. It was safe to assume that POs
that had been moved from the receiving fle had either been fulflled or
Sandra was following up on them.
I also suggested we could use these POs to track which suppliers were
frequently late. I wanted to see how we could become more proactive with
a late order, not just recognizing it afer the due date passed. But, for now
we had a simple visual system that showed abnormalities. Sandra indi-
cated she had a good feel for which were our worst suppliers, but I sug-
gested that we always speak based on data in the future.
Key point: Visualizing abnormalities as early as possible for all admin-
istrative and shop-foor processes should be every managers focus. Tis
requires defning the diference between normal and abnormal and
showing this visually and in a timely fashion.
Key point: Heres a good yet simple business tactic: If you fnd yourself
talking in circles because you do not have data available to make good or
clear decisions, then stop and put a quick method in place to gather some
simple data. Ofen, a simple tick list kept on the corner of someones desk
is sufcient. You will learn a lot, and nobody can argue with hard data.
Another good point to highlight is that, for many processes, it is fairly
easy to put a visual in place to track whether things are running on time or
behind, or whether or not things are functioning normally. Since my job
as a business manager was to understand when a process was in trouble,
to detect this as early as possible and then provide help or guidance, it
was key to make this apparent as early as possible for every task. Ten,
if everything was running according to plan I could sit around drinking
cofee all dayright?
3.6 schedulIng the shoP bAsed
on trIbAl knoWledge
Afer the orders had been entered and POs issued for the necessary mate-
rials, we could print out a shop-foor priority chart through our very basic
computer system. Te problem: our system only sorted the orders by cus-
tomer due dates and it did not contain their routings, required labor hours,
or the status of whether materials were available. Sandra informed me this
had not been a problem since Bob had an idea in his head about how many
hours each job required, and once the material was available he could
prioritize the jobs by how long he felt they would take to get through the
shop. (It was no surprise to me that all this had previously been managed
in this way.) She also seemed to think that a quick glance at any drawing
allowed Bob to understand all internal and external process steps (i.e., he
could visualize the entire routing for the part). Once again, alarms were
going of in my head: Bobs tribal knowledge was critical for yet another
step and I questioned how well he could manage the fve to twenty diverse
part numbers that moved through our shop each day. I needed to quickly
fnd a way to put some system in place as I had no system in my head that
could balance these fve to twenty part numbers each day, and since I
wanted to grow the business, this would only become more difcult. Also,
as a newcomer to the machining industry, I had no tribal knowledge, and
instead of the confdence that I could properly prioritize all this in my
head, I had only questions and concerns. I also questioned how efectively
this could be performed in Bobs head and what efect his mistakes would
have on lead time and on-time delivery. On the positive side, the new
orders now included the estimated times on each work package (as pre-
viously discussed) and the routing through the various operations. But,
repeat orderswhich accounted for about 60 percent of our volumedid
not currently include these time estimates and, to further complicate this,
the priority list (when it was printed) showed none of this information.
Again, all these issues were being discovered by frst observing and docu-
menting, and then working toward standardization.
3.7 Job-shoP PlAnnIng
As a general theme, job-shop planning is not only an interesting subject, it
is almost an oxymoron. Tink about it: Nothing is consistent. Tere are no
forecasts and no standard orders. And I basically did not know what the
customer would ask for, but I knew they would likely need it immediately.
Te positive side of this from a lean perspective is the concept of working
just in time; you only build what customers want when they require it.
Systems like MRP or ERP (enterprise resources planning) cannot ofer
much help since they work on two assumptions that likely do not hold
true in this type of business (and rarely hold true for any made-to-order
company). Tese systems assume accurate forecasts and are unaware of
changes in lead time (unless you are constantly updating the lead time
within the computer). In my case, forecasts did not exist, and if they were
available, they would likely be inaccurate. It should be noted, however,
that periodically we received an order with split delivery dates, although
even those dates frequently were changed.
Constant lead time is another oxymoron within most businesses.
Demand for each machine or process usually varies hour by hour or day
by day based on customer demands, and in the case of a job shop, you have
little infuence on customer demand (other than not accepting the order).
So, lead time is constantly changing based on the queue of work waiting
at that particular process or the amount of work the external plating or
painting shops have waiting. For example, at the beginning of the week
you may have only one day of work waiting in front of a certain machine,
but by Tursday you may have received enough orders to fully utilize this
machines capacity for the entire next week. In this case, lead time will
have gone from just more than one day to more than one week. Terefore,
it appears these MRP or ERP systems were created for another type of
business, predominantly those that build in either large batch production,
have fewer part numbers, have stable forecasts, or their capacity is more
evenly balanced. I think this is the reason people in made-to-order busi-
nesses ofen refer to MRP as More wRong Parts.
So, what can a job shop do to plan and accept only orders for which it
can meet the customers delivery date?
In my case, Bob had always worked all this out either in his head or on
a scrap piece of paper, which presented a few problems for the new guy on
the block. I was not as familiar with all the jobs as Bob appeared to be, and I
might not be around every day to develop the schedule or confrm whether
we could accept an order for the date the customer was requesting. (My mas-
ter plan was to develop a business that would allow me some freedom from
the day-to-day operations.) Terefore, the simple print-out from our sys-
temwhich only sorted the jobs by due date, did not relate their lead times to
the various routings though the diferent processes, and showed no relation
to working timewas not a method I could use going forward. Also, I was
fnding it nearly impossible to standardize Bobs thought process. For the
short term, my task was to quickly learn the methods through observation
and then standardize them, though this one would have to wait until I better
understood the related processes and could develop my own standardized
method as the baseline for improvement. In made-to-order businesses, mate-
rials planning is ofen more straightforward than capacity planning, which is
more difcult because of the continuously changing lead times. My interim
plan was to visualize the current workload for each process and to display
this in real time while quantifying where the available capacity resided,
essentially helping to determine the ever-changing lead time.
Lean principle: JIT (just-in-time) is an inventory strategy to improve
efciency and minimize waste by receiving goods only as they are
required for the production process, thereby reducing most of the costs
associated with inventory. Tis method requires that manufacturers are
able to accurately forecast demand, which in our case was also easy as
we performed no forecasting and only built to a frm order shortly before
the required ship date. Since we only ordered materials once we had
an order and had the available capacity to produce it, we were already
working to JITone of the advantages of working in a made-to-order
business or a job shop.
3.8 sIMPlIfIed short-terM PlAnnIng
Short-term shop-foor planning should involve ensuring you have the mate-
rials and capacity available before you start the work, based on its rout-
ing through the various processes and the lead times associated with the
required steps. Hopefully, this style of planning allows the part to be com-
pleted in time to arrive at the customer on or before the due date. Again, it
doesnt sound difcult when I sum it up in a simple sentence but there are
many nasty little surprisesor should I say variablesthat complicate
things. As a consultant, I encounter many companies that only material
plan. Ofen, this means that by the time they receive sufcient materials to
start on an order, they are behind schedule, and so they immediately release
the job to production and do not perform any capacity planning. Tis means
they cannot proactively analyze where the bottlenecks will be and work to
minimize themthereby becoming only reactionary frefghters.
Obviously, the frst concern is that the materials arrive on time so
that the work does not have to be expedited within your own process.
Experiences with poor suppliers have led some planners to infate delivery
times in the computer systems, thereby trying to avoid being caught out.
Tis just adds to inventory. In my business, I basically had no idea what
type of material would be required until I quoted the job. I could keep
some basics in stock that could be cut to size to reduce lead time (though
this had labor and inventory costs/risks associated with it), and there were
actually many suppliers that could provide most standard materials cut
to size within one to two days. So, that was not my biggest problem. Te
problem was the more exotic materials that came longer distances with
fewer choices of suppliers. Tat aside, I found historically there had been
no tracking of the suppliers on-time delivery, but I fgured that, with
my new system in the receiving area, it would be easy enough to visualize
when things arrived late. From previous experience, I had learned that
the really detailed planning (process by process) should not be performed
for a period of longer than twenty-four hours. Terefore, the best answer
was a planning board at each critical process that displayed the work
planned and showed the actual status periodically updated by the opera-
tor. Later, I implemented this system, and it is discussed in this book.
3.9 A More coMPrehensIve Work order PAckAge
Leaving the discussion on the planning system for a later point when I
arrived to kaizening of the process in my OSKKK strategy and also
achieved a better understanding of all the factors involved, I decided to
move on to learning other tasks. One of the other functions performed in
the front ofce was putting together the shop-foor work packages, which
consisted of only the customers drawing; a copy of the purchase order for
the materials; and a short form showing due date, any additional outside
processing, and any special instructions (like part to be supplied with-
out hardware or group with job number X for outside plating). Tese
were pulled together a day or more in advance and Bob would then decide
which machine they went on.
Te work package did not indicate whether it was the frst time we were
producing this part. Early identifcation of a frst-time build was critical
because it might require programming the part frst, increasing lead time.
Existing parts, on the other hand, already had developed programs along
with tooling and special fxturing when required. Naturally, this afected
the number of steps and the lead time through the system. For example,
a new part that required programming could add a half day or more to
the lead time depending on the programmers workload and the number
of programs (operations) the part required (typically, each separate CNC
[Computer Numerical Control] operation required a separate program).
Normally, any special tooling or additional process steps were identi-
fed during the quotation. But, from time to time a non-stock tool or an
additional external stress relief might be identifed during programming.
Terefore, the earlier all processes were identifed (e.g., the programming
step previously discussed), the fewer the chances for last-minute delays
due to some unforeseen requirement, thereby making it easier to deter-
mine an accurate routing and lead time and thereby maintaining high
on-time delivery performance.
Either Bob or Tom (the shop foreman) generally managed these plan-
ning/programming details in their heads. I did not want to change the
process at that point, but I felt that I could at least add a step. On the
work package form, I wanted to add a place for the front ofce to indicate
whether or not a program already existed, plus any additional external
process steps that may be required. Further, to allow a quick check for
any potential tool shortages, I asked for a preprint of the tool list from
the program. At least now I would know whether a program existed and
which outside processes were involved. Tat information would allow me
to more accurately determine the lead times and then organize priorities
for scheduling.
3.10 Poor Work order PAckAges
cAn IncreAse setuP tIMe
I was also thinking about how a poor shop-foor work package can increase
setup time. In many cases, half of the setup time was used for adjusting and
checking the part. Ideally, we should have been able to return as quickly as
possible to the parameters of a previously manufactured part (which had
been built to an acceptable quality standard). To do this, some notes regard-
ing the previous program, tooling, or the fxturing modifcationsall the
subtletieswould be required in order to take advantage of the information
and proft from it. Naturally, a small amount of time would be invested afer
a production run in capturing these details. In our case, we manufactured
thousands of diferent part numbers and although about 60 percent received
repeat orders at some point, the repeaters were so random and unpredict-
able that the operators could not readily recall the specifcs of any particular
set of parts. I did not feel it was the proper time to tackle this issue, but when
the time came to focus on setup time reductions, this was high on my list.
I continue to work with companies that have made the same part hun-
dreds of times, but because they did not document the peculiarities and
fne adjustments of previous production runs, they always fnd themselves
starting over. Tey pay the price each time they set up. Instead, they should
be racking up the proft from their previous experiences.
Tere were many other small tasks performed in the front ofce such as
the weekly payroll (another area my mother was overseeing during this
transition period). Paychecks were issued on Mondays. Bob had imple-
mented this policy to ensure everyone showed up to work each Monday,
traditionally the poorest attendance day for hourly employees. Other
small tasks like invoicing, chasing late payments, and accounts payable
were jumbled in the mix. Ten, there were the fnancial aspects: closing
the books and the nonexistent budget.
3.11 closIng the books
My experience with most large companies is that people live and die by
the closing of the books. Te exception was Toyota, where I never heard
the typical question, How are the closing numbers shaping up? Instead,
I saw everything focused on waste elimination by those who controlled
the processes.
I know for certain that I do not fully understand discussions or state-
ments dealing with standard costs and all their related variances. As I look
around meeting rooms when the discussion centers around a standard
cost variance that has increased or decreased, I always think that nobody
else appears to understand the root causes behind these variances either.
Tere always seems to be a controller or accountant in the room trying to
translate these statements so somebody can decide what to do about the
numbers. For my part, despite all my MBA courses, I am almost never sure
how to take a negative variance and directly turn it into an appropriate
action that will transform it into a positive variance on the standard cost
without acquiring a lot more information on what might be behind that
particular variance.
On the other hand, in my own company when I reviewed one of our
end-of-month closing statements, it was so straightforward that I could
ask intelligent questions and draw conclusions that I had some faith in.
I had concerns that in its current state, it might be a little too simplifed
to draw detailed conclusions. For example, we looked at the sales result-
ing from the current months shipments, but we simply compared those
to the current months costs by taking totals of all checks issued/paid
out during that month (this is also practical for many items like direct
and indirect labor, which are usually paid out in the period accrued).
However, we were really paying for last months material purchases dur-
ing the current month, and depending on whether we consumed more
or less material last month, the proft and loss could be misleading
(although our proft and loss determined for time and materials on each
job was more accurate). Utilizing an average percentage of material costs
could also be misleading in our business as this varied depending on the
features of the job. Our accrual system for certain costs was also simpli-
fed and would likely require a better understanding before suggesting
any improvements.
Te advantage was that it only took a few hours to close the books. My
mother was temporarily handling this process and had taken responsi-
bility to document it and make suggestions to improve it. I also fgured
that in order to really understand proftability in a job shop, I would need
to look at it on an order-by-order basis. So, for the time being I decided
to stick with this simple closing of the books and then let an accountant
explain things like depreciation to me at tax time. Te important point
was that I wanted to be in the position where the end of the month num-
bers held no surprises, all of our other real-time measurements ideally
would focus our activities, and closing the books would simply confrm
what we already knew.
3.12 collectIng the Money
Like all businesses, receivables were critical, though I had the feeling that
in a small business they were even more so. Bob kept a close eye on any
invoices that went beyond the standard thirty-day terms. Tankfully,
our computer system printed a simple report and grouped outstanding
invoices by those over thirty days, those over sixty days, and any that
might hit that dreaded limit of over ninety days. Apparently, those reach-
ing forty to forty-fve days outstanding initiated phone calls; afer that,
Bob usually told Sandra to be a little more forceful and he would start get-
ting involved. Since we were typically dealing with the same seven or eight
customers, we were usually working on a newer order and could always
threaten to delay delivery as a last resort. Here I saw no need to change
any of the current practices other than simplifying the measurement unit
to DSO (days sales outstanding) and planned to keep the payment terms
and the collection methods the same for the time being. DSOs also take
into account the time it takes to prepare and send the invoice, a process I
felt could use some improvements.
Lean principle: Its critical to focus on key measures and present them
in a simple format so that everyone can understand what infuences
the numbers. When measuring receivables, the recommended unit of
measurement is DSO: days sales outstanding. DSO is typically defned
as the average number of days it takes to collect the payment afer a
sale has been made. Te unit of measurement is dayswhich is easy
to relate to payment terms that are also in daysthereby quickly
making a determination whether a business is operating within a nor-
mal or an abnormal range.
Obtaining quotes for raw materials and subcontracted process steps was
also a responsibility of the front ofce. I have already explained some of
the limitations in the quoting system that was in place at the time I pur-
chased the business, one of which was that Bob did not use a clear format
to indicate what materials were required and what further processing was
necessary. I have discussed the new quoting sheet that John (the program-
mer/estimator) was starting for each part he quoted (shown in Appendix
C), and which Sandra then completed by adding the material and subcon-
tract prices and determining the awarded suppliers name (based on qual-
ity, delivery, and price). Te only faw in the system was that Sandra and
her assistant had all the suppliers names in their heads and some of the
phone numbers programmed into the speed dial. As usual, I was thinking
this needed to be standardized, and my overriding concern was the pos-
sibility that I would be required to complete the task myself in the event
neither Sandra nor her assistant were available (i.e., absent or quit). I put
together a simple tableincluding the type of material, the typical suppli-
ers we would use, who to contact, and a place for additional notesand
they began to populate it.
Tese were the majority of critical front-ofce tasks in which Bob was
involved. And although the plan was for Sandra to continue working with
us as the ofce manager, she had a close relationship with Bob, and I felt
she worked there mostly to support him. Terefore, she might leave at any
moment now that Bob was almost out of the picture. Just one more reason
to quickly learn, document, and standardize. Having my mother become
familiar with all the processes was a good backup.
3.13 suMMAry
Made-to-order businesses (and job shops) are already working to JIT
standards; therefore, systems in the front ofce must support short
lead-time orders (i.e., MRP systems ofen do not directly support the
predominant characteristics of these companies).
PDCA again proved critical in the administrative processes, with
most tasks requiring a feedback loop and measurements to support
continuous improvement.
Developing standardized work for business (administrative) processes
is just as critical as having standards for the shop-foor processes.
Although the OSKKK process indicates standardization is the step
prior to starting any Kaizen, I feel there are instances where imple-
menting an improvement should be done instead of standardizing
a poor practice and then improving it. Tough in most cases I was
just imagining the improvement at this time or adding an enhance-
ment to the existing process, the majority of ideas would only be
implemented when I reached the KKK point in my transforma-
tion plan.
47
4
The First Days Alone
4.1 IntroductIon
At the end of October, the former owner, Bob, was leaving so we closed the
books together with help from Sandra (the ofce manager), who typically
handled this. It had been my frst full month owning the business and it was
the acid test for me to gauge how I was doing. It is similar to being a man-
ager and getting rewarded if the monthly targets are met except it is a whole
lot more personal. It was especially critical for me as my $75,000 of seed
money to carry the business through until our receivables starting coming
in was basically bottomed out. We had a reasonable month in shipments
but I was unsure of the total expenses, especially during the last week. Afer
the numbers were checked and double checked, I had good newsnot nec-
essarily great news but good enough: despite making the monstrous loan
payment and the added expense of a new programmer, we still had a posi-
tive cash fow. I had only paid myself a $2,000 monthly salary, which could
not last forever, but I was surviving and damn proud of it.
With Bobs last day and a hearty handshake, I was now on my own and,
although content for the time being, next months numbers would really
separate the men from the boys. I fgured if I could make it a month on
my own and still be proftable, there might be an entrepreneurial future
for me. I was aware, however, that I could not live on a monthly salary of
$2,000 for long and if sales had been a bit lower or expenses a bit higher,
I would have either drawn no salary or written the company a check. So,
there was not a lot of room to sit back and relax.
Te obvious solution was that I needed to increase sales and/or propor-
tionally reduce expenses. As a lean guy, I was thinking of productiv-
ity improvements for all processes. Since Bob already kept expenses to a
minimumin other words, there were no big corporate perks to cutI
just wanted to get to the point at which he had arrived and have a few com-
pany cars foating around while paying myself a decent wage. For the time
being, with a big loan payment and a new programmer on the payroll, that
was not in the cards until we increased sales.
Although I was still single, I was involved in a serious relationship and
was contemplating the idea of popping the question some day soon. Not
knowing much about the opposite sex or having a lifelong partner, I was
just sharp enough to assume I needed a more stable fnancial situation
before anyone could comfortably commit. I had done some mental gym-
nastics and fgured if I could get six proftable months under my belt and
I could pay myself enough to live comfortably in the San Francisco Bay
area, then I could more seriously consider popping the question.
It was now Monday morning of week seven as the owner, which meant
the training period had ended and I was on my own. Bob had indicated I
could call him if I had questions. I had my mind made up to only call in an
emergency. In no way would I say I felt confdent to take over. Six weeks of
training had fnished and I now knew how to turn on the lights and had a
rough idea of a few other tasks. But, competently performing all that Bob
had been responsible for was another matter. Tis frst day alone had some
similarities to starting a new job when you reach the point where you are
starting to make your own decisions. Te major diference with a job is
the greatest downside could be unemployment and possibly stalling your
career. When you own your own business, you could upset a few large
customers and bankrupt the business, which in my case meant losing my
life savings and becoming indebted to the bank for a long time to come,
not to mention putting other people out of work.
4.2 My fIrst PrIcIng decIsIon Alone
Monday morning, my frst day alone at the helm, and everything was
rolling along fnewell, fne by machine shop standards, which seem to
be defned as OK if you only have a few minor problems simultaneously
TeFirstDaysAlone 49
on each machine. Te frst item of the daya request for quotationhad
been received. John (the programmer/estimator) had gone through and
completed the labor hours and required materials. Te front ofce had
quickly added in the material prices, and in short order it came to me for
approval. Te moment of truth had arrived. But, par for the course, it was
not an easy part to manufacture, and with a quantity of thirty pieces and
a high material value, roughly 38 percent of the total price (which is high
for a company that averaged 18 percent material costs), it was not an easy
decision without Bob. I had learned we could usually add a 15-percent
margin on top of our standard hourly rate for this customer and the par-
ticular buyer. Since it appeared there was no rush for this part, I decided
to take the conservative route and only add a 7-percent margin. Ten I
submitted the quotation and the waiting game began.
Historically, we had somewhere between a 60-percent and 70-percent
hit rate on receiving orders on our quotes. Tis number seemed reason-
able to me as there was a cost associated with each estimate. We wanted a
relatively high hit rate but too high would indicate we were undercharg-
ing. So, I set my personal target at 70 percent. For the time being, I still
lacked the experience to challenge Johns labor and material specifcations
but felt I would continue to track them against actual hours and costs and
review this every Friday (a complete PDCA [Plan, Do, Check, Act] loop).
We received the purchase order for this quotation and hoped it would yield
the margin we expected and would not turn out to be a loss because of a
poor estimate. At least we would measure and confrm this.
4.3 creAtIng extrA cAPAcIty WIth 5s
Creating extra capacity was naturally a theme to start on quickly.
Workplace organization (known as 5S to lean practitioners) is a simple but
efective tool to kick this of. Tis is especially true where time is lost look-
ing for items, which results in decreasing the efectiveness of a process or
an employee (especially where employees rotate between workstations and
jobs). Bob had traditionally run and organized the shop like his own base-
ment, meaning he knew where everything was and the others just had to
fgure it out. Te fact that things were not always in the most practical and
easily accessible places was just accepted. So, I fgured this would be an
easy win but wanted to let Tom (the shop foreman) and John handle this
afer I asked just enough questions to let them draw their own conclusions
about the current state. One advantage was John had previously worked
in other shops so he brought a new perspective that Tom did not have,
and I was sure he had observed superior workplace organization with his
former employers.
During my frst six weeks, I had noticed a lot of time being wasted by
the setup men looking for the right cutting tools for the CNC (Computer
Numerical Control) machines. It was almost comical as all tools were kept
in mostly solid-colored plastic cases (to help protect the cutting edges
from being damaged) and grouped into cardboard boxes only accord-
ing to their basic diameter. Tis meant a lot of diferent types of tools
were mixed in a single cardboard box (e.g., a quarter-inch, two-fute high-
speed steel end mill was stored with a quarter-inch, four-fute carbide end
mill). Te plastic cases for specialized tools were labeled a little better, but
the setup men had to open many plastic cases to locate the tools called for
on the tool list.
I did not want to dictate my solution to John or Tom; instead, I men-
tioned the frustration and wasted time involved in looking for tools. Tey
agreed. So, I asked John whether the other shops where he had worked
had stored tools in the same fashion. He indicated he had seen various
methods but none as cheap as this. So, I asked what these other shops
did instead. He indicated most organized them in drawers within their
tool rooms, where diferent drawers contained dedicated sizes, and they
were further sorted within the drawers by specifc type. Also, they were
stored in slots within the drawers so everyone could easily identify the
specifcs of the tool. It was the type of answer I was looking for, and Tom
seemed to grasp the advantages but pointed out Bobs reluctance to spend
money on things like tool cabinets. I did not view it as spending money
but as wisely investing money that would quickly be paid back. Now was
my time to get even for all those years of working with larger companies
where even small investments required many approvals and the delays
were endless. I immediately ordered $1,500 worth of specialized tool
drawers and had them delivered within two days. Tom was surprised by
the swifness of my actions, and I told him that as long as there were
sufcient benefts and the cost was not great, we would move quickly.
Organizing the tools into these drawers suddenly had everyone excited
and it seemed they almost forgot about running the machines as they
TeFirstDaysAlone 51
ofered to start helping. I asked for a little thought to be given to deter-
mining which tools got used most ofen so that they would be given the
most logical placement with the easiest access. Now we were on a quest
toward better workplace organization.
Although many businesses are not currently throwing things in card-
board boxes, there are always items that accumulate in mysterious places
and much time is lost in trying to locate them. Te trick is awareness
and prioritization.
Lean principle: Poor workplace organization (5S) had now been shown
as a source of non-value added time during a changeover. I consider
making this improvement part of the standardization step in OSKKK
(Observe, Standardize, Kaizen Flow & Process, Kaizen Equipment, and
Kaizen Layout); therefore, it is an appropriate frst step that ofen leads
to productivity improvements. Te productivity improvements are most
applicable in organizations that have cross-training and job rotation
(i.e., diferent people performing the same job).
Naturally, workplace organization became a topic we continuously pur-
sued. It is not worth going into considerable detail as I think the concept is
easily understood. I can only say that to get it ingrained in the culture took a
bit of patience on my part as well as asking questions every time I observed
time being wasted looking for items. Also, when I noticed operators losing
time because they were sharing a fairly common tool like a fxture, a wrench,
or deburring tool, I would ask how ofen this happened and how much time
they imagined was lost either waiting for it, going to get it, or looking for it.
Ofen, we came to the conclusion that each workstation should have its own
tool, and without hesitation I told Tom or John to order what was neces-
sary to eliminate this forced waiting time. Afer a few weeks, I found the
team making suggestions on its own. I only had to ensure the time that was
being saved was used productively. Time being used productively would be
further ensured in the future with the addition of scheduling boards helping
to make certain that everyone always had enough value-adding work being
planned. Tom had also been trained in a culture where employees never sat
around or worked slower to appear busy, so I was lucky that each minute of
productive time we gained was making me more money as long as I kept
bringing in enough work to keep everyone busy.
4.4 estAblIshIng norMAl versus AbnorMAl
My real challenge continued to be defning normal versus abnormal for all
aspects of the business. If I could accomplish this, then I could concentrate
on quickly applying my problem-solving skills to minimize the efects of
the abnormalities.
In the shop, I relied on Toms experience to help clarify abnormalities.
It was not as easy as simply asking him What is going on here that is
abnormal? Unfortunately, this would not make any sense to him. Instead,
as events took placefor example, an end mill would breakI would ask,
Does this happen ofen? I would get responses like, Tis usually hap-
pens when we work in stainless steel. Terefore, I could determine that
we needed to replace an end mill afer it cut a certain amount of stain-
less. Of course, it was much better to replace the end mill before it broke
because if it broke while cutting a part, it was also likely to damage the
part and cause scrap. Te problem was the wear was not always apparent
when looking at the tool. However, most tools have a tool life (in terms of
time or the amount of a particular material they can cut). In shops where
you continuously run the same parts, it is easy to determine. For example,
afer machining 100 pieces, the tool should be replaced based on cutting
time. Our problem was that our average order was only for a few pieces
and we could not aford to replace all the tooling afer every stainless steel
job. Terefore, I proposed each tool be given a number and replaced in the
tool tray according to that number, and each tool tray had a place where
the total cutting hours could be tracked by shading in a calibrated line (see
Figure 4.1 for an example). Te line was only as long as the total available
cutting hours for which the tool was rated.
Lean principle: Standardization is what helps in clarifying what is nor-
mal and abnormal in the administrative and manufacturing or service
processes. Instead of accepting excessive frefghting as the norm, try to
determine a standard expectation and visualize when you are meeting
1 5 4 3 2 6 hr
Figure 4.1
TeFirstDaysAlone 53
the standard (i.e., a normal condition) versus when you are not meet-
ing the standard (i.e., an abnormal condition). Remember: your job
as a manager is to quickly realize when a process has crossed into the
abnormal range and to help to rectify the causes with a permanent
countermeasure.
We knew from the program the total cut time being utilized by each
tool, so the operator only needed to multiply by the number of pieces he
had manufactured and this was only required for stainless steel and a few
other hard materials. Tis also drove home the point that stainless steel
parts should not be priced (in other words, not have the same hourly rate)
as easier machining materials like aluminum. I was already beginning to
see all the pitfalls of Bobs simplistic $60/hour fat shop rate. Tis is an
example of where we identifed a problem because there was no standard,
so the process was Kaizened (the frst K in OSKKK), which drove us to
create a new standard process for the operators.
I decided that as each problem was encountered or when I was required
to make a decision, I would try either to develop a process that helped
minimize the problem or simplify the decision making. I also wanted to
work to understand when any process encountered an abnormal condi-
tion and try to move frefghting to the lowest level possible. In my case,
this basically meant that I wanted either my shop foreman or programmer
to handle most of the shop-foor problems but this would require some
training in problem-solving skills.
Previously, I noted that in many cases when Tom encountered a problem,
he would involve Bob before fully understanding the root cause. I basically
wanted my foreman to act as a team leader. Ten, as either the setup men or
operators encountered problems, Tom (acting as a team leader) would solve
them with little or no involvement from me. I did not want each operator
to struggle for some period of time and then call Tom. Instead, I wanted
him to be in the position to immediately respond and quickly resolve the
problems, and while he was working on the problem, the operators could
be assigned to other tasks like deburring, tumbling, or preparing the next
jobs material. I knew this would require some problem-solving training
nothing as in-depth as Six Sigma (we did not have the volume to apply sta-
tistical problem-solving methods)but some simplifed one-page problem
solving (also known as A3 problem solving) and simple thought processes
like the fve whys and fshbone diagrams. Basically, Tom just needed some
confdence and a few simple, structured methods, and I was sure he could
work through it.
Lean principle: Within a lean context, a team leader does not have
management authority over the team members but is challenged to keep
processes running, resolve problems, perform root cause analysis, fll in
for absenteeism, etc. Many of us who deal with the categorization of
direct versus indirect employees think that direct employees add more
value and are therefore easier to cost justify. Some managers view the
team leader as an indirect employee and this is unfortunate. Toyota
realized it is better to have directs only doing the direct work and spe-
cialists available to handle problems and irregularities. Tis allows the
directs to have productivities in the high 90-percentiles and skilled prob-
lem solvers resolving the issues at their root cause, yielding a better over-
all productivity when executed correctly.
Te second line of defense was my new programmer. In his position,
John already possessed the ability to think more abstractly in terms of what
was happening at diferent points in the program as the part progressed
through its various steps. I fgured if John and Tom received a little theory
and structure, they could handle most of the technical issues, though it
frst required coaching them through the reasoning behind root-cause
problem solving. Bob had always handled the problem solving and he had
not developed those skills within the organization. I did not want the busi-
ness to rely solely on my skills. Instead, I wanted everyone to develop these
abilities, which would free me from this as a daily responsibility.
Lean principle: Problem solving can be accomplished in as few as fve
steps:
1. State the problem as specifcally as possible.
2. Analyze the problem (other perspectives).
3. Formulate a wide range of solutions.
4. Evaluate solutions.
5. Choose solution to trial.
It can also evoke more complex methodologies like Six Sigma depend-
ing on the nature and characteristics of the issues. Another, ofen better
TeFirstDaysAlone 55
option is A3 problem solving, a structured approach used to engage those
involved and to help in developing future leaders.
4.5 ProbleM solvIng versus fIrefIghtIng
I really believe that some people thrive on frefghting, especially when
they solve a problem that is delaying a process and walk away as the hero.
Tis is not a personality trait that I possess. Instead, I always thought that,
with a problem-solving process in place for identifying problems early on,
it would be better motivation for all involved if the team leader led them
through the problem solving and then had the time to put a more per-
manent countermeasure in place. Although this sounds simple, it takes
a lot of time to evolve and it requires a manager who asks questions that
lead employees to their own solutions (instead of just telling people what
to do). I was not born with a lot of patience, so asking questions to allow
employees to draw their own conclusions is something I had to learn. My
Toyota training helped me along in developing these skills. Tis is a key
point since I feel it leads to more successful problem solving, which helped
to reduce a lot of our lost machining time.
Key point: When encountering problems, coach employees through
root-cause problem solving instead of assuming you already understand
the cause and/or the solution. Tis is much better than dictating what
should be done since evidence shows most of us are wrong about half the
time. It may take longer but you are engaging the brains of the entire
organization, and in the long run you will develop a team of problem
solvers who will ultimately minimize lost time by reacting quicker.
4.6 IdentIfyIng the need for shoP-floor vIsuAls
Defning normal versus abnormal for many processes was another prior-
ity for me, but I knew this would take some time and require observation
followed by standardization. Te more non-standard a process is, the more
difcult it is to say what is normal. Typically, when you ask, What normally
happens next? the answer is ofen, It depends on this or that. Terefore, I
wanted to identify the most critical processes frst and concentrate on those.
I knew some of them would be on the shop foor and others would be in the
ofce. Although both our quality level and on-time delivery rates were rea-
sonable, I knew based on our short lead times that our on-time delivery was
always vulnerable. Terefore, I decided to start some visual management on
the shop foor, not just to ensure we were running in a normal condition
and would deliver on time, but also to develop a simple process that ensured
the front ofce could decide when an additional order could be rolled into
the schedule. Tis would provide them a tool to easily commit to a delivery
date (in other words, creating a job-shop planning system). Previously, this
was all done by Bob and most of it took place in his head.
I started with the idea of a simple scheduling board for each machine.
Te board was to be used by the foreman, programmer, and me to indicate
whether we were on schedule (normal) or not (abnormal), and to help the
front ofce to determine the lead time for new customer orders. I felt the
boards needed to display the current days plan in detail and ideally have
this short-term capacity well utilized, plus have the option to display the
approximate workload going out for the next week. Tis was discussed in
the previous chapter and an example of the type of board we frst used is
shown in Appendix D.
We normally worked with such short lead times that anything requested
outside a two-week window was an exception and consequently rarely a
problem. I felt the scheduling board should function in hours because that
is how we estimated and planned each operational step. I also wanted the
ofce to consider pulling ahead some of the work that was not immedi-
ately due and to plan the schedule allowing sufcient lead time for any
external processing steps (plating, painting, stress relief, etc.). Tis would
provide the opportunity to group orders with other materials going out for
the same external processes, reducing cost and transportation.
I chose a simple dry-erasable board for each machine and planned to
start only on my critical machine (the Haas VF-4) as a pilot. Since most
of the CNCs had the same capabilities and were only diferentiated by the
size of the material they could handle, the largest machine was the one
most likely to become the bottleneck as it was the only one with the capac-
ity to manufacture the complete range of parts. Consequently, it was fre-
quently in higher demand.
TeFirstDaysAlone 57
Before purchasing the pilot board, I wanted to frst discuss the idea with
Tom and John and get their buy-in. I was not sure of Johns background
with visual management. But, I knew that during Toms eighteen years
at the company, visual management had not been embraced. Instead,
his experience was the frustration of trying to read Bobs mind for what
should be done next. I wanted his support in making things more trans-
parent. I explained the advantages: Tom and the operators would always
understand the plan for the day and the quantity of upcoming work. Te
expectations would be clear and I felt we could minimize any schedule
changes. Also, Tom and John could have input to improve the sequences,
helping minimize changeover time as long as it did not impact the on-
time delivery, and we would all quickly see other abnormalities.
It seems my sales pitch went well as they bought into it, and Tom felt
confdent the operators would also see the advantages. I basically wanted
Tom to handle scheduling on the boards and told him I would make a
simple fowchart to help him in following a standard process for determin-
ing priorities, ideally based on FIFO (frst in, frst out) between processes.
I wanted to push the order in at the front end (based on its lead time) and
then keep it fowing/moving. I also asked John to review the boards at least
once a day and give Tom any input if he saw a better changeover sequence
or felt Tom had not allowed enough time for the external processes or other
potential problems with the complicated parts. Starting with only the frst
board, the entire plant would not be completely optimized as many parts
could be built on various machines; therefore, the best sequence for change-
over could only be determined at a later time when all the machines were
viewed together. Although I was starting with a bit of suboptimization, it
is always better to pilot ideas on one process, make a few improvements to
the concept, and then ensure those involved see the benefts before rolling
it out across the plant. Imagine the buy-in to the schedule now that Tom
and John were going to be developing it each day.
Lean principle: FIFO (frst in, frst out) is a process usually associated
with material control, though it is also recommended for scheduling
work-in-process, especially in manufacturing where parts are routed
through second, third, and fourth steps, etc. Tis is critical in maintaining
lead times for all work-in-process. Since I was allowing my team leader
limited fexibility to change the sequence based on optimizing setups,
it might appear I had disregarded FIFO. I clearly only allowed reshuf-
fing within one shifs worth of production; therefore, the lead time to
move though a particular process would be the current queue time plus
a maximum of one shif. A particular part could only be shifed back-
wards one time, always allowing me to quantify when it would arrive at
the next step.
Appendix D shows the idea behind our pilot board with the in-trays
for the shops information packages (quote sheet, routing sheet, tool sheet,
setup, page, etc.). Tis was only the beginning, and this board changed in
format as we found better ways to plan our work. I found it critical to start
with a pilot, which is owned by those who use it; therefore, my role was to
coach and facilitate by providing a starting point and the guidelines as to
what should be accomplished, and let them create and customize the tool.
Key point: When starting any visual process, it is better to set the guide-
lines, clarify the purpose, and start out simple. Te entire team should
have input on taking the pilot/prototype (i.e. of the visual) through its
various improvement stages. You never get it right the frst time.
In all honesty, the reason I began visual management with a planning/
attainment board was to ensure I was made aware of all available capacity.
I could then try to sell it in the short term, thereby increasing my proft-
ability. Selling unsold capacity brings in more proft and whether it was
sold based on Bobs $60 per hour rate or some discounted number within
reason, it still made us more proftable than letting a machine sit idle.
4.7 convertIng sMAll ProductIvIty
IMProveMents Into ProfIt
One of the biggest problems with making small productivity improvements
in large businesses is that they get lost either within a poor management
system or by a lack of understanding of how to transform these small and
sporadic amounts of additionally created capacity into proftability. Ofen,
TeFirstDaysAlone 59
poor habits like not adjusting a standard process time afer an improvement
has been made can cause the planning system to continue to schedule in the
unnecessary time on the next occasion this part is manufactured. Te great
part of my experience was I knew the necessity to adjust these times and
that I could continue to push more work through the shop, even if I took it
at a lower margin. Even at a lower rate, it was still going to increase my over-
all proftability. Naturally, a productivity improvement has to do with get-
ting more volume through in a shorter time, assuming the additional work
contains the same range of proftability. Here lies one mistake I have ofen
observed and did not plan to make myself: the inaccurate proft margin per
part number. It sounds like an impossibility to understand each individual
part numbers true proft with any level of confdence. And I knew I would
not get it exact, especially with as many allocations as we had in our over-
head costs. But, at least I planned to get my direct labor hours and mate-
rial costs as close as possible and then group some of the other more costly
indirect activities into categories. I fgured I might start with more than
one standard shop rate depending on some characteristics or the activities
required for the particular part being quoted. I was not looking for preci-
sion, just more accuracy than what I had observed we were applying.
Most fnance departments do not appear to have a PDCA cycle to ensure
any accuracy in their allocations and do not appear to work at continu-
ously improving the process. If I could hold someones interest, I think I
could write an entire book about the mistakes I have witnessed in leaders
assuming the fnance departments proft margin per product is reason-
ably accurate without challenging it. To sum up, the biggest problem I
have observed, overhead allocations aside, is the fact that so many compa-
nies do not have feedback systems for their quoting (comparing quoted to
actual hours), indicating the chances of their direct hours being accurate
are minimal. You should not be thinking that your direct labor is only 7
percent or 10 percent of the total cost and dismiss this point because the
direct labor is only the tip of the iceberg (since it is ofen the basis for allo-
cating most of your overhead costs, which are signifcant).
Tink of it this way: you start a weekend project at home, and while
driving to the hardware or garden supply store to pick up supplies, you
might try to guess how things will go and how long it might take to get it
done. At least that way you can plan when you will be fnished. Will the
job take just a few hours, the entire afernoon, or the next few weekends?
If you are like me, things rarely go as planned and no matter what time
you promise your signifcant other you will be fnished, it usually does not
work like a Swiss watch.
Now, imagine that you were to do that same job againwhat would
be diferent? First, you would have learned a few tricks, so likely the job
would take less time the second time around. Also, you would have a much
clearer idea of the exact time and steps involved because you performed
the job yourself, in which case you are directly providing the estimator
(yourself) with the feedback.
Apply this experience to the work environment. Te frst complication is
you usually have one person estimating and another performing the work.
Without a feedback system, how is the estimator to learn about inaccura-
cies in the way you learned about them during your weekend project? Te
second thing to consider is that if the work were to repeat, how could you
be better prepared afer having gone through it the frst time? Even in
the case that you had originally estimated the correct time, imagine how
you would be able to make improvements for your second run that would
make you more competitive. A smart business person would be aware of
these savings in terms of improvements, would document or capture the
learnings for the next time, and not plan for the unnecessary time and
waste. Naturally, if your customer is still willing to pay for this extra time,
go ahead and let them. But, you need to understand the true time (in other
words, the true proftability) so you can make the best business decisions.
Terefore, a system without feedback or a method to continuously update
the time it takes to perform the job as you constantly introduce improve-
ments is comparable to having someone else perform your weekend project
the second time around, and not sharing with them what you learned on
the frst occasion. Likely, they will make the same mistakes and you will
end up paying for this one way or another. Tis analogy can be extended
to more than just direct labor hours; it also applies to indirect tasks and all
processes in which you are engaged.
4.8 chAllengIng the rePorted ProfIt MArgIns
Te advantage I had was I did not plan to make the same mistake that I
had seen others making for years. I have sat beside many managers when
asked by the big boss, What is our proftability on this job? In many of
TeFirstDaysAlone 61
these cases, I had been on the shop foor and observed the actual time,
materials, and efort required, and compared those to the available stan-
dards (routings, bills of materials, etc.). Terefore, I knew the times were
not correct as noted in the routings and there was no system in place to
continuously update this. Consequently, when these managers answered
something like, We have a 17.5 percent proft margin on this job, I knew
that this was inaccurate since it was based on some poor data and lots
of assumptions in allocations of the overhead. But, everyone else seemed
content to take this number at face value. What if, in reality, they only
made 2 percent because they did not account for additional overhead costs
(i.e. extra Q.C. efort) and then further orders for the same part/service
were received? Ignorance can be bliss but if this happens too ofen, real
fnancial jeopardy can follow.
I have already explained my Excel spreadsheet that was being utilized
as our format to ensure standardization for quoting the jobs (shown in
Appendix C). Te sheet was then attached to the shop-foor information
package that traveled with all jobs. Te gray boxes were completed to show
actual times, problems, and later the actual material that was consumed.
My concern was now to capture improvements and encourage us to con-
tinually improve the times. Setup time is the traditional starting point to
be pursued during a lean journey, but I also wanted to look at the cycle
times (the machine operating time) for each operation. I wanted to drive
this continuous improvement and always capture the savings without
encountering a lot of paperwork or extra documentation costs. However,
this would come a bit later as the OSKKK method indicates to frst Kaizen
fow and process because they are the least costly to change and have the
most impact. Te next step is to Kaizen equipment (fxturing, cycle times,
etc.) as this involves incrementally more cost. Te fnal step is to Kaizen the
layout as this usually has higher cost implications and a longer payback.
Lean principle: Runners, repeaters, and strangers are segregations for
various types of parts when implementing lean (also referred to as ABC
analysis or stock and non-stock). Tis is even more applicable in high-mix
plants. Runner partstypically, the 10-20 percent of your part numbers
that account for approximately 80 percent of your sales volumeare the
ones on which you frst apply many of the lean improvements. Tis is
where you will obtain your best payback.
Simply put, you need to continuously monitor and improve the times for
parts that are produced over and over. You need to understand and docu-
ment these improvements, not necessarily passing the savings onto the
customer if the market does not demand it. Also, never forget that proft-
ably selling any additionally created capacity is the best scenario.
4.9 suMMAry
5S is a tool that improves productivity, especially where you cross-
train and have job rotation (which are critical elements).
It is crucial for managers to quickly distinguish between normal and
abnormal conditions. Well-executed visual management can aid in
quickly determining the status.
Team leaders are vital in problem solving and resolving issues at
their root causes.
FIFO is important when managing work-in-process as these jobs
route through the various process steps (push at the beginning and
then fow), and without FIFO the lead times will fuctuate.
Determining more accurate proft margins for each part number
requires a PDCA cycle as the basis to direct continuous improvement.
63
5
Creating and Selling Capacity
5.1 only thInkIng In terMs of full
AbsorPtIon costIng
I might not be the sharpest tool in the shed but I completely under-
stand that the more volume you move through a manufacturing plant, the
more likely you are to be proftable. I cant argue with those of you who
are thinking, Yeah, but it has to be work at a decent margin. But, in the
case where you have already covered fxed costs with baseline volume, it
is basically hard to lose money on additional jobs as long as they absorb
more than just the variable costs. Te problem most companies encounter
is their quoting and accounting systems are stuck in the mind-set of full
absorption costing (includes all manufacturing costs: i.e., direct materi-
als, direct labor, and both variable and fxed manufacturing overhead),
instead of having the aptitude in certain cases to apply only variable costing
(includes only variable manufacturing costs: i.e., direct materials, direct
labor, and only variable manufacturing overhead) plus a bit of proft. In
those special circumstances where you look at only variable costs, it would
lead to arriving at a lower quoted price and might put some work in your
shop that you might not otherwise have been awarded.
Luckily, I was not burdened with such a difcult system. I was able, not
only to judge each job on its own merit with real numbers, but to look at
it also in terms of incremental costs compared to full absorption costing
and make an educated decision: Did I want to make money (assuming the
customer was unwilling to absorb a fully burdened rate containing excess
allocations) or be burdened by a rigid pricing system that did not take into
account the realities of my available capacity versus demand? Naturally, it
was always a gamble that this particular part (quoted at a lower rate) could
come back later to haunt me if it were to receive repeat orders (in particular
for larger quantities), especially at a time when my demand was exceed-
ing capacity and I could instead be selling that capacity more proftably. I
was willing to take some calculated risks because lean methodologies ofer
many possibilities to increase capacity. I also knew that as I reached the
Kaizen stages during my OSKKK (Observe, Standardize, Kaizen Flow &
Process, Kaizen Equipment, and Kaizen Layout) strategy implementation,
I would need some money to fund these improvements, especially in the
areas of Kaizening equipment and layout; therefore, any extra proft could
be quickly reinvested in the business.
5.2 overtIMe As A PosItIve
I fnd it amusing that some companies view overtime as a negative. I
agree that it is a major concern if you do not manage your daily business,
and overtime is driven by poor productivity, poor uptime, or employees
manipulating the system for a bigger paycheck. In my case, almost all
of our overtime was the result of additional demand that exceeded our
capacity, this was ensured by controlling nearly every minute of straight
time with planning boards and time management. Tis made almost all
my overtime very proftable, even when paying one and a half times the
normal hourly rate. I could still put money in my pocket since about two-
thirds of the standard shop rate went to cover my overhead, which was
basically already fully absorbed in our standard forty-hour work week.
Te weeks that we had a full shif running on Saturdays even encouraged a
tight guy like me to buy a round of cofee at Starbucks on the way to work.
My formula was simplemy rent remained the same, my loan payments
remained the same, my insurance remained the same, and those extra
hours billable near $60 per hour or more earned a lot of proft. I was not
about to overlook that by putting some incorrect metrics in place, always
applying full absorption costing, or thinking of overtime as a negative.
Even the jobs I sold at less than $60 per hour to secure the order were still
proftable when estimated and managed correctly.
CreatingandSellingCapacity 65
Before going on, I want to capture this point about overtime, as I feel it
is ofen misunderstood.
Key point: Overtime is a negative if you do not manage your daily
business, and this additional cost is necessitated due to poor produc-
tivity, poor uptime, or employees manipulating the system for a bigger
paycheck. On the other hand, overtime is a positive and is proftable
as a short-term solution to meeting temporary increases in demand if
you proftably manage your straight time.
5.3 setuP-tIMe reductIons
I do not want to give the impression that I willingly paid overtime. On
the contrary, I was continuously looking for productivity improvements to
create that extra capacity, achieving more output without overtime. Being
a lean guy, I naturally focused on setup time as one of the frst steps
within the OSKKK strategysetup falls under the second K, Kaizening
of Equipmenttherefore, any extra proft could be reinvested when fx-
tures and tooling were required thereby further enhancing proftability.
Our frst successes were low-hanging fruit. For example, originally we
did not have complete sets of tools and fxturing at each machine, caus-
ing the setup men to walk away to fnd and periodically wait to share a
common tool. Tese were mostly inexpensive hand tools so the frst steps
were some small investments. Some items like cutting tool holders for
the various machines were a bit more expensive, and we had to gather a
little data frst to see how ofen a particular operator had a delay caused
by not having sufcient tool holders or a particular type of tool holder.
Afer identifying a concern, a simple tick sheetrequiring minimal time
to completewas put up on each machine for a few weeks. Using the tick
sheets, we quickly quantifed how ofen either a few minutes were lost due
to searching for a tool holder or how ofen signifcant time was needed to
confgure an alternative tool holder. We further noted how ofen it was
necessary to revise the program to use another cutting tool, and we also
tracked how ofen we used various fxtures stored close to the machines.
Te results showed that we needed some sorting to be done (5S, workplace
organization) so that infrequently used tools could be stored further from
the machines or even disposed of and vice-versa the runner tools were
located close to their point of use. Once the data regarding delays was in
front of me, I applied an average time lost for each occurrence, and from
there it was easy to justify some investment and determine the particular
type of tooling to purchase to minimize forced waiting times. Real-time
Pareto charts were the method we employed to quantify the amount of
time lost for every occurrence when a tool was not available (see the exam-
ple shown in Appendix E). Te advantage of the operators putting the data
directly into this real-time Pareto chart was that I did not have to spend
time transferring the data into a diferent format to interpret its meaning
and draw conclusions. I could do that in real time as the data was entered
by the operators.
Lean principle: Kaizening of Equipmentthe second K within
OSKKKconsists of:
Reducing setup time using SMED (single minute exchange of dies)
methodology.
Improving machine feeds and speeds.
Measuring and improving OEE (overall equipment efectiveness).
Balancing operator workload to machine cycle times.
Making improvements in TPM (total predictive maintenance).
Simplifying machines to improve OEE.
Te importance of basing decisions on data acquired through obser-
vation is another key point because, without data, I ofen fnd people
talking in circles or making poor decisions or not able to draw frm
conclusions.
Key point: When a decision requires some investment (either time or
money) and not all parties clearly buy-in, it is easy to have some simple
data gathered over a few weeks or months by the people performing the
processes, then regroup and review the facts. Usually, the dataespe-
cially in the format of a real-time Pareto chartwill speak for itself and
make the choice clear, one way or another.
Many other setup reduction ideas presented themselves, some to do with
improving our work order package by including better tooling lists, others
to do with standardizing locations for certain expensive, high-usage tools
that were shared among the operators (further simplifed as we purchased
more of the same types of CNC [Computer Numerical Control] machines
that permitted us to share more common tooling). We standardized some
fxturing that would remain set up on certain machines and some of the
bigger machines were standardized with multiple types of fxtures lef per-
manently in place spread across the larger machining tables. One advantage
in any machine shop is that once you get everyone focused on improving
the setup, you engage everyones mind and people get creative. Machine
shops have a further advantage in that they can manufacture most fxtur-
ing in-house; not only can someone come up with an idea for fxturing, but
he can also usually produce it if it is not available in a catalogue.
5.4 contInuously MeAsurIng setuP tIMes
Most changeover time-reduction eforts founder because of a lack of
support or because there is no measurement or incentive put in place.
Fortunately, this was not the case for me.
As previously mentioned, the estimate sheets for all new jobs (or the
previously achieved times for repeating jobs) were now being attached to
each work package and the operators were required to fll in their actual
setup and run times for each operation. Terefore, the measurement sys-
tem already existed, and we were not going to have the problems that I
have encountered at other companies: they run setup reductions or SMED
workshops but do not continue to focus and follow-up on them, or they
do not constantly measure the results. It was far too important an area for
me and we already had our monitoring system in place. Te monitoring
had been functioning for about two months by the time we focused on
the setup, and now this data would help to confrm the accuracy of our
estimates, highlight problems with setup times, and allow us to capture
improvements by reducing the allocated time.
Te incentive for my employees might not have been directly fnancial; but,
since I had explained that the longevity of the business would be jeopardized
if we were not achieving or beating the estimated times, their motivation
came through job security. Also, I announced that I planned to develop a
bonus system that would take productivity into account. Productivity was
important to meand what is important to the boss should be important
to the employeesand that meant I was going to be constantly monitor-
ing and trying to improve it. Tis is probably one of the biggest reasons
lean transformations are not successful: the managers do not signifcantly
change their behavior (or their standardized working methods).
Key point: What is important to the boss should be important to the
employees, and continuous monitoring, reviewing, and supporting
improvements can only lead to success. In lean terms, you must always
Plan, Do, Check, Act (PDCA) in a continuous cycle.
As far as incentives for the entire team, I was working to create a bonus
system driven by these productivity improvements. At this time I did not
have enough background in the fnancials and had not put the proper
metrics in place. But I promised that once they were in place, some form
of bonus would be paid. I had yet to fgure out how, but I was confdent it
could be done. And afer seeing some poor bonus systems in other com-
panies that frequently promoted behaviors contradictory to what was
intended, I felt I would not make the same mistakes. I later learned that
this was easier said than done and that designing a reasonable bonus sys-
tem was beyond my abilities.
Lean principle: A formal suggestion system encourages employees to
submit ideas to address waste, safety concerns, and environmental
improvements. Naturally, it supports a structured way to channel, eval-
uate, and track ideas. It also encourages continuous improvement by
providing the motivation and form to evaluate and discuss these ideas,
instead of encouraging only a few at higher levels in the organization
to provide the majority of suggestions for improvement. Tese systems
can become unruly if not thought out and supported with resources and
management involvement. I was discussing a bonus system with the
employees, but a suggestion system is also important.
Returning to setup-time reduction, I was really looking to separate
the internal and external setup elements, using the SMED methodology
pioneered by Shigeo Shingo. Keeping in mind that we typically changed
over a machine between three and six times a day to one of a few thousand
part numbers, few if any of which were familiar to us, we had our work cut
out for ourselves. I had a few further concepts to implement in conjunc-
tion with this theme, the biggest being that I really wanted my foreman to
function as the team leader.
Te great advantage of having Tom (the shop foreman) as team leader
was his work ethic. He was a working machine, and those who could
not keep up fell into his bad graces and I learned later that they would
not survive working with him. I wanted Tom to be freed from running a
machine and instead to be dedicated full time to solving problems, pre-
paring setups, and minimizing any machine stoppages. Bob (the former
owner) frequently had Tom running a machine, feeling that the highest
productivity (or proftability) was achieved when Tom performed this
direct labor function in addition to his other responsibilities.
I was not hung up over the traditional separation of direct labor and
indirect labor, and did not view the latter as having a negative cost impact.
I had been trained by Toyota, where there was no such mentality. Most
businesses are willing to accept lower uptimes or OEEs (overall equipment
efectiveness, defned in more detail in Appendix F) and let everyone solve
his own problems; however, they do not demand a higher OEE by provid-
ing the resources to solve the problems and keep the machines running at
or above 80 percent. I knew the superior solution was to have one skilled
problem solver running around instead of relying on the various problem
resolution skills of all my setup men and operators to deal with the various
abnormalities.
So, before I liberated Tom from the day-to-day running of a CNC, I
gathered some data in order to compare the before and afer, and deter-
mine whether we achieved a productivity improvement. For almost two
weeks I had everyone attempt to record every time his machine was
down and the duration of the delay. Again, the simplifed, real-time
Pareto charts (shown in Appendix E) were utilized. I fgured I was about
to have one less part-time machine operator, and I needed to justify this
cost by having the other fve or six machines running more produc-
tively. Tis team leader concept was one of Toyotas proven successes,
so I fgured I could get it working in my little world. When a machine
is involved, OEE is a critical measurement for improving productivity.
We began with the operators capturing the downtime data and placing
it into categories while I was able to calculate the OEE based on the good
parts we produced.
Lean principle: OEE is the ideal measurement for monitoring and indi-
cating where to improve manufacturing processes (ideally, auto cycle
machines or cells containing auto cycle machines). More details can be
found in Appendix F.
5.5 creAtIng A teAM leAder
Although most would think anyone would be glad to be more or less free
from running a machine, I had discovered at Toyota that many people who
were promoted to team leader would later ask to go back to working as an
operator on the assembly line or running a machine. All that running
around and dealing with everyones problems can sometimes be too much
pressure and some feel it is mentally and physically easier to just operate a
machine or work on the line and call for help when there is a problem. (A
list of traditional team leader responsibilities within the Toyota system is
captured in Appendix G.)
Tom had been the shop foreman for many years and was used to com-
ing in early to start things up and get prepared. But, most of the day he
struggled to keep a machine running while simultaneously performing
his other dutiesrealistically, he was only leaving his machine to frefght.
Instead, I now wanted him to continuously try to reduce any reason the
machines might not be cutting chips. So, it was time to take Tom out for
a beer or two and present the idea to him, to share the reason for suggest-
ing a change, and to see how he felt about it.
Tom seemed to like the idea and the challenge; he also felt he could learn
to be a better problem solver. I started to explain a few theories on how
problem solving could be applied but thought the conversation might go
better at the shop rather than at the bar. We agreed not to implement this
new role immediately but over the next few weeks we would clarify the
responsibilities and discuss how things would work diferently. I also indi-
cated that there would be no hard and fast rule that Tom would not go
back to running a machine if the business case required it or if he needed
to fll in for an absent operator.
Key point: When choosing a team leader, not only do you need someone
like Tom, who had earned everyones respect and knew all the jobs in the
area, you also want a leader who is trained in problem solving and sets
an example in working methods/ethics and dedication. For me, it was an
easy choice because Tom was one of the most dedicated workers I have
ever met. He knew the jobs and had earned the trust and respect, all he
needed was some training in problem solving.
While we were planning for Tom to move into the team leader role, we
started discussing all the setup tasks that would be externalized (prepared
while the machine was still operating) and how we would best leverage
the other setup men (who functioned as operators when not performing a
setup). Te simple fact of externalizing setup activities implies that either
the operator must have the time (outside of keeping the machine running)
to perform these tasks or a team leader (or similar individual) with the
available time must handle them. For example, you cannot just say that we
will have all the tools prepared in tool holders prior to the beginning of the
setup without a plan for who will be doing this and making time available.
(While the machine was running in our shop, the operators were ofen
busy running a second machine, deburring, measuring, etc.; therefore,
they did not have a lot of available time for preparing tools.) Keep in mind,
though, as with most companies I was selling my machine time at a much
higher rate than the cost of the employees hourly wages, so the more time
the machines could be producing, or in this case cutting chips, the more
chance I had for proftability.
5.6 IncreAsed ProductIvIty
through vIsuAl MAnAgeMent
Te scheduling boards I had put in place about a month before were also
beginning to help in improving the available capacity. I found just by set-
ting a targeted completion time, then monitoring against it and being pro-
active in reducing problems was motivating the team. (Naturally, there is
a danger if you set the target too low, but PDCA helps to verify the times
and minimize this).
It took a lot of work on my part in the beginning to monitor these boards
and try to get everyone to see it as a positive. But, I think it was successful
because I always tried to focus on what went wrong within the process
instead of making it personal. Everyone could then look to improve the
standard processes instead of feeling they were being personally attacked.
In all honesty, there were times when someone had just screwed up and
ofen I sufered the fnancial consequences of the lost time or scrapped
material. I tried not to lose my cool very ofen; instead we would discuss
what should have been done by the operator and I would document the
serious instances. Te upside was that if I decided an employee had too
many of these problems, I could let him go without being accused of unfair
treatment or being challenged for a poor labor practice. Ofen, supervisors
do not take the time to document infractions related to quality or produc-
tivity but it is a necessary part of the job.
Instead of setting explicit expectations with relation to time, Bob had
always just given out the next job. It was matter of prioritization, and he
was just spoon feeding Tom. Nevertheless, I noted during my training
period that Bob periodically told me a job seemed to be taking too long
(based on his experience). Ten hed either ask Tom if he knew what the
problem was (though frequently Tom was unsure of the reason because he
had returned to running a machine) or he would ask the operator himself.
Te problem with this random system was Bob had not set the expectation
with the operator, and the machinist obviously was not trying to achieve
anything other than staying busy. He did not know if he was ahead or
behind and only looked at the clock to determine how far of break time or
quitting time was. When I began setting time-related expectations, there
were instances when an operator did not agree with the scheduled times;
but it is always better to discuss this and set the expectations up front.
About this time, I had owned the business nearly six months and had
closed the books for four months since my training period had ended.
Overall, things were still looking good. I had continued to earn enough
to cover the loan payments and was even able to pay myself a decent sal-
ary. I felt I was getting a little closer to fnancial stability though I was not
in any shape to weather a storm, and it would only take a few months
of poor sales or some major quality issues to throw me into a downward
spiral, losing money instead of drawing a paycheck. Te unwritten rule
among business owners is that you need to survive for fve years (appar-
ently the magic number) and to have your debt level signifcantly reduced
before feeling there is any breathing room or longevity for the business.
However, I was not getting any younger and I knew I could not wait fve
years to pop the question. And if I had to bet, I was sure my girlfriend
would not want to be strung along for that length of time.
5.7 PdcA of the estIMAtIng Process
I was continuing my weekly reviews with John (the programmer/estima-
tor) regarding the accuracy of his quotes. I realized that any actual times
that were within 10 percent of the estimate were not worth dragging
into our Friday reviews, and there were weeks I lef out some that were
15 percent above or below the time estimates. Te various hourly rates
we had begun using (accounting mostly for the type of material we were
machining) to better account for our tooling and direct shop costs were
also beginning to better align actual costs with estimates.
I wanted us to improve as much as was reasonable and I was also trying
to learn and improve the processes as quickly as the organization could
absorb. So, the more quotes we reviewed each Friday, the more I was able
to learn, and periodically I was able to further evolve or refne our quoting
process. I had even started doing some of the simpler quotes myself and
then John would look them over and give me pointers. I had good days and
bad days but I learned something from each of Johns reviews. I was even
more interested if we got the job to see how the actual times compared to
my estimates. It basically becomes a point of pride when you get it right.
On some occasions I was even getting some acknowledgments from Tom,
who wasnt exactly free with a compliment. I was also tracking our hit rate
on quotes from week to week and month to month. In the early days, you
may recall, I had set a goal of winning orders on 70 percent of our quotes.
With our type of business and the short lead times, we either heard back
quickly by receiving a purchase order or it was not meant to be. Our hit
rate was averaging anywhere from 50 percent to 65 percent and was utiliz-
ing a reasonable level of the shops capacity on one shif.
I was still learning diferent buyers habits, how to read into the serious-
ness of the quote, and how much price fexibility there might be. I was also
trying to compare this to the proft margin per job and look at an average
margin per customer. Bob had given me a gut feeling for which customers
were more proftable, and my simple data analysis had basically proven
him correct. It was also interesting to track the hit rate by customer and
for each particular buyer. Tere were wide swings, and in some cases we
were getting over 80 percent of quotes for a particular buyer turned into
orders; others were as low as 15 percent. Tese numbers indicated which
buyers shopped around or just used us for a second or third price quote
and which buyers were too busy or too lazy to get alternative quotes and
just regularly sent us the order. I guess the other possibility from those
buyers from which we had a high hit rate was that the alternative quotes
were always more expensive or our on-time delivery and quality level
ensured us a lions share of the work. In these cases, I needed to ensure
we had a good proft margin from these jobs and we werent just getting
orders where we made mistakes on the quotes and underbid. Remember:
winning a job was only a good thing when it had been correctly quoted
and yielded a good margin.
Just when all seemed to be going well, John did not show up for work
one dayno show, no call. Te second day, I tried to reach him as I could
not let the difcult quotes wait any longer. His wife would only tell me
he was not at home, and so I assumed I had lost my estimator. Although
John had not been late or absent in the previous three months, I would
have to terminate him if he failed to show the next day on the basis of the
strict attendance policy I had implemented: mandatory termination for a
combination of three late arrivals or unplanned absences within a rolling
ninety-day period. I had learned this attendance rule when working for
Toyota and it was part of a larger program that kept absenteeism down
near 2.7 percent during my tenure.
5.8 becoMIng the estIMAtor
I had feared something like this happening since I took over the business
and had been planning for it by learning and documenting all the critical
tasks. However, the quoting was one of my greatest fears as it was a high-
level task that required an in-depth understanding from the perspective
of both a machinist and a programmer. I had no other choice, so I started
working through a difcult batch of drawings and working up the quotes.
Te frst day John was absent I had taken the frst step of sorting though
the drawing packages and deciding which parts we could not build because
they were not within our capabilities or they were sheet metal or fabricated
parts. When Bob owned the business, he would try to subcontract some
of these parts to other machining or sheet metal companies and then add
a markup (though he did not let the customer know that the parts were
manufactured elsewhere). I noted that most buyers were not concerned
or involved enough to understand where the manufacturing took place.
Tey just wanted a good price and good quality for all the drawings in the
package and it was easiest for them if the entire package could be placed
on one purchase order with one vendor. I had further capitalized on this
outsourcing and generally tried to quote all drawings within the package,
which required digging around and fnding additional subcontractors to
those that Bob had used. But, I hoped that the average 15 percent markup
that some of those orders brought in was worth the efort. I later learned
that sometimes the difcult parts were more trouble than they were worth
and outsourcing work with some new suppliers could periodically damage
my reputation in terms of quality and on-time delivery. Bear in mind that
I was not outsourcing to reduce my costs. I was only going outside when I
did not have competitive capabilities internally.
Of the quotations I was working on that morning, I had already sent out
for subcontractor quotations all the drawings I had determined we could
not manufacture in-house. I had resolved that I needed to do the in-house
estimating, at least in the short term, while I worked to replace John. So, I
began to conceptualize how to manufacture each part; in what form I would
request the raw material (i.e., bar stock, cut to size, etc.); what the frst, sec-
ond, and third operations would be; and I began flling out my standardized
quotation sheet. Tis sheet was a real help my frst time alone at this pro-
cess. It helped keep my thinking structured and worked like a check sheet
to ensure I did not overlook anything. I was really glad that I had not only
learned the process but had also taken the time to document and structure
it. All those Friday afernoons spent reviewing the quotations were really
paying of, especially when we found something had been overlooked and
we went back and improved our standard quotation form. I would not say
I liked the position I had been thrown into but I was much more comfort-
able and better able to handle it because of having a standardized process in
place and having pried all the tribal knowledge I could from both Bob and
John. I really had a good standard process instead of just following Bobs
words: You learn it as you go and then youll know what to do.
I worked through the two drawing packages that were waiting for pric-
ing and put my fnal numbers together. A little follow-up and I had all the
numbers in from the subcontractors. Ten, I had to decide what margins
could be used with these customers. Tis all happened within the day, and
before 5:00 p.m. I was ready to fax back the quotations. Again, the proof of
how well I had done was not whether I won the bid for the job because that
might just mean I missed some important items and underbid it. Te bot-
tom line was in comparing the real costs to my quoted costs. I was typically
quoting individual parts that ranged from $20 to $2,000 eachjust think
how critical estimating is for companies quoting large, expensive systems.
For one customer, I had quoted eleven diferent parts and received the
orders for six at the smaller quantity levels quoted (periodically, customers
asked to have parts quoted at various quantities, which usually meant they
were looking at inventory management or they did not yet have an order
themselves). Only getting part of the order sometimes indicated that I had
overlooked something that my competitor had foreseen and included in
his price. But, afer a quick review of the six parts for which we received
the order, I did not see anything obvious I had overlooked. From the other
quotation, we received the order for all nine parts we quoted. I had expected
this as we usually received what we quoted from that particular buyer. She
was my favorite type of purchasernot familiar with the commodities she
was buying, always overloaded, and always in a hurry. Normally, I did not
have to sharpen my pencil too much with her. She frequently laid her
cards on the table, indicating how desperate she was for the parts by the
way in which she sent the request for quotation: a hand-scribbled note sent
via the fax machine with some hastily conceived drawings, a due date of
ASAP, and asking how soon she could have the quote back. It was a clear
indication that if I could quickly get the quote back with a short delivery
time, price would not be the determining factor. I think this is another key
business point to capture.
Key point: It is critical to have measurements in place that indicate when
the purchasing or estimating departments are overloaded or only given a
short period in which to place an order. Under these circumstances, you
will not have sufcient time to obtain the best prices and quality, thereby
not providing competitive quotations to your customers. Tis normally
results in signifcantly increasing costs or using poor quality suppliers. If
your real-time purchasing workload indicates an overload, you should
have other people trained to temporarily support the purchasing and esti-
mating departments. Another tactic to pursue is to reduce departmental
workloads. For example, in purchasing you can put some frequently con-
sumed parts onto Kanban reordering, utilizing blanket purchase orders
that will reduce the daily workload, and creating more available time to
obtain the best prices on other purchased components.
Afer we built the jobs for which I had prepared the quotations and I
reviewed the actual versus estimated costs, I found I had done well and
decided I would still look for a new estimator, but I did not need to rush.
Again, my goal was to be freed up from day-to-day operations as much as
possible, so I wanted an estimator. But, I did not want anyone to be able
to hold me for ransom because he or she was the only one that had that
specifc skill. I was on my way to proving that no one could pressure me
into thinking I could not live without an estimator.
5.9 suMMAry
Full absorption costing is not the only option; to better utilize your
capacity in certain instances, variable costs plus adding some proft
could be utilized when quoting.
Reducing setup times is a critical and early step in lean improve-
ments, and when you are able to proftably sell the additional capac-
ity you have created, you have a winwin situation.
You must continuously monitor setup times to truly beneft and con-
tinuously improve.
Team leaders can be some of the most productive employees if they
are assigned the correct responsibilities and receive proper support
from management.
Putting a PDCA loop in the estimating process is a vital step for any
made-to-order or job-shop type of business.
Monitoring the purchasing departments workload to recognize when
they are overloaded is critical to maintaining cost control. Tis is
especially important when you have a high percentage of your spend
on either raw material, purchased components, or outsourcing.
79
6
Getting Everyone Motivated for
Improving the Machines Output
6.1 IntroductIon
I had been drawing a regular paycheck, enough to live on but nothing to
brag about. So, I was periodically giving myself a small bonus at the end
of the month, assuming it was in line with our monthly proft. Te good
news was the monthly bonus was becoming steadierwhich was great
because when I had popped the question during a surprise visit to my
girlfriend in Spain, the answer was a big yes. Te surprise visit to Spain
was two fold: I partially wanted to astonish her, but it was also important
that I do it on her turf and in her mother tongue. On the fight over I
was busy digging through an English/Spanish dictionary trying to fgure
out the necessary words and put them into proper sentences. Te Spanish
woman sitting beside me asked if she could be of some help. I was a little
embarrassed, but once I explained to her what I was attempting to do, she
appeared to be so taken with the idea, she couldnt do enough for me. I
found it very strange practicing and repeating the phrase will you marry
me with a complete stranger. Tough prior to landing she told me that
my pronunciation was fne and apparently it worked. Can you imagine
what the others on the fight were thinking as I repeated over and over
again will you marry me? Now, attaining a bit of fnancial stability and
slowly developing a cross-trained workforce, the idea of marriage was feel-
ing a little more comfortable from my working perspective. Maybe even a
little time of for a honeymoon would be possible.
6.2 MAnAgIng eAch Job vIsuAlly
I continued looking for a new programmer/estimator while sharpening
my own skills in these areas. With only about six months experience
under my belt, I was outsourcing the more complicated parts; however,
with each passing day I was able to estimate and program progressively
more difcult jobs. When I reviewed my estimates against the actual
times, I found there were good and bad days. In reality, I did not even
need to wait until Fridays to know how I had performed. I was visually
scheduling jobs according to the estimated time. Tom (the shop foreman),
the operators, and I were immediately aware when certain jobs began
taking more time than planned, and we started observing/monitoring
and asking questions. Of the jobs requiring extra time, it seemed about a
50/50 split between those that had encountered operational problems that
caused the overrun and those that I had incorrectly estimated. It wasnt
as bad as it sounds because, although the errors frequently were under-
estimates, periodically I overestimated the time, and it put a smile on my
face to see a very proftable job moving through (although I still wanted
to get the quoting right). Also most of the incorrect estimates were only
of by 10 to 20 percent. As they say, You win some, you lose some. I just
hoped those very proftable jobs would be the ones to repeat.
Lean principle: Tis visual scheduling of jobs is really how a made-
to-order business accounts for its takt time (the customer demand).
Remember: takt time helps set the pace/beat based on customer demand,
and this must be compared to the cycle time, which is how long the pro-
cess itself takes. So, the time estimated for a job is basically what the
customer has agreed to pay for; therefore, this is as close to takt time as
you have available. Te visual board was measuring this time against
how long it actually took (the cycle time). Since this helps to improve the
machines output, it is a further form of Kaizening of equipment, the
second K in OSKKK (Observe, Standardize, Kaizen Flow & Process,
Kaizen Equipment, and Kaizen Layout).
GettingEveryoneMotivatedforImprovingtheMachinesOutput 81
6.3 the desIre for IndePendence
froM dAy-to-dAy oPerAtIons
No viable candidates had responded to the newspaper ad I had run the
week afer losing John, the programmer/estimator who had quit. I had
been holding of running another ad for a few weeks when I received a
call from my former employer, Delphi Automotive, asking if I could come
back and support them. Teir frst priority was to evaluate a particular
plants current lean status before a visit from the CEO of General Motors.
Following this engagement, they needed support for some other plants on
a lean consulting basis.
As challenging and exciting as it was running my new business, it was
also intriguing to think about keeping my hand in my former profession.
I had not even mouthed the word consulting until receiving this call. I
had wanted to get the business to the point where I could step away from
the day-to-day operations for a holiday, a practice I had become accus-
tomed to while living in Europe. Tough the thought of part-time consult-
ing was well worth considering, the business was not at a point that I felt
comfortable enough to pursue this idea. I explained this to my contact at
Delphi, who felt the ofer would still be open in the future; in the mean-
time, could I spare a day or two to help them get ready for this extremely
important visit, especially since I had overseen the lean transformation
at this plant? I made this trip, but in reality I was busy working ten-to-
twelve-hour days refning and putting new processes in place. I was also
functioning as the estimator/programmer and handling other functions
that I had taken over from the former owner, Bob, and still had not had
the time to cross-train others to perform. Terefore, I felt six months was
about the best I could do before I would feel confdent to leave the busi-
ness for a week at a time. So, with the possibility of consulting as a sideline,
fnding someone to handle the estimating and programming responsibili-
ties once again moved up on my priority list.
WeTom and Istarted interviewing again. One of the most impor-
tant reasons I wanted Tom involved in the interview process was that he
had more experience in the industry and he had some street smarts that
I was lacking. Another key reason was that he was going to be working
more closely with the estimator than I would be. Actually, I wanted the
estimator to report to Tom (even though the position would likely pay
slightly more than Toms) because I trusted Tom to always get the job
done. I also wanted the interviewee to be clear that the estimator position
would report to Tomwho never smiled and was very direct. I felt it was
only fair for the interviewee to consider who he would be reporting to, and
for Tom to know ahead of time who hed be working with.
Afer more than a month of searching, we found Jake. He brought over
twenty years of experience and had worked with various materials and
machines. Again, I wanted to pick his brain to learn what we did diferent
than other shops and where he thought we could improve. I wanted to hear
his opinion on everything, but I preferred he tell me his ideas prior to sharing
them with everyone else in the shop. I had to keep in mind that the others had
been working in the industry signifcantly longer than my seven months of
experience, and if I were to let the new guy infuence too many changes with-
out letting everyone else voice an opinion, I was sure to make a few waves.
6.4 PrePArIng for further setuP reductIons
I was now ready to further intensify our eforts at reducing our setup
times. It seemed to me that now the average quantity of pieces per order
we were receiving had decreased from what they had been when Bob had
owned the business. As I looked back through the history of past pur-
chase orders, it seemed to be moving slowly in the direction of decreasing
lot sizes per order. During recent weeks, Tom had been moving into the
modifed role of shop supervisor/team leader, which would enhance our
setup methodology. Te big change was that I really only wanted him to
operate a machine afer all the other machines were up and running; the
tools, fxtures, and programs were lined up for the next work order on each
machine; and our most signifcant recurring problems had been resolved
or at least received attention. We had been working through guidelines
and issues with his new role just as Jake came on board. Te advantage was
that Jake was faster at programming than John had been so he ofen had
a few spare minutes. It became Toms responsibility to understand when
Jake had time available and keep him busy preparing tools and fxtures.
Tis created the opportunity for Tom to get ahead of the game and start
to either run a CNC (Computer Numerical Control) machine or start a
manual job on our more traditional lathes.
Lean principle: Runners, repeaters, and strangers are terms used to
classify various parts (sometimes referred to as ABC analysis or stock
vs. non-stock). Runners are usually the top 10 to 20 percent of the part
numbers that typically account for 80 percent of total sales. Strangers
represent a high percentage of part numbers but account for little sales
volume (order frequency ranging from unknown to monthly or once or
twice a year). Repeaters fall in betweensome likelihood to repeat more
regularly. It is critical to group parts this way so you can manage them
diferently. Runners might be applicable to controlling with some type
of pull system (Kanban), meaning you will likely keep a small inventory
(usually as WIP [work in process] before the process step where signif-
cant variation is introduced, not in fnished goods) to balance the sup-
plying lead time with the customers required lead time; on the other
hand, strangers would always be built to order. Setup improvements
should always frst be focused on the runner parts.
6.5 contInuIng the kAIzen of equIPMent
by IMProvIng the AdJustMent
PhAse of chAngeovers
Since we had already purchased additional setup tooling and tool hold-
ers, things were beginning to move along faster. Jake was able to show us
how to premachine some jaws and fxtures, allowing faster changeovers. I
was trying to create a competitive environment encouraging us to always
improve on the previous changeover times, which had been noted on the
estimating sheet that followed each job through the shop. Te danger in
making things too competitive, however, was the possibility of quality
being sacrifced, along with it being difcult to keep that motivation up
in the long term. Tough where I saw our biggest opportunity was in the
adjustment phase of our changeovers. I had a short brainstorming session
with Tom, Jake, and the two setup men to determine how we could quickly
return to the previous point where we had been manufacturing a quality
part. We concentrated this adjustment phase improvement to our runner
parts. It became clear that we needed a way to capture even more notes
and details of where we lef of during the last production run regarding
ofsets, adjustments, and specifc tooling. Since this documentation added
a cost to the business, I only wanted to perform this on parts for which we
expected repeat orders. I indicated that this was something on which we
could make an educated guess during the estimating phase and should
indicate which jobs were likely repeaters by adding and marking a box on
the estimating form. Tis determination would be based on the customer,
the type of part, the current revision, discussions with the purchaser, and
some other general hints like certain customers part numbering systems
that were designed so the number indicated whether they were designated
as production parts or only tooling/fxturing, etc.
For these new parts that were expected to repeat, a standard setup
sheet would be attached to guide the setup man through documenting
the important items, thereby saving some time during repetitive setups.
Tis setup sheet was to be designed with input from everyone in the team
and was to be continuously updated when we had improvement ideas. Te
concept was to utilize the tool sheet as the basis with certain columns to
be added afer each tool to cover ofsets, specifc tool references, and com-
ments. It would also contain a sketch area where the part orientation could
be shown along with other details and notes written in it (or digital photos
attached). Jake had only seen a similar sheet used at one shop where he
worked, but he was able to add a few more improvement points.
6.6 runnIng A MAchIne WhIle confIrMIng
the quAlIty of the fIrst PIece
Since my previous experience had demonstrated that over 50 percent of
a changeovers time was consumed in the measurement and adjustment
phase, I knew we were on the right track. I also explained to the team that
for repeating parts on the setup sheet we would highlight the dimensions
that would likely be afected by a mistake during the setup, and I would
indicate whether material was inexpensive and/or additional pieces could
be cut in-house. If these conditions were satisfed, then the setup man
could quickly check the highlighted dimensions (those likely to be incor-
rect due to a problem caused during the setup) and, if everything checked
OK, he could start running the next part while he fnished qualifying the
other dimensions. Tis would buy us anywhere from a few minutes to
about one hour additional cutting timeassuming all other dimensions
were correct when completing the check on the frst part. And it would
only cost us some material the few times we might have missed some-
thing. Remember: on these repeat parts we already knew our CNC pro-
gram was good so we really only needed to check for correct tooling and
ofsets. Naturally, our internal quality system and our agreement with our
customers required a full dimensional check for each batch or order but
that did not have to be completed on the frst part. It might sound a bit
like gambling (a Las Vegas approach), but every minute of machine time
was worth at least $1 and ofen it could take ten minutes or considerably
more to completely verify all dimensions on a part. We still completed the
necessary quality checks; we just did this while the machine was produc-
ing the second and third parts of that batchpresumably manufacturing
good partsand we were going to contain and scrap these frst pieces if we
found a deviation from the allowable tolerance. Overall, I found we won
much more ofen than we lost, meaning the frst part passed so the next
few were also good.
Lean principle: Fify percent or more of changeover time is typically
utilized for verifcation/measuring and adjusting (quality control). Tis
provides a signifcant opportunity to improve the time by being able to
quickly return the process to the exact point (parameters) where you lef
of the last time you produced the exact part to the correct quality stan-
dards. Tis requires observing the process and documenting where all
variables are set for the particular machine/process so you can quickly
return. Tis is frequently done by the operator keeping certain notations
(in a standard and organized format); these notes are usually reissued
in the work order package. Physically scaling and marking machines is
also a common technique.
I found with this setup sheet continually being improved upon that we
were reducing our setup times during the frst few production orders for a
particular part. I think this a learning point worth highlighting.
Key point: Afer focusing on improving the physical changeover, you
should work to reduce the measuring and adjustment phase, which usu-
ally accounts for more than 50 percent of the total setup time. Tis should
be done by working to document (or marking the reference points on the
machine) and quickly return to the same conditions that allowed you to
produce a quality part during the previous production run. Taking the
time to complete this documentation in an organized format can have
huge payofs.
I thought I could further link these improvements in setup times to an
employee bonus system. When I refer to a bonus system, I am really refer-
ring to some type of proft sharing instead of an incentive system. Incentive
systems usually fy in the face of lean ideologies because they encourage
standards be set that can be exceeded, and then workers are naturally
encouraged to exceed those standards (i.e., a pieces-per-hour target). Tis
does not encourage standardized work or continuous improvement.
6.7 dIffIcultIes In creAtIng A
ProfIt-shArIng PlAn
I did not have the resources or the wherewithal to design and monitor
a bonus system that linked each employee to their daily performance.
However, by monitoring items like the OEE (overall equipment efective-
ness) of the machinery (for which the reduced setup time would play a
part), I could really motivate the team to continuously work toward these
improvements. Although I told them I was going to work toward linking
these machine improvements to proft sharing, I later found that in a high-
variation business where thousands of diferent parts could be randomly
ordered, there was too large a range of variables that afected OEE. Some
of these were outside of the employees control; therefore, I was not able to
link this level of detail to the bonus system.
Other problems were that some share of profts should be allocated to
the bonus system, and logically this system would be used to distribute
those profts fairly. Tere were a lot of reasons we did not show proft some
months and those were not always related to the teams working eforts
or even our sales fgures. It also related to what constitutes proft: one
simple defnition is when operating income is greater than adjusted
operating expenses. In my case, this was not always a black-and-white
issue. In addition to my large loan payment (an operating expense), there
were investments in new equipment and tooling that increased operating
expenses, and my compensation was likely to be whatever was remaining.
Terefore, I did not have a rigidly defned accounting practice, making it
hard to base any bonus on proft.
Lean principle: I was attempting to design a bonus system or at least
implement a more Western-style suggestion system, which stresses the
economic benefts of a proposal and also provides an economic incen-
tive. A more idealistic approach stresses the boosting of morale through
positive participation from the employees. Tere are usually three stages:
encouragement, education, and improving efciency. I wanted a simple
system for its economic benefts, but I had previously experienced how
labor-intensive they can become if not well planned out.
In the end, afer playing with some diferent mathematical formulas I
gave up on creating an ideal bonus system based on items like quality,
cost, and delivery. It seemed in all cases that a critical element of my vari-
ous proposals was the monthly or quarterly profts, which was not work-
able in my case. Although clearly defned by the IRS (Internal Revenue
Service), tough decisions were necessary when so many necessities called
out for money and a lot of proft was being reinvested in the business. Te
fnal bonus systemin place by the end of my frst yearwas based solely
on an individuals attendance. It was based on a fxed percentage of the
employees wage, being paid in full if they were late less or absent less than
a prescribed limit during the prior six months, and for each occurrence
that exceeded this limit a small portion of this amount was deducted.
Another factor ofen brought to the forefront when working to imple-
ment a bonus system is whether or not the base wages are competitive
with local industry standards. It was pointed out to me that the wage scale
established under Bobs tenure was a little low, so I did a bit of research
and established comparable local wage rates for our various skill sets and
provided milestones so the team could move up to those rates. I listed the
criteria for what an operator and setup person should be able to handle
and then pointed out where I felt the individuals needed to improve. Each
employee was to pick one task at a time to work on, then Tom and Jake
would try to support them in practicing and learning that skill. Tis devel-
opment plan was included in our cross-training (or skills) matrix (this is
also shown in Appendix H).
Lean principle: A skills matrix (cross-training matrix) for either a
department or team quickly shows how much cross-training is in place
for diferent tasks (and can also display a future training plan). Tey
are used to help in identifying and assessing shortcomings in skill sets as
well as in aiding managers and employees in planning employee devel-
opment. Naturally, these higher skill sets are openly encouraged with a
skills matrix and are frequently linked to pay rates. See an example of a
visualized skills matrix in Appendix H.
So, motivating the team to work on setup time improvements through
the bonus system never came to be. Instead, I had to provide the motiva-
tion. Going back to what I mentioned in an earlier chapter, if it is impor-
tant to the boss, then it should be important to the employees. Since things
like setup time and schedule attainment were always important to me and
never reached a program of the month status, the motivation remained
there for the duration of my tenure. My continuous monitoring and help-
ing to improve these measurements (PDCA [Plan, Do, Check, Act] in lean
terms) is what really made it important for everyone. Ive explained my
shortcomings with respect to creating an ideal bonus system but if some-
one ever designs the near-perfect manufacturing bonus system, I would
sure like to hear about it.
6.8 stAyIng hAnds-on sInce the
devIl Is In the detAIls
I found my time was split between the ofce and the shop foor, although
with Tom supervising and Jake in the annex ofce programming, estimat-
ing, and helping prepare setups, I could have spent more time in the front
ofce with the administrative or sales tasks. But, I wanted to continue to
learn the business from the ground up and I could not sit in the ofce
making decisions without having a true feel for the details.
We all know the phrase the devil is in the details, and in a techni-
cal business where you build or assemble a completed part from a com-
plex drawing, it holds a lot of truth. Terefore, I spent time on the shop
foor observing as well as setting up, running, and programming jobs.
Especially being necessitated when we were behind or working on a rush
order, I would take over running a machine during the teams breaks or
lunch time. Naturally, this led to me continually coming up with my own
improvement ideas along with all the suggestions that were coming from
Tom and the others. Tis hands-on approach also helped me hone my
skills with front-ofce tasks like reviewing quotes, purchasing tools/mate-
rials, and really confrming where investment was required.
One example that comes to mind from early on: We had two rather
small, old-style band saws for cutting material to size. Each had an add-on
coolant system that functioned miserably on a good day. One morning, I
noticed we were slightly behind on a job (which included operating one of
these saws) and I ofered to run it during the morning break. Tom indi-
cated to me that keeping the saw running to cut the steel bar to size was
part of keeping this particular job going. Te cycle time on the saw was a bit
longer than the machines cycle time. So, it was a bottleneck and Tom was
planning to set up the second saw afer his break. I ran into a problem with
every piece I tried to cut: the saw blade would pop of a guide, the coolant
system would plug up, etc. Naturally, the guys came back from their break
laughing and asking how I was enjoying my time with the saw.
My standard procedure in a case like this would be to coach Tom or
another team member through problem solving andhopefullyelimi-
nate the biggest problems without making a major investment. I estimated
that during my ffeen minutes of operating the combination of the CNC
and saw we had incurred about four minutes of lost run time on the CNC.
Relatively speaking, the saw is an inexpensive piece of equipment com-
pared to the CNC and is not much of a proft center itself. But, it caused
such a distraction of valuable resources that I found myself losing critical
time on the CNC machine (the proft center). Te bottom line afer a short
investigation was that the saws could not be brought up to an acceptable
standard because of their inferior quality and age, and a new saw that
could exceed the capacity of the two old saws required less than a $2,000
investment. Plus, it gave us some additional cutting capabilities (relating
to size and accuracy) so we could keep more work in-house instead of out-
sourcing (keeping this cutting in-house also allowed us to better control
our lead time). A quick calculation indicated payback would be between
three and four months. Te point: had I not been working hands-on and
able to quickly assess the overall cost impact of this poorly performing
saw, it would have been more difcult to convince me and the team would
likely have gone on struggling with it. It highlighted that we still did not
have a good system in place to identify and quantify where we were losing
money within the individual processes.
As a consultant, I have ofen observed operators struggling with insig-
nifcant parts or pieces of equipment and managers either looking the
other way or accepting it as a given. It always comes down to how ofen it
is happening, multiplied by the cost impact, versus the cost of the simplest
and least expensive solution.
I found similar things happening with tools, fxtures, and even a poorly
functioning mop bucket being used to clean up oil spills. If you are getting
the impression I am easy to convince to spend some money, you only need
to speak with my friends and they will tell you that it is usually necessary to
pry those dollars out of my wallet. Tis desire not to spend money foolishly
is the exact reason I spent money to eliminate these ridiculous little prob-
lems that were really costing me much more money. I really view all these
non-capital expenses as wise investments with great returns. Te trick is
that you must be involved with the details to understand the efects and the
savings. Once again drawing on my earlier analogy related to home projects,
when you buy a new, superior tool to replace an old, infuriating tool at home,
you think of it as a great savings. It can be harder in the work environment if
you are not involved in the task and only seeing it from the angle of signing
a purchase order instead of really understanding the full impact. In the case
of just signing purchase orders for equipment or tooling, you are also forced
to assume that whoever proposed the solution is a good problem solver, that
he or she has done some root-cause analysis, and this is really the best and
least expensive solution to eliminate the problem at its root level.
6.9 cost JustIfyIng A teAM leAder
One thing I still had lef to do was justify (mostly to myself) through a
costbeneft analysis that changing Toms role essentially to a team leader
was a wise investment. Under Bobs direction, he had functioned more
then half the time as a direct employee by running a machine; under my
new plan, he was operating a machine less than 25 percent of the time. It is
like I turned a somewhat direct employee into an indirect employee. Most
of us have been brainwashed into thinking of direct employees as the real
money makers and indirect employees mostly as overhead or burden cost.
I had learned while working at Toyota that they did not have this direct
versus indirect mentality or vocabulary. Most companies are content to let
their employees intermittently perform most supplemental tasks related
to their primary job. In manufacturing, this frequently involves getting
your own materials, making small repairs, and basically handling other
problems with little guidance or support and only asking for help when
you have determined it is beyond your means. With this comes, most
managers unconscious acceptance of about a 70- to 80-percent uptime
or efective operating time for their processes. On the other hand, Toyota
targeted an uptime in the mid-ninety percentile, recognizing this could
only be achieved if operators could remain at the workplace focused only
on performing their direct tasks. Tey further understood that not every-
body had the same problem-solving skills and it was better to have some-
one trained and skilled to solve a problem at its root cause than having an
individual operator band-aiding it. Toyota also knew these team leaders
were critical to resolving sufcient problems, thereby allowing them to set
up connected processes that operated with a respectable efciency without
signifcant WIP to hide problems.
Since most of my sales dollarsand therefore, proftcame from man-
ufacturing parts on the machines (I provided other value-added services
like assembly, deburring, painting, plating, etc., but they were not my
bread and butter), I had to keep good parts coming of the machines. Te
best way to measure this and to assess the additional value being added by
Tom as a full-time team leader was to compare the OEE before and afer
Tom had been in the position for a few months (and was likely to be having
an infuence). Previously, I indicated that when Tom and I frst discussed
his interest in taking over, we agreed to defne the team leaders respon-
sibilities before we made the change. Te biggest reason for this was that
I wanted to gather data on the current OEE level of each machine. I knew
that if this number increased, it would be simple to convert it to dollars
and compare it to Toms wages.
Lean principle: ROI (return on investment) measures how efectively
the frm uses its capital to generate proft and for machining processes
this can be determined by measuring OEE (overall equipment efective-
ness) improvements. I have found it best when companies convert what
a 1-percent OEE improvement means in terms of productivity or poten-
tial proft for each machine or process. Tere is normally an investment
made when improving OEE and that should be openly compared to the
benefts. In my case, there was a cost in making Tom the team leader,
though the ROI more than justifed this cost.
Afer a few months of Tom assuming most of the responsibilities of the
team leader (summarized in Appendix G), I compared the previous and
the current OEEs. As it usually goes with data analysis, it took a little bit
of interpretation, but basically I was seeing an average of a 7- to 8-percent
OEE improvement on our critical machines afer the frst few months. A
rough translation of this to dollars from our fve main machines, which
at the time were unfortunately only averaging thirty hours scheduled pro-
duction per week, was
30 hours/

5
8% 13 additional hours
week machines improvement
=
of production per week.
Multiply the additional thirteen hours per week that I gained (assume I
was able to sell most of that time, which was realistic for us) and multiply
by the average of $60 per hour for which we sold our machining time, for
a total of an additional $780 per week of sales (with no other additional
costs). Tis almost covered Toms wages (without benefts). But remember:
Tom also still ran a machine up to 25 percent of the time and performed
many other value-added activities like functioning as the supervisor. It
proves the correct decision was made and implies that you should not get
hung up on the number of direct and indirect employees; instead, focus on
the value each position is adding.
Key point: Do not be concerned with whether an operator is classifed as a
direct or indirect employee; instead, measure the value added and quan-
tify the productivity through a measurement like OEE improvement.
Having a team leader in place allowed the operators and setup men to
always be at their machines, which not only kept the machines producing
more good parts, it also made it easier to manage. Now, if people were
observed not to be at their machines, Tom or I would instead think of this
as an abnormality, instead of just accepting that they were probably look-
ing for material, looking for tools, or working on a problem. I shared my
calculated savings with Tom, showing him the thirteen hours per week we
had gained. Naturally, I wanted further improvements and asked for his
input. In reality, he threw out a few small ideas to fll in the silence, but
I think he had been too busy to give it much thought and seemed satis-
fed with what we had already achieved. I asked him to scribble down any
ideas he came across during the next few weeks. Unfortunately, I had no
suggestion plan to fnancially reward individual ideas; but, for the most
part everyone was motivated. My team had not become so accustomed to
overtime that their lifestyles depended on it, so they were not motivated
to slow things down during their normal shif to create the need for over-
time. Tis led to being open-minded in helping the company develop and
implement improvement ideas. Also, I was too involved with the day-to-
day operations to let wasted time go unnoticed, and we had a measure-
ment where we continuously compared our actual times against either our
quoted times or our previous build times.
Te ofer to increase pay in line with local industry standards was serv-
ing two purposes with relation to improved productivity: (1) it motivated
the employees to better themselves by acquiring skills they were lacking,
and (2) it benefted the company as I had a better trained workforce, which
minimized manpower shortages (creating fexibility to better balance
capacity in line with demand). With the planning boards being utilized,
the operators could not slow down their work pace when they noticed the
workload diminishing; instead, they would either work on my ongoing
miscellaneous task list or cross-trainand since I paid based on abilities,
this could put money in their pockets.
All of these ideas as well as others I have likely forgotten helped get more
parts out in a shorter time, one of every manufacturing managers goals.
6.10 closIng the books becoMes AntIclIMActIcAl
Te monthly closing of the books was becoming a non-event for us for a
few reasons. We were monitoring productivity and material consumption
so closely that we knew, job by job, whether or not we were proftable.
We also monitored weekly spend on indirect purchases, payroll, etc.,
and our accounts receivables were continuously updated. Further we had
improved our accounting system to include accrued liabilities each month
to cover items like equipment maintenance, new capital purchases, etc., so
continuing to show proft was even a more signifcant event. Terefore, we
had our fngers on the pulse in real time and the end of the month was as
simple as tying up a few loose ends.
6.11 suMMAry
Managing each job visually is one way to judge things in smaller
time intervals and react quicker, which is the backbone of lean.
Te measuring and adjusting phase of a changeover frequently
accounts for over 50 percent of the total changeover time; therefore,
it is critical to improve this part of the process.
To minimize the time necessary for the adjustment phase of a
changeover, quickly return the process to the point where you lef of
the last time (with a quality part). Tis requires documentation and
reference markings.
Proft sharing is ofen difcult, though a good suggestion system that
rewards implemented ideas is a step in this direction.
Team leaders are ofen cost-justifed positions with very high rates
of return. You should not dismiss this idea, especially on the basis of
the position being considered that of an indirect employee.
95
7
Seeing the Administrative
Processes from a New Angle
7.1 IntroductIon
I was on the downhill side of making it through my frst year as a busi-
ness owner. I would also like to call myself an entrepreneur because the
word has a nice connotation. But, I have always felt this word describes
someone who starts something from scratch, comes up with a new idea,
or invents a new product. I had essentially taken a proftable business and
further improved its proftability using the OSKKK (Observe, Standardize,
Kaizen Flow & Process, Kaizen Equipment, and Kaizen Layout) strategy.
Sales levels remained fat, which was tolerable only because the industry
had been sufering a slowdown at the time and many of my competitors
had seen sales decreases near 40 percent. Terefore, I deduced that main-
taining my sales level in and of itself was an accomplishment while simul-
taneously learning the business and starting to improve proftability. I was
also envisioning what would be possible during the good times within the
semiconductor equipment manufacturing industry, plus I was starting to
get a feel for where I might be able to generate sales in new markets.
I had just returned from Spain, where I tied the knot. Other than a few
embarrassing moments with Spanish mispronunciations, including one
in front of a packed church where I apparently said a word that came out
sounding like a sexual organ, the wedding was great. Tis was the frst
time I had been completely away from the business for more than a week.
A key goal in buying the business was that robust processes would allow
it to function without continuous on-site management. Te former owner,
Bob, had never been away from the business for more than a day or two.
In my absence, Tom (the shop foreman) kept the shop foor running while
Sandra managed the ofce. So, the fact that I was able to get away proved
that standardization of the shop and ofce processes was paying of.
7.2 vIsuAlIzAtIon confronts obtAInIng
InforMAtIon eIther froM
coMPuters or In MeetIngs
I needed to get refocused on growing the business, partially through its lean
transformation. Te OSKKK methodology was helping to reduce the time
necessary both for ofce tasks and shop tasks, though the best way to proft
from this lean transformation was to sell the additionally created capacity.
Te shop was functioning better, and with the few visuals I had in place I
was able to quickly grasp where problems were being encountered when I
was on the shop foor. I cannot tell you how many businesses I walk around
in as a consultant and feel sympathy for the production managers. Although
they are typically made aware of the larger catastrophic problems, there are
typically no indicators as to what should be happeningno visual plan to
establish normalI am always forced to assume that these managers just
get historical (afer the fact) information in meetings and reports and cannot
quickly discern abnormal conditions in the workplace. Te frst problem
with using historical data is you have to be able to pay for previous problems
and the related losses because by the time you learn about it in a meeting,
the money is already spent. Or, the manager is lef to assume he or she
would not have had any valuable input while the problem was unfolding.
Second, you have to be content with only ofering potential solutions for the
smaller problems whenever they might occur again or learning about the
large problems when they are already into their later phases.
Since I did not have the money to support unproductive machines or peo-
ple and felt my problem solving was better than average, I wanted to know
about the problems from the start. Most organizations have trouble defning
exactly when they began encountering a problem and become accustomed
to all the daily frefghting of the small issues. It is not always easy to determine
SeeingtheAdministrativeProcessesfromaNewAngle 97
when a task or process can be said to be in an abnormal condition. But to
arrive at this point you must have clear boundaries for normal and abnor-
mal, which are usually clarifed through visualization. Most organizations
feel they have so many problems that they can only respond to the big ones.
So, everyone is lef to his own devices to determine when a problem is big
enough to call for a resolution. Consequently, most abnormalities are swept
under the carpet. Terefore, throughout my tenure I spent considerable time
defning and refning what we considered abnormal and then began pro-
cesses to quantify and deal with the signifcant issues.
7.3 not AlloWIng excuses for A lAte stArt
Another change that I made afer putting Tom in the team leader position
was to have him start preparations somewhere between ffeen and thirty
minutes prior to the start of the shif. During this time, he made sure the
air compressor was started, he turned on all machines and let the older
CNC (Computer Numerical Control) machines warm upit seemed that
if the older machines were started with a dummy program, running the
machine through movements similar to those for which they were frst
scheduled, they would hold better accuracy. I guess it is like people getting
older, needing more time or cofee in the morning before we can function
properly. Depending on the part the machine was to run, it ofen seemed
that ffeen to twenty minutes of warm-up sure helped in quickly obtain-
ing good parts. Tom was able to start a few machines and let them run
simultaneously while he prepared other tools and materials. As a result,
there was no need for the entire team to be delayed waiting for machines,
looking for tools, waiting for compressed air, etc.
Lean principle: TPM (total productive maintenance) is a philosophy
to integrate maintenance into the manufacturing process and involve
machine operators. Te goal of TPM is to eliminate downtime losses
caused by equipment stoppages that are restricting the production of
good parts within the cycle time. An indicator of how successful TPM is
can be seen in the OEE measurement (overall equipment efectiveness).
Although Tom handled many parts of our TPM program, the operators
were also trained to be proactive in supporting it.
During this preparation time, Tom also put out any special measuring
tools required, checked their calibration, and made sure the machines
were in mechanical order and were topped of with coolant, lubricating oil,
etc. He also stayed afer the shif to perform anything he felt he would not
have time for the next morning, like more involved TPM activities. Tis
machine maintenance is part of the second K in the OSKKK strategy:
Kaizening of equipment. I felt this was some of the best overtime money
ever spent; there were rarely excuses for late starts other than those Tom
encountered, like periodically having a machine or material problem. In
the cases where he could not correct the problem, he would call me (prior
to 6:00 a.m.) and ask what to do. My wife and I both dreaded those calls,
but we agreed they were necessary since some of this saved time would end
up increasing proftability. Also, we were working to permanently resolve
these issues so the same ones would not be encountered again. Tis also
helped in our measurements. For example, the hour-by-hour planning was
simplifed because we could arrange for production to start at full speed by
6:00 a.m. and did not have to factor in any start-up time. I learned this
technique during my training period at Toyota while they had me supervis-
ing an assembly line. All four team leaders on my line were paid to come in
early and get everything prepared. Toyota knew it was more expensive not
to do this; for example, to stop an entire line while waiting for one operator
to warm up a sealing machine could cost thousands of dollars. I recom-
mend this to most manufacturers; therefore, I want to highlight it.
Key point: Having a team leader or supervisor come in early to prepare
all workstations is money well spent. It not only eliminates the excuses,
it also lets you plan your production better. It is better to have one person
paid a little overtime rather than to have the whole team milling around
looking for tools, or waiting for a machine to warm up or a problem to
be resolved.
7.4 sMAll busInesses look At the
recruItIng Process dIfferently
I recall that during this period my attendance policy (a combined total of
three times late or absent in a rolling ninety-day period meant dismissal)
had caused us to lose two operators. As I ran ads to replace these opera-
tors, I received a lot of interest. It was not hard to fnd trained operators.
Between the San Francisco Bay area and San Jos, there were nearly 3,000
machine shops, and a lot of trained operators already lived in the area.
Tere was a lot of development work for high-tech industries that sup-
ported this, although most of these industries were also cyclical.
Te basic rule I had learned from Bobs experience of running a small
company for eighteen years was to avoid hiring those who had worked for
large companies. Tis probably sounds harsh, but when speaking to other
small business owners I found a general consensus on this point. Te basic
reasoning was that many employees in large companies were used to an
easier working pace, and they thought about things like unions, cigarette
breaks, and how they could get overtime. Tey were ofen not accustomed
to being so directly tied to proftability and to handling the wider range of
skills and tasks required by each employee in a small business. Naturally,
generalizations are always dangerous, but I can attest to a diferent type of
worker within the small shops. Having observed horrifying work habits
and bending of the rules while working for General Motors, with certainty
I can say that nothing similar happened in my little world. Tom and the
other setup men I inherited with the business had always worked in small
shops where you were expected to be a team player and do a fair days
work, handling everything from running a machine to sweeping the foor.
And a small shop meant the boss was more involved and concerned with
the results. Tom had actually worked for Bob for the previous sixteen or
seventeen years so he had really been mentored and molded into a respon-
sible and hard-working machinist.
During this period, I had made the mistake of overlooking this advice
when we hired a setup man who had worked for a large machine shop
in a semiconductor equipment manufacturers planta large company.
Since big businesses frequently have larger training budgets and undertake
more formal schooling, this man had achieved a great pedigree including
many courses and certifcates he had completed. With all his training and
his experience in a large and varied machine shop, I really felt that he
would bring a lot of insight and new ideas to my little world.
Because it is better to make such things clear during the interview to ensure
you are not wasting each others time, I explained the environment of our
shop: We did not have a one-man-to-one-machine mentality, and a setup
man might fnd himself operating more than one machine simultaneously
and would even be required to take his turn at certain cleaning tasks. He
indicated that he understood what I was telling him. Afer about two weeks,
however, I began noticing his pace was much slower than the others and he
was taking many additional breaks to have a smoke.
Te deal breaker, so to speak, happened during his third week. His work
habitsthe slower pace and the cigarette breaksand his know-it-all
attitude were already getting on Toms nerves. Ten, Tom overheard him
telling the others that a ffeen-minute morning break and a ten-minute
afernoon break were really not enoughat his previous job, they had
twenty minutes in the morning and twenty in the afernoon. So, I cut his
last check and quickly walked him to the door. It is just that easy in a small
business, especially within the frst three months of a new hires employ-
ment. Te last thing I wanted was a bad apple poisoning a good team, and
we all know it frequently only takes one.
More regulations apply to terminating an employee in larger compa-
nies, but I fnd the rules are not the reason many managers put up with
the underperformers. Te reason has to do with indecisiveness and a lack
of documentation in the employees fle. Again, not documenting each
case in which an employee is not performing as expected is either a prob-
lem in the process or a lack of disciplined management. Whether or not
you have had similar experience or agree with this suggestion of a prefer-
ence for hiring employees with their previous experience in only small- to
medium-sized companies, I want to capture it as a key point.
Key point: When hiring, assuming you fnd candidates with similar
backgrounds, the preference should go to those with previous experi-
ence in smaller frms as they are likely to have better working habits.
Documentation of all instances when an employee is not meeting expec-
tations is also key to successful termination of these workers.
7.5 focusIng to IMProve the offIce Processes
While the shop was running well, I wanted to focus some more on our
administrative processes. I felt I could further cut some costs and lead time,
especially before soliciting work from new customers in other industries.
I wanted to better understand our processes and refne them, and then I
could use these points to convince potential clients that they would want
to work with us. I also wanted to look into getting some accreditations
from regulatory organizations to distinguish us from others. Te accred-
itation that I was fairly familiar with was ISO (International Standards
Organization) 9001. ISO was not as critical for second- and third-tier sup-
pliers in the industry we served, but it was becoming more common to
see the larger machine and fabrication shops promoting it. (Most custom-
ers individually qualifed us by having representatives from their quality
departments periodically assess our quality systems during on-site visits.)
Lean principle: OSKKK is an ideal process for improving both the shop
foor and the administrative processes. Deciding where to frst focus your
eforts (and limited resources) should be based on where the major bot-
tleneck resides with respect to getting products out the door. Remember:
a shop foor can only function efectively when all the support (admin-
istrative) processes provide the information and materials on time and
tasks are performed in a timely manner.
One of the frst areas I looked into with a view to administrative process
improvement was the speed and ease of locating information, which I felt
caused several bottlenecks. With only three of us working in the ofce, the
lack of a computer network may not seem like a big issue, but without it
there had been no easy way to share many tasks and have everyone work-
ing from the same data. For most of us, computer networks are a given, but
when I took over the business all the PCs were stand-alone. I had fnally
gotten all the computers networked, including the shop computer which
was further connected to all the newer CNCs, allowing programs to be
sent back and forth throughout the entire network. I could even access
the network remotely, and any programming I did at home could be sent
directly to a machine. Tis had been completed for some time but since
I had not begun cross-training or taking advantage of the network, few
benefts had been realized and everyone still kept all the data they man-
aged on their respective hard drives. Setting up a network is not worth
discussing in more detail, but it is important to note that without prop-
erly organizing this taken for granted infrastructure, the optimizing of
many tasks and information is signifcantly more limited.
Te perfect time to organize data is when you start fresh with a big
change like a new network. I knew any efort spent looking for information
was costing me time and money. Because everyone was now networked,
they needed training on how to properly store data flesit would have
been counterproductive to lose time looking for data in a randomly cre-
ated structure and fle-naming system. For lean practitioners, this would be
considered 5Sworkplace organizationin the administration area. Since
we had an eighteen-year history of everyone saving their fles on their indi-
vidual PCs, you can imagine how many variations of fling systems we had.
I did not want to go back and sort out all those records, so we decided on
some new rules going forward, pertinent to all new fles and information
being created. When the network was installed, we only put in one shared
drive. We encouraged everyone to segregate all necessary and timely com-
pany information currently residing on their individual hard drive and to
consider how it should be organized according to the new protocols.
Te biggest problem we had was our outdated, DOS-based order-entry
sofware (similar to an MRP [material requirements planning] system),
which had not been designed for a network with multiple users simulta-
neously extracting and saving data. Using Microsof Access, a friend of
mineone of those technology whiz kids who boggle your mind when
explaining how things are working inside that boxwas able to extract
data from this antiquated sofwares database and ultimately move it to an
Excel spreadsheet that everyone could access on the network. I frequently
fnd when the central computer system cannot perform a task, an auxiliary
Excel fle is necessary until a better system can be found. Using Excels
tracking feature, we could all search for and extract the data we needed
from the DOS program and understand who had made any changes. My
friend even included an automatic update function so the data was fre-
quently extracted from DOS and refreshed in Excel. Now, everyone had
access to our orders and the material on order. We also brainstormed some
other ground rules regarding how to store and retrieve new data.
7.6 PurchAsIng Is A crItIcAl AdMInIstrAtIve
functIon to look toWArd IMProvIng
One area where I frst focused on taking advantage of the networks
abilities was in the purchasing function. I wanted Sandra to move the
purchasing data over to the new shared drive in the organized format I
mentioned earlier. Sandra alone had been handling this data in the same
way for so long that it begged improvement, mostly because no others had
been involved (ofen we think our own process is OK until others bring in
their ideas andhopefullyopen our minds to the possibilities).
Up to this point, the only change of any substance in purchasing had
been the simple tracking system for late deliveries that I implemented dur-
ing my frst few weeks. Although most purchasing improvements center
on price or inventory reductions (usually brought about through reducing
the minimum order quantity or the lead time or both), I had only about
an 18-percent material cost on the average, which we tightly controlled.
And although I worked to provide sufcient time for the buyers to col-
lect multiple supplier quotes, I was now more interested in establishing
some standard stock material. I determined this as a necessity because,
although our supplier lead time was only one to two days for cutting plate
material to size, I was fnding we periodically had problems obtaining the
minimum order size and I was beginning to see more requests for even
shorter lead times from our customers. A lot of these short lead times were
not possible to achieve if we had to order our raw materials by 3:00 p.m.
to receive them the next day or the day afer that. Terefore, keeping some
strategic inventory would allow us to quote on those really-short-lead-
time RFQs (request for quotations) if they could be cut from an on-hand
stock of standard bars or sheets of material.
I provided Sandra with the sizes of fourteen diferent bar stocks and
sheet materials I wanted to keep in stock and told her to start with a
maximum of four pieces and minimum of two pieces for each of the arti-
cles. I also asked her to think about a simple reorder system. Naturally,
a lean practitioner would think in terms of a Kanban system; however, I
did not want to tell Sandra how to redesign the reorder system. Instead,
I wanted to hear her ideas and then guide her in the right direction. She
basically nailed the Kanban concept with her suggestion: somebody from
the shop would notify her every time a complete bar was consumed, and
when two of any type were consumed, she would order more (the lead
time for these sheets and bars ranged from one to three days). I pointed
out we should have the system standardized in case she was on vacation
or the guys in the shop forgot to let her know. (Mentioning the need
for vacation is always a good way to get people to buy into documenta-
tion or standardization. Knowing someone else can handle a specifc
task allows the person who usually performs it to take their vacation).
We agreed to attach a simple (Kanban) card to each bar; as the shop
guys fnished a bar, they would bring the card to Sandra and she could
track it on the wall with predetermined slots for the various cards. I fur-
ther indicated that she needed to monitor (on the same board where the
cards were hanging) how ofen we reordered, how ofen we were short,
or how ofen we immediately required more of a particular material. My
maximum of four, minimum of two had been just a lazy mans guess as
a starting point, but with a little thought and data she could likely adjust
that up or down for each material. I further asked her to monitor other
typical materials she frequently ordered and to consider stocking some
of those. Tis also helped reduce the administration time and efort in
purchasing since orders for stock materials can be grouped together on
a blanket purchase order and release against it instead of having to
generate separate orders.
Lean principle: Kanban is a Japanese word that essentially means a
system of notifcation to replenish what has been consumed. Tis is in
contrast to MRP, which uses forecasts and lead times (looking forward)
to anticipate what the customer will likely consume, and advises on
ordering materials and planning internal production. Unless you have
accurate forecasts that do not exist in made-to-order businesses, replen-
ishment style reordering is superior to MRP order systems. Kanban
replenishment is a better option for runner materials (discussed
in Chapter 6) and others should only be purchased or manufactured
according to the customers orders.
You might think it is crazy for a lean guy to suggest new or additional
items to stock in the raw material inventory in what is essentially a
made-to-order business. But I knew that although some of our custom-
ers were price sensitive, the priority periodically became the lead time.
In that case, a lead time of two to three days could take me out of the
ball game on a potentially high-margin order. I would also like to clarify
there is a big diference between unplanned, unmonitored, and unneces-
sary inventory and what I call strategic inventory, which is determined
as necessary and is kept at a minimum by continually monitoring and
adjusting it.
Lean principle: Inventory is a necessity to compensate for the difer-
ences in the lead times and minimum order sizes required by customers
and the lead times and minimum purchase quantities ofered by suppli-
ers (or the internal supplying processes). If the customers desired lead
time is shorter than the suppliers (or supplying processes), it is necessary
to hold enough inventory to compensate for the diference (until you can
improve the suppliers lead time). Tis was becoming the case for us as
the customers demanded shorter lead times to compete in this proftable
market segment. Te inverse is also true: When the customers lead time
is longer than the suppliers, no inventory is required and you can either
buy or build to order. When the inventory is there because you have
planned around this and have worked to further reduce this diference
in lead times, I like to call it strategic inventory so it can be distin-
guished from unplanned or unnecessary inventory.
When organizations have signifcant capital tied up in inventory
(which was not our case) and want to work to reduce this, not only do
they need to involve the right people and departments (one of which is
purchasing), they also need to use the proper measurements to focus their
improvements on the root causes of the problem. Remember: inventory
is most notably the result of two factorslead time and minimum order
size (accuracy of forecasts also has a strong infuence). So, measurement
should refect these infuences. Some organizations still refer to inventory
in fnancial terms (i.e., We currently have $500,000 of inventory), which
makes it difcult to judge whether there is too much and on which prod-
ucts to focus. Others have moved to referencing inventory turns, which
is annualized cost of goods sold/average inventory. However, the correct
lean measurement is DSI (days supply of inventory, also called days sales
of inventory). DSI is the converse of inventory turns but, because the
units are in days, it can be directly compared to the lead time (which is
also in days) for the parts involved and it becomes easy to judge whether
you have too much or too little.
For example, if someone tells you that you have a DSI of twenty-fve days
(meaning, under normal usage you have a twenty-fve-day supply on hand
for that part), and you know the lead time to either purchase or manu-
facture the part is four days, then you have too much inventory and your
target should be four days plus possibly some safety stock.
Lean principle:
Days supply of inventory or days average inventory/
sales of inventory (DSI)
=
cost of goods sold

365.
Or, another way to look at it:
DSI = average inventory/cost of goods sold per day,
which is the converse of inventory turns but a much better measurement
because the units are in days and can be compared to lead time, one of
the most infuential reasons for inventory. Many organizations struggling
to reduce inventory do not clearly understand the reason for it and the
factors infuencing it. But if you know your lead times for the various part
numbers and compare those to the DSI, you can have a focused discussion
on what is a reasonable target for that inventory (by specifc part num-
ber) and what can be done about the lead time to further drive improve-
ments. Tis also links to the DSO (day sales outstanding) measurement
(discussed in Chapter 3). Remember: if you could order a minimum of
one piece for every part number (without cost implications) and have it
the next day, there would be little need for any inventory.
Since I arrived at the stage in my lean transformation that I viewed the
supporting processes from the front ofce as currently generating the
most signifcant bottlenecks, I was to the point in my OSKKK strategy
of working on Kaizening Flow & Processes in the ofce. Another oppor-
tunity in the front ofce was to train Andrea, the ofce assistant, to handle
the purchasing (and minimize delays to our customers) when Sandra was
on vacation or sick. Tis requirement to train someone else on the purchas-
ing process had been previously identifed on our cross-training matrix, as
shown in Appendix H. As a starting point, I provided Andrea with the doc-
umentation I had made up while I was being trained by Bob, and I asked her
to further detail this while she was being trained and complete any steps I
had missed. Again, there is no better time to document or improve the exist-
ing written procedures than when someone is learning the process.
I was also seeing RFQs involving materials for which we had little or no
previous experience. Determining how to cut and work with each mate-
rial was important in the shop, but in the ofce it was critical that various
materials could be purchased at a reasonable price with a short lead time. I
had asked Sandra to document the available suppliers and pertinent infor-
mation for each material. I worked with her to further develop this with
a more refned Excel sheet in which we could search for primary and sec-
ondary material descriptions as well as vendor, lead time, and price. I also
wanted her to continue adding information on any new materials that were
being requested, whether or not we received the order. It would always be
good for future reference, and building a valuable material database helped
to justify the lost time on quotes for which we did not receive an order.
7.7 unsuccessful At sIgnIfIcAntly reducIng
the tIMe necessAry for quotAtIons
Quoting was another area that posed some challenges in terms of produc-
tivity. In particular, the signifcant amount of time that was necessary for
putting together some quotes (calculating setup and run times, specifying
and quoting material) could be reduced or eliminated if a similar quote
that had been previously prepared could be located quickly. Bear in mind,
customers periodically asked for the same or similar parts to be quoted.
So, a reference system to locate quotations would be a great beneft if it was
quick and simple. Like a lot of database-type projects, this begged the
typical question of how much detail was required for quick access to past
quotations. I had to think about the expected outcomes and the expected
savings. At that time, our system was only to fle a hard copy of the quote
by customer with any attached documentation. Once the quotation fle for
a particular customer got too thick, it was put on my desk to sort through
the old quotes and decide whether any were worth keeping, on the chance
we might still receive the order. Most of our business was represented by
short lead times, and the majority of quotes for which we did not receive
an order within one to two weeks were lost opportunities.
Te time and (potentially) money to set up this database were not the
only considerations. Te time to populate it with each quote and the time to
search through former quotes for all RFQs with the probability of no match
being found were highly likely, which would mean we would have wasted
efort on both the inputting and searching. Naturally, this needs to bal-
ance against the time savings once a similar quote was found and against
how ofen this was likely to happen. I came to the conclusion that it would
be less than 10 percent of the time when we built a part similar enough
to share quality information between the previous quotation and the new
request for quotation. Terefore, Sandra and I decided to abandon this idea
and leave that part of the quoting system as it was. I think she was relieved
that she was not going to inherit the task of entering all this data for each
quotation into the computerwe frequently quoted between three and ff-
teen part numbers per day, so there would have been a lot of data entry.
However, I still wanted to make one simple change that involved a more
systematic follow-up on quotations for which we did not receive an order.
We currently did it on a random basis, and I wanted a meaningful system
that could help to improve the quoting process without too frequently
contacting the customers purchasing agents for feedback (afer all, how
ofen would you want to explain to suppliers why they did not get their
order). I also fgured that periodically it might provide a second oppor-
tunity to quote a portion of the order if we had come close or the buyer
preferred to work with us, etc. I thought we could somehow designate for
follow-up only the critical quotes that we felt were either our type of part
or ones in which we had put a lot of efort. Sandra could also keep track
of our hit rate with various customers, and where we had a low rate on
receiving orders, she could then choose a few quotations to follow up on
and feed back her thoughts to me and Jake (the programmer/estimator)
about why we had not received an order. So, Jake and I began marking
quotes we thought would be good candidates for follow-up. Sandra noted
on her standardized work calendar that on Wednesdays she should go
through the quote fles and determine any follow-up to be made based on
customers with low hit rates.
You might be wondering what a standardized work calendar is. It is
a simple monthly calendar for each person in the ofce to mark periodic
tasks for which he or she is responsible and when to follow up. Ofen, peo-
ple plan to follow up or monitor something each week but they get busy
and forget. When consulting, I frequently ask questions like How ofen
do you confrm the suppliers on-time delivery? and a typical response
(afer a long and painful pause) is More or less every other week. Which
really means rarely, if ever. I did not want that to happen; so, my two
administrative employees and I each hung a calendar near our desks,
showing the day on which we planned to perform periodic (usually weekly
or monthly) tasks. Ten, we would mark of each task as it was completed.
Naturally, some tasks fell behind a day or two; much longer than that,
and I was bound to take notice and ask about it. Te hardest part of this
was to lead by example, as I frequently got busy and missed a task or two
myself. Because I had made it one of my weekly tasks to review the other
two calendars, I had to get my own things in order before pointing out
anyone elses shortcomings.
Lean principle: Standardized work for managers is really a critical
element of managing within the PDCA (Plan, Do, Check, Act) cycle. It
puts some structure into everyones week so they dont get busy putting
out fres and forget the weekly/monthly monitoring/follow-up that they
should be committed to. Usually, it is best to visually display the items to
be followed up on and when they will be done so everyone is held to the
schedule. Some managers like to set up this management audit on a ran-
dom schedule so the various departments do not know when a review
will take place and must keep everything up to date at all times, though
this can cause resentment.
7.8 5s for shAred coMPuter drIves And fIles
Returning to transferring all pertinent data to the network drive instead of
going back through eighteen years of data on individuals PCs: We decided
to start immediately creating most new data fles on the network. Files
that were determined to be common and relevant were also moved to
the network drive. I had implemented a standard convention for naming
and locating the directories and the fles and I had established maximum
levels of subdirectories, etc. I was not sharp enough to understand both
our current and future storage requirements on the shared drive; instead,
all my guidelines were open to change and adaptation but only afer we
tried the current suggestion and later agreed on the need to modify or
adapt it. No two people will ever fle something the same way when there
is no preexisting system; therefore, others will lose time trying to locate
items or fnally give up and ask. My system was not perfect but, because of
certain rules pertaining to the title of the directory or fle (how things were
worded, word order, numbering, date formats, etc.), the search function
became a much more applicable tool.
I was hoping that all the ideas I was suggesting would encourage oth-
ers to suggest various improvements, and slowly they began throwing out
more of their own ideas. Since it is not reasonable or possible to implement
all suggestions, I had to explain why certain ones were worth doing in the
near term and why others were not feasible or did not seem to have enough
beneft to ofset the implementation costs. I was also hoping to develop
Kaizen eyes, not only in the shop foor team but also among the ofce
personnel, and I believed from the suggestions I was getting, they were
starting to see the waste within many of the processes.
7.9 outsIde quAlIty AccredItAtIon
One other item I was determined to work on during this time period was
to receive a quality accreditation from an outside source to help in being
recognized by larger organizations as a qualifed supplier. Most quality
departments of the larger customers for which I was currently manu-
facturing parts had diferent processes in place to certify their suppliers.
Someone from the quality team would typically audit our processes at
various time intervals to ensure we had robust processes in place to meet
their quality standards. We appeared able to meet their various require-
ments, but I also wanted to hold an externally assessed accreditation to use
as a marketing tool.
My priority was accreditation for quality standards, not for environmen-
tal or health and safety standards. Te ISO was what I was most familiar
with and was mainstream when I had been in the automobile industry.
As I started down the ISO path, I recall it was relatively inexpensive to
download the manuals for the ISO 9001 (the quality accreditation) for our
industry. Te manuals contained surprisingly little direction on specif-
cally what was required. Te basic guide was to defne the companys pro-
cesses for either X, Y, or Z, and then ensure they were robust and being
adhered to. When I looked into the cost of having an auditor come to
our plant for review and accreditation, I was shocked. I do not recall the
exact cost but it became apparent why so few small companies pursued
it. I decided instead that the quality systems we already had in place only
needed some small tweaks to meet the requirements spelled out in the
ISOs manuals. So, we put the additional process steps and documentation
in place and, instead of telling our customers we were ISO 9001 accred-
ited, I mentioned we were ISO 9001 compliant. A bit of a play on words,
but we were following the guidelines. We just had not spent the money for
the actual accreditation.
7.10 suMMAry
A team leader should be brought in prior to the start of the shif to ensure
all materials, information, tooling, programs, documentation, etc., have
been prepared, so there are no excuses for late starts and planning can
expect a productive frst hour with no abnormalities encountered.
It is worth considering where potential employees have previously
worked and whether those companies are known to be lax in their
policies (sometimes this is infuenced by company size).
Administrative process improvement should follow the OSKKK
strategy and if there is no apparent bottleneck process, purchasing
is always a good place to begin, especially for companies with a high
purchasing spend.
Strategic inventory is necessary when the expected customer lead
times are shorter than the supplier lead times (or manufacturing
lead times). In these cases, a replenishment system (Kanban) can be
utilized for regularly repeating runner parts (also known as A
parts in the context of ABC analysis).
Standardized work for managers should be a visualized plan of
their commitments to monitoring/follow-up. Tis is a necessary
step of the PDCA cycle, especially in organizations consumed with
frefghting.
113
8
Moving the Business
8.1 IndustryWIde doWnturn
Te semiconductor equipment industry continued through one of its cycli-
cal slowdowns. Of my seven major customers, fve had noticeably slowed.
Up to this point, I had only been able to diversify with one new customer
in the medical device industrya contrast to the semiconductor equip-
ment industry in that it appeared less cyclical and price-sensitive. Tis
industry was a reasonable size in the Bay Area so I had decided to pursue
new customers in this sector. I had built about eight diferent parts for
the new customer and only one had repeated, for a total of nine orders. It
was a start but not enough to ofset the efects of a major downturn in the
semiconductor equipment industry.
Unfortunately, I had to let one of my operators go because even afer
cutting the work week to thirty-fve hours, there was not enough work for
everyone. One of the great lean ideologies is the value of the employees
and the importance of ofering them job security. But I had to face the
reality that I did not see much changing in the near term. In choosing
whom to let go, I had more freedom in this small, not strongly regulated
business than I would have had in a large organization. So my decision
was strongly infuenced by the employees ability to add valuehis skills
and attitudeand not based on seniority. While it is true that the opera-
tor I let go had the least seniority at my shop and the least experience in
the industry, he also had the least abilities, a mediocre attitude, and, in my
opinion, not much potential to be further developed.
I now fnd many frmsespecially large ones and those in countries that
are strong in protecting job securityutilizing a strategy of employing
some temporary or subcontracted workers in most areas of the business
to create fexibility with their capacity and to help in getting around rigid
regulations. Tis is one way to create some fexibility within more rigidly
regulated industries, allowing you to follow the business cycles. But it is
never ideal because a lot of training and cost is invested in temps and then
you pay the price when either there is a slowdown of work or they decide
to leave for better opportunities. Also, you pay an hourly premium for
temps, which could be rationalized in the short term if you hire people
through agencies as a screening process for existing openings. But when
you use temps to create fexibility, it usually becomes more long term and
therefore more costly. Tere is no easy answer but, hopefully, cross-train-
ing and fexibility with your permanent workforce allows you to adjust
to some of the smaller cyclical swings with fexible hours, overtime, and
helping each other (this should be visualized by the cross-training matrix
shown in Appendix H). In a short-lead-time business like mine, where all
parts are made to a customers order, there is little fexibility in planning
or leveling the schedule; therefore, lots of fexibility in the workforce is
needed to absorb these peaks and troughs in workload.
8.2 bAckground for MovIng the
busIness At thIs PoInt In tIMe
While Bob (the former owner) and I were negotiating the sale of the busi-
ness, a new investor had bought the building in which Bob leased space.
Te new building owner had increased the rent so dramatically that Bob
was worried it would afect the sale. So, he prepaid the rental increase for
a one-year period to ensure the sale went through and to smooth the tran-
sition. I had accepted this and hoped that by the end of the frst year the
new landlord would realize he was overpriced in the market and would be
willing to negotiate a reasonable contract with me due to experiencing suf-
fcient vacancies. However, when the year was up, the building was about
half vacant but he would not budge on the rent. So, with the current slow-
down in our workload and the new layout opportunities a move ofered, I
fgured it was not a bad time to shop around.
MovingtheBusiness 115
Our layout at that time was a hodgepodge resulting from years of growth
by acquiring various machines and placing them wherever there was space
available. Does this sound familiar? I could not go too far because my
employees all lived close by and did not want to travel. And, my custom-
ers wanted us close by to maintain the high level of service and short lead
times. Tey were accustomed to having someone there right away to pick
up special drawings they needed quotes for, or they wanted to review some
drawings with Jake (the programmer/estimator) or me, or they wanted to
meet to discuss a quality concern.
Lean principle: In OSKKK (Observe, Standardize, Kaizen Flow &
Process, Kaizen Equipment, and Kaizen Layout), Kaizening the layout
is reserved as a fnal improvement step predominantly because many
other improvements require little to no capital while yielding better
returns on investment (ROIs). Tough we were far from arriving at the
last K during our frst cycle through OSKKK, the business case to relo-
cate created the opportunity for Kaizening our layout. Keep in mind
that OSKKK is not a one-of process; rather, it is a continuous loop.
Key point: Any type of slowdown where employees have additional time
available ofen represents an opportunity to work on improvements that
you might otherwise be too busy to consider.
For all these strategic (or large) decisions like moving the business, I
was still using my father as an advisor. Having held executive positions
in companies as well as owning his own business, he had a pragmatic
and logical approach. Also, having invested in my company, he had every
right to ofer his advice on strategic decisions like this. He was in complete
agreement that it was the right thing to do and should not afect the cus-
tomers if done correctly. Naturally, it is great to bounce ideas of someone
you trust and who has the right background.
Te search took only a few weeks and we located some industrial space in
San Jos, only about ffeen minutes from our current location and well-cen-
tered between our customers. Te price was actually less than the portion of
the rent we had been paying during the previous year (remember: Bob had
already subsidized the rental increase for one year). It was 250 square feet
smaller than our current location, which was fne with me. I fgured even
with any expansion I would likely be considering we could make it work.
Also, the unit directly adjacent to the one available for lease (along its south
side) was being utilized for storage by one of the other businesses within this
multi-unit building, and they agreed to consider renting part of the space to
me for future expansion. Ironically, the adjacent business along the north
side was a plating shop that handled many of the processes we currently
outsourced (anodizing, alodine, passivating, etc.)talk about a lead-time
and handling improvement. Te other real advantage was the previous ten-
ant had also been a machining business who had lef all the power lines,
transformers, and air lines in place. Te icing on the cake was the ofce was
air-conditioned, which was not the case in our current ofce.
8.3 creAtIng A neW lAyout froM A blAnk cAnvAs
Before involving the team in the layout discussion, I wanted to ensure
everyone was OK with the proposed new location. I had already taken
Tom (the shop foreman) to the new site to ensure it would not greatly
afect his commute. Te team studied a map of the location: For some, it
was closer to home. For others, it was a little longer drive. But, everyone
was on board.
Planning a move and working on a new layout was right up my alley.
Naturally, I wanted a lean layoutgreat material fow, visualized fow and
visual management, reduced walking, more fexibility of tasks to be shared
between the operators, keeping bottleneck processes out in the open, etc.
I involved Tom and Jake in frst reviewing my layout proposals and then
received input from the others.
I wrote out a list of what I believed were the critical considerations
(naturally based on lean principles), then included three proposals with
material fow lines drawn in to get the thinking started. My list of critical
considerations included the previously mentioned ideas as well as not cre-
ating islands of workers but instead incorporating work cells, the shortest
fow path for large parts and frequently utilized process steps, the criti-
cal and bottleneck processes being visible from the main aisle, and fnally
the shared processes and resources to be centralized (certain measuring
equipment, tumblers, tooling, and the programmers desk as he frequently
supported program changes and improvements).
When considering layout options bear in mind how each operator will
spend his time, especially on CNCs (Computer Numerical Control) or
other machines with automated cycle times, try to balance the machines
cycle time with the workers available time. If the worker cannot perform
another task (and is forced to wait on the machine), then leave out some
small step of the machines cycle and make this a secondary operation
on a manual machine to fll the workers time, thereby also shorting the
machines cycle time thus increasing machine capacity. Tis type of opera-
tor work balance should also be taken into account in any layout proposal
(only completing it for typical parts or runners) or when purchasing
new equipment.
Lean principle: When proposing a new layout (or a layout change), it
is always necessary to draw in the major fow lines. Tis is critical in
evaluating layout options (of which there should be at least three propos-
als). Even if you are familiar with the processes, put in the fow lines. It is
best to use thick lines for frequent trips or large/heavy materials. If your
layout looks like a plate of spaghettiespecially flled with lots of long,
thick linestry another option.
I frmly believe that any manufacturing plant layout diagram that
doesnt include the major fow lines is a waste of paper. Even if you are
experienced in working with layouts, you want to clearly see and ana-
lyze the fow. Bottlenecks and heavily utilized processes should be well-
marked on a layout, and process steps that ofen fow in sequence can be
marked with the same color, thereby highlighting poor layouts when one
color spreads throughout the entire layout. You can get pretty fancy with
fow lines incorporating diferent colors for diferent types of products,
thicker lines representing the movement of heavy materials or more fre-
quent movements, and thin lines representing small parts or less frequent
trips. When the fow lines and these other criteria are drawn in, you can
more easily compare layout options.
Naturally, you must force yourself to draw up multiple options. At
Toyota, they liked to encourage four to fve diferent proposals. Te frst
one or two are easy; then, you start stretching yourself and become cre-
ative. I guess I was a little lazy with only three, but I fgured we would
settle on some combination or adaptation of the ideas in front of us.
Key point: When creating any layout proposal, always draw in the major
fow lines and mark up other considerations like bottleneck processes,
U-cells, highlighting of frequently utilized processes, minimum space for
WIP (work In process), and overall optimization of square footage.
8.4 A nondIsruPtIve Move
We needed to work on a plan for a nondisruptive move that would not
afect our production schedule. Fortunately for the move and unfortu-
nately for our proftability, we were sufering through a slow time in the
industry. We had one thing in our favor: We had very fexible machines,
many of which had similar capabilities allowing a fexible move sequence.
Conceptually, we could move a machine or two and start up production
at the new location while continuing to operate the remaining machines
at our existing shop. An electrician could prep all the wiring and connec-
tions prior to moving each machine. Renting a second compressor and
making a temporary connection into the existing air lines was also fairly
simple. Terefore, we could move one or two machines and their related
tooling, stabilize them and get them back into production before mov-
ing the next machine, essentially operating the two shops simultaneously.
Tom could be based at the new location while I handled regular business
(at least as regular as you can make it during a move) and hold down the
fort at our old shop.
Afer receiving three quotes from equipment movers, we had made the
choice and starting looking at the timetable. Te movers fgured they
could easily move the majority of the equipment during four days, fnish-
ing over the weekend, during which we could move the ofce and remain-
ing material, tooling, and odds and ends. So that was the plan. We went
with a seamless transition.
We had worked through the layout options and arrived at a compro-
mised planas you know, every time you make a new layout, there is a
compromise between various advantages and disadvantages, and rarely
are there only advantages making it an easy and clear choice. We had a
somewhat long, narrow building with the ofce and shipping/receiving
at the front, so we were able to basically form a long thin U with a few
smaller cells incorporated into it. We had only one loading dock to be
used for both shipping and receiving in the new location (not the best
layout but I think it is typical in smaller plants), so our options regard-
ing ideal fow were limited. Our larger CNCs, which handled the larger
material, would be located closer to the shipping/receiving area. Shown in
Appendix I is a partial view of the relocated CNCs; you can see one side of
the CNC machining areas U cell (essentially forming an L with some
of the machines). Other processes would basically follow the process fow
in as close to a large U shape as possible, with the start and end points
meeting near the shipping/receiving area. Following the fow of our typi-
cal process steps: afer our receiving inspection benches was the tumbler
room, where the material was cleaned and tumbled prior to going on the
machines, and again afer the fnal machining process, prior to going to
outside contractors for plating or painting, then parts were moved onto
the various machines. If assembly was involved, it was usually afer any
plating or painting (outsourced processes), so we located the small assem-
bly benches and equipment near shipping. Tis made sense because afer
assembly the parts were packaged and shipped to the customers.
Lean principle: Te third K in OSKKK, Kaizening the Layout, has the
following sub-steps:
Ensure previous steps of OSKKK have been worked on before rear-
ranging layout (there is no point in relocating wasteful processes).
Collect data: process fows, capabilities, new products, bottlenecks,
OEE (overall equipment efectiveness), value stream maps, informa-
tion fow, etc.
Compile a minimum of three layout proposals, all showing fow arrows
and complete correlation matrixes.*
Review new layouts to measure against all lean principles and remove
non-value-added layouts to justify costs.
* Refer to Made to Order Lean: Excelling in a High Mix, Low Volume Environment by Greg Lane for
additional information.
Work to improve the man/machine/materials ratio.
Consider new machinery only afer working to improve existing machines
and incorporating all learnings into specifcations for new machinery.
Simulate the new layout. either marking the foor (for existing layouts)
or with cardboard mock-ups of the equipment (for new processes).
From a visual standpoint, this was going to work fairly well as we could
walk down the middle of the U and clearly understand the status through
the continuously updated day-by-hour or planning boards placed on all
critical machines. Naturally, the U shape also lent itself to fexibility for
the operators to share work. We did not have a one-man-to-one-machine
mentality and as we frequently had longer cycle times in the range of ten
minutes to more than one hour, one man could load and deburr parts
from more than one machine simultaneously. Since the longer run times
were typically on the larger machines, we made more or less a mini U
cell out of those. Tis lef fexibility to share tasks, not only between the
larger machines but basically across the entire shop (the whole shop was
not that largea man could walk through all processes within about a
minute). We put the shop ofce basically in the center of the U, ensuring
it did not block the view (everyone in the room had visual contact with one
another). In the new layout, we planned no walls for the shop ofce; just
the programming area and common/critical measuring tools were located
there with everything out in the open. Tose of you thinking about noise
and distractions while programming because you would be surrounded
by loud machines, not to worry; a good pair of industrial headphones to
block noise and a big screen to work from would allow Jake to enter his
own little world of concentration.
8.5 sIMulAtIng the neW lAyout
Once we had settled upon our new layout, we marked out everything on
the foor at the new site. Tis was not only to help the equipment movers
to know where things went, it also allowed us to simulate and imagine the
new environment in actual scale instead of just reviewing it on paper. We
made a few small changes, mostly moving things a few inches one way or
the other. I have found that you can get away with just marking out the
foor to get a proportional feel for the existing processes; however, when
you are developing new processes, you must work in three dimensions to
simulate as much as possible. Tis is best done through cardboard mock-
ups. Naturally, you try to do this as close to scale as possible and then
simulate the actions and movements to take place within the processes.
Key point: On the foor of the area you will be moving an existing pro-
cess to, mark out the new layout (that was developed on paper) so that
you get a feel by simulating activities. If you are developing a layout for
a new product, you should simulate this with three-dimensional card-
board mock-ups (made to scale) before agreeing on the equipment to
purchase and the new layout. Remember: always consider the operators
work balance when making a new layout proposal and locate equipment
in work cells that will best balance the machines and operators time.
One other point was that all of the shop services (electrical, air, data
cables) in the new location came from overhead as is common or at least
desirable in new construction. (Te exception is when you have cranes or
other machines moving overhead that do not permit this; but, this was
not our case.) Tis leaves considerable fexibility for future layout changes,
especially if you have the ability to install some extra wire and air lines
that can be coiled up for the future. We preinstalled a few extra feet of
fexible air line at all the air drops, then coiled and attached it to the hard
piping. I also had the electrician put about ffeen to twenty feet of extra
wire at each electrical drop, coiling it and attaching it to the conduit. Te
goal was no conduit directly to the machines, and so far the local electri-
cal codes were allowing us to get away with it. Tis would not only leave
us some breathing room if we didnt get it right, it would also leave a lot of
options open for continuous improvement and expanding the business.
8.6 IntroducIng ProcedurAl
chAnges durIng the Move
Another great opportunity to take advantage of when moving to a new
location is to change any work practices or standardized work you have
identifed that necessitates improving. It might appear I was jump-
ing from Kaizening the Layoutthe fnal OSKKK stepback to
Standardizationthe second step. But remember: OSKKK follows an
endless loop of continuous improvement. Te frst day an existing team
member must drive to a new work location, he or she is expecting some
new experiences and is as open-minded as that individual will ever be.
I felt my team was relatively open to new ideas, and this point was con-
stantly driven home with the frequency of changes taking place over the
past year and a half. Tey were starting to understand what continuous
improvement meant, but I still wanted to take advantage of this oppor-
tunity to introduce some additional changes. I frst wrote down a few
possible changes to be introduced and then invited Tom to brainstorm
with me. Te list included ideas like everyone wearing new shop aprons
that contained pockets for standard tooling (e.g., deburring tools, Allen
wrenches, dimensional pocket reference guides) and anything else that
the operators did not like to share or leave at the workstations but required
continuously. Providing everyone with all necessary tooling also helped to
reinforce the idea that all team members were responsible for deburring
and other tasks. Marking the foor for items like current job and mate-
rial for next job was helpful as well as introducing a lot of workplace
organization (5S).
When moving to a new location, it is also a good time to introduce change
within the procedures; in other words, any new work standards should be
incorporated as part of the move. One of the critical wastes I continued to
observe (even afer Tom was in the team leader position) that necessitated
a change was the operators were still walking away from their machines
to perform non-cyclical tasks like emptying chip bins, tumbling material
whenever they felt they had sufcient parts to fll a tumbler, or perform-
ing random measurement checks that required centralized instruments in
the shops ofce, etc. If the cycle times were long enough that they could
do this without the machine stopping and waiting, I saw no problem. But,
if there was no way to reduce or improve the time the task required, then
I did not want the machine waiting; instead, I wanted the team leader to
handle it. Remember: you always want things balanced, but it is better
to have the man waiting a few seconds for the machine than to have the
machine stopped and waiting for the man. If Tom was too busy, then Jake
could handle it if he was not in the middle of programming. As we started
out in our new shop, I wanted to reinforce that we were all there to keep
the machines running and every minute we were not cutting chips and
making good parts was time (and money) lost. Again, this is measured by
OEE. Tere were a few other procedural changes I introduced, but those
discussed here were representative of the critical ones.
Key point: A great time to introduce changes to processes or to introduce
new processes you want implemented is when making layout changes. It
is the perfect time to start workplace organization (5S), and people are
more accepting of changes at this point.
8.7 usIng A MAJor event to Introduce
chAnge In the offIce
Tis concept of introducing new procedures during a transformation or
big event is not unique to the shop foor. It also applies to the ofce. Te big
change for our ofce regarding the layout was that we were going to insti-
tute an open ofce. I would now be located closest to the door to the shop
foor (as I was back and forth most frequently). Te ofce assistant would
be next in line with access to the shop foor as she was frequently organiz-
ing the daily shipping and performing some part-marking and packaging.
Sandra (the ofce manager) was placed closest to the front entrance and
furthest from the shop as she spent most of her time handling the extreme
level of administrative processes that are associated with a small business
and she also dealt with visitors. We also looked at where various ofce
hardware (printers, faxes, copy machines) were located, based on who
used them more frequently and where they were less distracting. Just as
with the shop-foor layout, we also ensured everything was networked and
that all new information was now being put on the network drive. Some
ofce visualization boards were also installed, so available machine capac-
ity (updated daily) was visible when speaking to customers. Other boards
included a priority board and a vacation planning board. We also had a
place for our skills matrix, which showed the current cross-training sta-
tus and our plan going forward. Many of these visual improvements were
duplicated in the shop ofce (for the shop personnel), although without
walls it was a bit more difcult to fnd a location to hang these boards.
Lean principle: Lean office layouts are a consideration of the rela-
tionship between three factors: the f low of work through various
departments, the employees, and the equipment. The primary con-
sideration should be given to workf low and achieving straight lines
for the most frequent processes without backtracking or crisscrossing.
This can be further linked to aligning employees (or departments)
with the value streams they support. An example from manufactur-
ing is placing the responsible supervisor, purchaser, planner, and
engineer for a particular family of products together in a small office
in the middle of the production area; it will be hard to point fingers if
everyone sits together resolving the daily issues.
Overall, our move went smoothly, and afer a hectic weekend we were up
and running Monday with everyone in the new location. Because we kept
all our existing phone numbers we had not afected our customers with a
lapse in communication or on-time delivery issues. Naturally, we sent out
a notice about our new location beforehand as we did not want to catch
anyone of guard as it might appear we were doing something detrimen-
tal to our customers. Later, we could work through changing the mailing
addresses with all customers and suppliers, but for the time being we were
focusing on the orders at hand and maintaining our near-perfect on-time
delivery record.
Refecting on the move, I think it was successful because we worked
hard planning and preparing many weeks beforehand. We also talked
it up as a positive and made sure to clearly explain the purpose and
advantages behind the changes being introduced. I also openly brought
up the perceived disadvantages; for example, now the operators would
be spending more time in front of their machines instead of going to
the tumblers and measuring areas (as this is frequently viewed as job
enhancement by having a wider range of responsibilities). I always
spoke openly and worked to clarify how the advantages outweighed
the disadvantages and that all improvements helped with job security,
not guaranteeing it but keeping us ahead of the competition. I have
witnessed a few other organizations very chaotic relocations. I think
traditionally the majority of the people are so involved with their day-
to-day tasks prior to a move that by the time they sat to think through
the implications, it was too late to properly plan and there was only
time to react.
Now I really felt like the companys owner. Te new location was cre-
ated by me and my team, and little of Bobs presence was felt there. I just
needed to expand into new markets and survive the latest slowdown in
the semiconductor equipment industry. In this case, the move had served
many purposes. It had not only reduced our fxed overhead (by reduc-
ing our monthly rent), it also allowed us to work in an improved layout
that was sure to save money each day with less movement of people and
material. We were also closer to our supply base and could now shorten
our lead times for outsourced services like plating and painting. I could
more easily visually manage the shop and, if we expanded in the future,
we could also grow into the adjacent storage area. And the air conditioned
ofce was a critical factor and a productivity improvement for those work-
ing there (afer all, temperatures in California are frequently warm).
8.8 suMMAry
Although the OSKKK strategy relegates Kaizening the layout to the
fnal step because of the cost implications, there are times relocation
is preempted for economic or market reasons, determining a diferent
sequence in an implementation strategy. Since OSKKK is a continu-
ous improvement loop, you will be returning to work on the other
Kaizen steps as you follow the endless cycle through this strategy.
Flow is a critical element in any shop-foor or ofce layout and all
layout diagrams should include fow arrows.
Simulationeither through marking the foor when relocating exist-
ing processes or mocking up scaled cardboard three-dimensional
machines/processes for new layoutsshould be completed to allow
everyone to visualize and simulate the new layout.
127
9
Accurate Pricing through
Better Cost Allocations
9.1 IntroductIon
I wouldnt say that every detail had been worked through and resolved
with regard to our move to the new location, but we had focused on the
major problems and, for the most part, things were now running smoothly.
I was proud of the team and happy with our new location. We were in an
industrial area with a strong presence of machining and fabrication shops,
which accounted for the large localized supply base. We now had a plating
shop next door, a painting company across the street, and a machine tool-
ing supplier within a block. I worked out a simple pricing structure with
the plating shop, which allowed us to determine the plating costs we would
incur and to include them directly into our quote instead of going through
the internal efort with each RFQ (request for quotation), which specifed
plating, thereby necessitating subcontracting. So, we avoided the delay and
administrative costs of sending out an RFQ to the plating company and
receiving a price, then rolling it up in our quotation. Tis further reduced
non-value-added work in the administrative area; better yet, it only took
about thirty seconds to transport the parts to the plating shop. Tey prom-
ised us priority if we utilized them for most of our volume, further reduc-
ing our lead time and administrative costs to organize transportation.
Since it was a slow sales period, we were coping in the ofce and shop
without serious issues. But, like many businesses experiencing slow times,
we were only able to shed some variable costs. Our fxed (mostly overhead
costs) remained about the same, so our proftability had been low for a
few months. As is customary in slow times, not only were we receiving
fewer RFQs but our customers were feeling more price-sensitive and their
purchasing departments had more time to shop around. I felt confdent
that our material estimates and quoted labor times were very accurate as
we still tracked all labor and material against every quote and on Fridays
we reviewed those that were in excess of 15 percent. I had methodically
kept this process going as well as keeping myself in practice by completing
quotes and reviewing them. Although Jake (the programmer/estimator)
was handling about 85 percent to 90 percent of the quotes, I randomly
jumped in to keep in practice and receive feedback. However, one cost
factor with which I was not comfortable was how we charged those min-
utes and hours of laborconverting them from time into dollarswith
respect to our indirect labor and overhead costs, which had been incorpo-
rated into our hourly shop rate.
Previously, I have mentioned that the logic behind OSKKK (Observe,
Layout) also applies to ofce processes. You must frst observe the current
situation and then work to standardize it so you have a basis from which
to make improvements. In administrative processes, the key improvement
comes from the frst K, the Kaizening of fow and process, which most
ofen involves improvements based around computer systems or programs
that dictate the fow of the process. Whether companies are using a larger,
more powerful sofware package like SAP or Oracle, or a smaller one, many
improvements are related to data input, output, or some type of change to
the application. Most administrative Kaizen takes place within the processes
(or related sofware). Some companies utilize the second K of Kaizening
the equipments productivity. Te third K, Kaizening the layout, is where
many additional improvements are discovered but a more signifcant cost is
incurred. For example, having the planner, buyer, supervisor, and possibly
the engineer for a value stream sitting beside one another ofen minimizes
a lot of fnger-pointing and saves time in communicating. Now, the issue I
had identifed for my business was to Kaizen the process related to how we
charged customers for our indirect labor and overhead costs.
I have met managers who feel that in reality they do not need very accu-
rate product cost information because they cannot choose the jobs; they
can only do what is necessary to retain the customer. I fnd this thinking
AccuratePricingthroughBetterCostAllocations 129
incorrect because the customer ofen chooses what jobs you will be ofered
or allowed to quote, and if you are only taking jobs to keep the customer
happy without comparing the prices to your actual incurred costs, you are
walking a dangerous line.
9.2 deterMInIng More AccurAte
AllocAtIons durIng tough tIMes
I indicated that when I took over the business, the former owner, Bob, had
been using a straight $60-per-hour costing for all quoted labor. He felt
this covered all overhead costs as long as sales reached a minimum point
every month. I had adopted this rate. But earlier I began to break it down
into some various other rates based mostly on the type of material we were
machining, or adding an additional mark-up for some customers, though
I had done this randomly and haphazardly. In other words, I had done a
little observing and started changing the process without a lot of PDCA
(Plan, Do, Check, Act).
During this industry slowdown, not only were we quoting and selling
less but our average hit rate on receiving orders from our quotes (which
previously hovered near 70 percent) had dropped to less than 50 percent.
Naturally, we could drop our labor rates to keep people busy and attempt
to cover our overhead but that seemed a hit-or-miss strategy and was not
refective of the indirect work that went into the various orders.
Being a lean guy, I had learned and attempted to apply some lean
accounting in my consulting work. One methodology I had success with
was a simplifed version of activity-based costing (ABC). To me, ABC
helps to allocate as many indirect and overhead costs as possible based on
the level of activity they consume (not necessarily getting it exact but it
becomes a more precise cost). I fgured our material costs were competi-
tive (we ofen received two or three RFQs for material, and according to all
indicators we received a great price per pound) and our direct labor times
were accurate. Plus, we had made a lot of improvements in setup-time
reductions, we always programmed aggressively, and we used the newest
technology in tooling to keep our cycle times at a minimum. Terefore,
the only way I fgured our competition could beat us on price was either
with lower overhead costs or mistakes in their quoting. Te only other
diferences I could see in our costs were my loan payments and my lack of
experience, which I ruled out as a factor because Tom (the shop foreman),
Jake, and the other operators had signifcant industry experience. I knew
that some of my competitors, who were also approved suppliers at cus-
tomers like Applied Materials and LAM Research, were larger shops with
considerably higher overheads. So, how could they be underbidding me?
Lean principle: Lean accounting works to provide accurate, timely, and
easy-to-understand information that supports lean activities. It helps
with making correct choices regarding growth, proftability, pricing, and
where to focus improvements. It will meet generally accepted accounting
practices and is more understandable and directly applicable to mak-
ing pricing and investment decisions while justifying lean ideologies like
growing the employees skill sets.
9.3 usIng My ModIfIed ActIvIty-bAsed costIng
I was convinced that a great share of the diference in quoted prices lay in
my competitors various rates and how they allocated costs. So, as much
as possible, I wanted to get my rates to refect the activity level (i.e.,
the real ofce and indirect shop costs incurred) for each product, which
would directly address many indirect costs andhopefullywould bet-
ter help proportion our overhead. I could not aford to continue to turn
only 50 percent of quotes into orders, especially with as few RFQs as we
were receiving in those days. But, on the other hand, I could not aford
to lose money by making low-ball quotes and not understanding what
my actual costs were on a particular job. Having previously evolved a
few diferent rates based on the type of material we were machining, I
felt I had learned something and could spread that concept into other
areas, especially the administrative functions, where a large part of my
overhead resided. Job shops like mine naturally had much higher over-
head when compared proportionally to high-volume manufacturers. For
example, we had much more indirect activity associated with each part
we built (therefore, we had higher indirect costs that were being allo-
cated arbitrarily).
Te question was how to more accurately assign these costs to the vari-
ous products that consumed them. Here again, I thought I would start
with the direct and indirect shop activities as they related to the various
parts, and then tackle the ofce activities, which were all considered indi-
rect. I did not want to create too complicated a system and fgured I would
not be able to come up with a way to distribute all costs to specifc parts.
But, I wanted to reduce the large amount that had simply been generalized
into the one rate we had been using. I liked the ABC accounting concept of
creating some general categories and not attempting to be exact because I
did not have the time or ability to get the accuracy to two decimal places.
I would more accurately allocate based on the customers and parts char-
acteristics, which consumed time and resources in the ofce, and those
costs or activities I could not specifcally account for would be rolled into
these rates. I did not plan to make things too complicated and did not
necessarily advocate strictly following a detailed ABC approach. Instead,
I felt I could have one rate for the ofce and another rate for the shop. Te
rates would be based on the indirect activities and costs involved and the
sum of these two rates, along with the cost for direct labor, would make up
the total. Te total could then be summed with directly attributed costs
like the materials and outsourcing. Tis new rate would be inclusive of all
overheads that could not be based on the activity they consumed like rent,
loan repayments, etc.
Lean principle: My modifed activity-based costing (ABC) attempted
a realistic means of determining the real cost of providing a product or
service. It worked to accurately allocate some direct and indirect labor
costs and overhead to the particular products/services or customers
that consume these activities. Te general methodology encourages
grouping products or services by categories that afect their required
activity level and then assigning weights or costs to these categories,
creating accuracy but not precision.
For each area (both the shop and ofce), I envisioned a separate type of
matrix that would take into account the principal indirect activity levels
and costs incurred for the necessary tasks to produce and ship a certain
type or family of parts. For example, in the shop the activity level involved
(the persons or machines time that was consumed) was infuenced by the
material from which the part was to be manufactured and also the spe-
cifc tolerances. Te operators necessary skill level, along with checking
activities and the level of programming complication, also dictated how
much time was required for estimating. Trying to categorize everything
could get very involved. But remember: Te modifed ABC strategy I
was attempting to implement encouraged working toward being more
accurate than I had previously been but not exact down to the minute (or
cent). I wanted to be able to account for the majority of efort involved
and to be specifc about the true costs involved. In this way, I felt I could
increase our hit rate on the quotes during the slow times, and during the
busy times it would still help in better qualifying my true proft margin
per specifc part number (and customer).
Working to improve our pricing by more accurately distributing indi-
rect costs is obviously a Kaizen to the process, the frst K in the OSKKK
strategy. Te reason this became a priority at this point was based on busi-
ness needs caused by the industrywide slowdown and the necessity to
quote more competitively. I had previously improved some other admin-
istrative processes, then jumped to Kaizening the ofce layout (the third
K), and had now come full circle to further Kaizening administrative
processes. Keep in mind that the sequence of OSKKK is based on logic and
cost of implementations versus the benefts, but it cannot take into account
unique changes in the business environment that dictate the opportune
timing of certain improvements.
9.4 stArtIng ModIfIed ActIvIty-bAsed
costIng In the shoP usIng A MAtrIx
It is probably easiest to frst discuss the matrix I made for the shop, and
then I will explain the one for the ofce. I will reference the frst version
of the matrix, though it naturally lent itself to a continuous improvement
process and was still being evolved when I sold the business. A copy of
the original matrix is shown in Appendix J, Part I. I simplifed the two
axes of the matrix, the frst refecting all direct activities infuenced by
the type of part (attempting to account for additional activity not already
accounted for as direct labor for machine or setup times). Te other axis
further subdivides these categories by the indirect shop tasks likely having
an infuence and consuming time for a particular part (measuring/pro-
gramming). I am sure you can imagine all the infuences in a business and
envision this going into three dimensions or more, but I kept it to a simple
two-dimensional matrix and accounted for diferences by further segre-
gating into subcategories. Starting with the shop infuences that were most
strongly afected by the activity level required for the particular part, there
was the material (the additional machining time for tougher materials was
already factored in but this material also consumed more tooling, which
needed to be accounted for with a higher rate). Tis was further infuenced
by the amount of deburring involved as some could be done within cycle
while the machine was running versus signifcant deburring that would
ofen require additional time outside of the machines cycle. One more
factor was the tolerance on the part, which afected whether additional
checking and adjusting time would be required. Naturally, this additional
measuring and adjusting time frequently involved Tom or Jake on the
more complicated parts. And the tighter tolerances or excessive features
with critical dimensions equated to additional time for the operators as
well as the programmer and team leader, thereby increasing the cost. All
this additional work also infuenced whether one man needed to be dedi-
cated to a machine or he could instead operate more than one machine
simultaneously. Te number of operators required has a direct cost cor-
relation and the point system developed (within the matrices) helped to
refect this. Te idea behind the matrix was simple: the more difcult the
task, the more time it consumed; therefore, it received a higher point value
(shown as +0, +1, or +2 in the spreadsheet in Appendix J).
On the other axis, I put the various categories for the indirect support tak-
ing place in the shop. Again, the various categories and subcategories were
based on the time and efort required from the indirects in the shop, which
included Tom and/or Jake and me. Tese were then also categorized into
levels of +1 to +3 to account for the time or activity involved so that a total
combined efort level from direct and indirect shop efort could be summed
up in the matrix. Consequently, the higher the cumulative total of points
from both axes, the higher the activity level required for the part; therefore,
the higher the associated cost. In theory, we had already accounted for the
direct persons time on the machine through the machine cycle time and
setup time, but we were not always factoring in the additional direct and
indirect activities and time required for the particular part.
Te items afecting how the indirect people in the shop spent their time
were dictated primarily by how many measurements were involved (more
measurements generally equated to more difcult parts), more compli-
cated parts generally required support to help in determining how to
measure them. Te next factor afecting the shops support staf is how
difcult the program was to write and debug. Obviously, simple programs
usually worked immediately and, at the other end of the spectrum, the dif-
fcult CNC (Computer Numerical Control) programs required more time
to develop and debug. Naturally in some cases to quantify which row or
column you utilized was subjective (i.e. easy or difcult customers), but it
was a starting point.
In summary, the following items were identifed as afecting the activity
level in the shop:
Ease or difculty of measurements
Ease or difculty in writing CNC computer programs
Type of material (sof or hard)
Ease or difculty in deburring (cleaning) the part
Quantity of difcult tolerances to maintain
9.5 hoW to relAte the MAtrIx to cost
Te principle in the matrix is simple: the higher the cumulative numeric
value, the more efort being required from both the direct and indirect
shop personnel; therefore, the customer should be charged more for this
higher activity level. Tat is not to say improvements could not be made
to reduce the efort level or time involved but, at this point in time, it was
a refection of reality (part of the goal of any type of ABC is to help in
identifying areas or tasks that require improvement). Logically, a lower
numeric value would be linked to a lower cost for the customer and a
higher number to a higher cost. Before explaining how I arrived at link-
ing this to fnancial numbers, let me clarify what I did with the related
level of ofce-staf activities.
9.6 My ModIfIed ActIvIty-bAsed
costIng MAtrIx for the offIce
I followed the same principles in the ofce. I created a separate matrix to
refect the variables that infuenced the activity level required in the ofce
based on the part and the customer. Across the top, I placed Infuence
from the Customers (i.e., the difcult vs. easy customers) as this afected
the time involved, and it was further impacted by the level of packaging and
the method of shipping. Te original ofce matrix is shown in Appendix
J, Part II. Te other axis dealt with the major factors afecting the activity
level in the front ofce. For example, any changes to the specifcations (or
having to resolve unclear issues with the customers engineers) had a large
efect on the time required from the front ofce (this was more common
with certain customers and part types). Also afecting the time was the
number of items on the bill of material for a particular drawing; natu-
rally, the more items, the more time it took to purchase and receive those
items. (Historically, we were progressing into performing more assembly
work, and this was having a direct efect on administrative time.) Te
last issue afecting the ofces efort level was the amount of outsourcing.
Each item being sent out required obtaining a quotation, shipping, receiv-
ing, tracking, and a payment. Te same logic was applied in the matrix;
a higher numeric value represented more efort being required. Te com-
bined ofce scale ranged from a total efort level of four to nine, while the
shop had more factors infuencing it; therefore, the scale had a larger range
from four two twelve.
In summary, the following items were identifed as afecting the activity
level in the ofce:
Level of customer support required (changes to orders or specs, sales
support, standard specifcations)
Number of items on the bill of material
Number of services requiring outsourcing
Customer payment (on-time, good terms and conditions)
Level of packaging requirements
Now, this arbitrary scale had to be converted to a dollar value for a cus-
tomers quote. Tese two matrices in no way took into account all of our
overhead costs or all the tasks necessary to operate the businessthey only
helped to relate more accurately some major infuences on how our direct
and indirect time was spent in relation to processing various jobs according
to the characteristics that most directly consumed our time. (Tough the
rate I was to utilize had to account for all the fxed overhead costs that were
not accounted for in the direct and indirect activities within the matrices,
those would be more arbitrarily rolled in). In terms of ofcial activity-
based costing, this was not a complete analysis or plan for how to reallocate
all of our costs based on their activity level, but I knew it was better than
simply utilizing one rate or the few general labor rates I had previously put
in place, therefore I refer to it as my modifed activity-based costing.
9.7 relAtIng the MAtrIces to hourly costs
Based on the typical hours of work we would usually sell, I knew that in
slower months a rate of $60 per hour had traditionally covered most of our
costs, including the various overhead charges. Terefore, I wanted to use
that as a basis and somehow distribute it more accurately, based on the direct
and indirect activity levels I had determined in the matrices (previously I
had only been trying to account for direct labor and lef both indirect and
fxed costs arbitrarily distributed). I did not see how it was possible to fairly
distribute many of the other overhead costs (though this would likely be
pursued in full-scale activity-based accounting). Historically, the $60-per-
hour rate had satisfed the overhead costs and I felt this was the best I could
do at that point in time. I knew the costs incurred for direct labor including
benefts averaged about $20 per hour, so I could assume the other $40 per
hour ($60/hour shop rate $20/hour direct labor cost) covered all indirect
and other overhead costs. Tis was the proportional part I needed to more
accurately distribute. I had determined that about 60 percent of our overall
costs (excluding direct material and direct labor) resided in manufacturing
and the other 40 percent were in the ofce (salaries, overheads, and other
directly assignable ofce costs). Terefore, I decided to split the remaining
$40 per hour rate, attributing 60 percent ($40 0.6), or $24 per hour, to
shop-foor activities and 40 percent, or $16 per hour, to ofce tasks. Ten I
linked this fgure to the numeric scale in the shop and ofce, respectively.
I linked the $24 to the shop matrixs middle value (#8) of the scale and felt
that increasing or decreasing $2 for each point was relatively representa-
tive of the costs we incurred. Tis is shown at the bottom of the chart in
Appendix J, Part I. Naturally, this method of assigning costs was still based
on the labor hours associated with the job but since we knew (through our
continuous feedback system) that these were accurate, it seemed the best
baseline to utilize at that point.
Te remaining 40 percent of our costs (including administrative labor
costs) were assigned to the ofce matrix. In this case, the remaining $16
per hour ($40/hour 40% of distributed costs) was assigned to cover all
administrative costs other than direct materials; it was also evenly dis-
tributed on the ofce scale. Te middle point (#6) was therefore linked to
the $16 per hour, again with an increase or decrease of $2 per hour being
assigned to the efort levels above or below #6, respectively. When I totaled
both the shop and ofce matrices and combined this with the $20-per-
hour direct labor rate, I arrived at a minimum hourly rate of $48 per hour
($12/hour from the ofce + $16/hour from the shop + $20/hour for direct
labor) assigned to the parts requiring the least activity from direct and
indirect labor or a maximum of $74 per hour ($22/hour from the ofce +
$32/hour from the shop + $20/hour for direct labor) for jobs consuming
the most indirect efort from both the ofce and shop. Tis was quite a dif-
ference from a blanket rate of $60 per hour for all jobs but was much closer
aligned to our true activity levels.
Naturally, this could be further developed and refned to include con-
siderably more variables but I felt it was a good starting point. It was more
refective of the actual costs we incurred with each job and I hoped this
would help us in winning more work. I was not sure how our competition
determined their hourly rates or how sophisticated they were in allocating
all their costs within their quoting processes but I fgured we were better
of than before. Even though we could now easily associate a variable labor
rate with each job afer we had calculated the actual setup and run times,
we still had to adjust this for the customer and the current market condi-
tions. Te important point is that we now had a more accurate cost fg-
ure that refected reality in allocating our indirect and overhead costs. So,
when we were confronted with a choice about how we should price to win
a job, we could more accurately determine our proft margin and make a
better pricing or repricing decision. Tat by itself would allow us to com-
pete better. Pricing jobs with an hourly rate from $48 per hour to $74 per
hour was a considerable diference that we had not previously accounted
for. And I was sure it had cost us some business at the low end of the scale
when we applied our $60-per-hour general rate and we probably had a
poor margin or lost money when we won business for which we really
incurred costs of nearly $74 per hour. Now, we might not receive as much
work for the difcult jobs but we were likely to make a more stable margin
on the work we did win. Time would tell if we were on the right track.
To summarize what had been done to this point: I discovered an easy
way to better account and relate our pricing based on the activity levels
of our indirect labor that various types of parts consumed. Tis method
also helped in better proportioning some of our overhead costs. I still
had not determined a better way to allocate some of the fxed costs to a
particular part number, but traditional ABC accounting does ofer some
suggestions in this area.
Key point: Indirect costs can be proportionately better allocated to the
products that consume indirect time based on the activity level, as dem-
onstrated using a simple matrix.
In the end, the market determines the price based on the various suppli-
ers quotations (this will be infuenced by which suppliers the customer is
aligned with, the type of part, the number of competitors, etc.). Difcult-
to-manufacture parts and rush jobs frequently commanded a higher price
because they demanded a higher level of activity, and certain customers
were less price-sensitive (though all still demanded superior quality, and
some were pickier than others regarding subjective issues like surface
quality or cleanliness). In any case, I now better understood what my true
costs were, based on how the various parts and customers consumed my
teams time.
9.8 hoW others often AllocAte costs
In my consulting work, I fnd most companies either content with apply-
ing traditional accounting allocations or uncomfortable challenging
them; therefore, they easily accept the proft margins based on which-
ever methods the accountants choose to utilize for allocating the various
indirect and overhead costs. Tis is likely caused by a variety of factors,
but I frmly believe most manufacturing managers and executives feel
uncomfortable with the idea of challenging the accounting methods
(unless they are accountants themselves). I fnd they might ask their
controller to look into lean accounting but do not fully understand the
implications within their current system. Another factor I feel plays a
major role is the high cost of materials in many businesses, which creates
a feeling of keeping the focus on material costs and not worrying about
the accuracy of direct and indirect costs. I ofen fnd manufacturers with
whom I work have material costs near 50 percent of total cost (or the sales
price) and periodically it exceeds the 50-percent mark. In my case, mate-
rial averaged 18 percent of sales so we had a high value-added portion
and it was critical to quote this correctly.
Direct labor today ofen accounts for only 7 to 10 percent of sales value in
some manufacturing industries. Terefore, managers ofen consider this
of lower importance in relation to how critical it is to estimate and assign
it correctly, though the problem is that many accounting systems allocate
most or all of the indirect and overhead costs based on these direct labor
hours (as I was doing). Remember: the total indirect and overhead costs
frequently account for nearly 30 percent of the sales price. So, if you are
allocating those costs based on poor direct labor hours and not assigning
them based on time (or activity), you can end up misallocating a large
percentage (30% + 7%) of your costs, losing considerable business based
on incorrect pricing estimates. I have demonstrated simple methods to
more accurately account for the indirect time, though the fxed costs are
considerably more complicated to correctly allocate.
Not being an experienced accountant myself, I am sure many could
poke holes in my methodology; the truly analytical person would say it
was oversimplifed but I fgured it was a better refection of reality than
what was previously in place. And later I was able to further learn and
improve this methodology. In a way, there were some easy measurements
to determine the success of my new pricing strategy; I would immediately
start to notice if my hit rate on orders received climbed above 50 percent
for the jobs I was quoting, then I could also compare my end-of-month
proftability to previous months. Naturally, I was not just hoping to receive
more work in a troubled and competitive market; I was really hoping to
improve my proftability, the number one goal of every business. I had
to factor in the slow market and lower sales because obviously I received
more overall proft in very busy months. But, at least I could look at the
recent proft compared to the new proft as I introduced this new labor
scale ranging from $48 per hour to $74 per hour. I felt good about realizing
the pitfalls in an oversimplifed costing system and felt I was making step-
by-step improvements. Time would tell.
9.9 keePIng the teAM leAder focused
durIng the sloW tIMes
Naturally, during the slow times we were not working overtime. However,
Tom was still coming in a little early to prepare the machines, materials,
and any paperwork so all the operators could start on time. Even during
a slow period, it was better to pay a little overtime to one man and avoid
having many others get a slow start. Tese were also ideal times for cross-
training, preventive maintenance, and other items we had put of during
the busy times. As I mentioned earlier, I even had a cross-training matrix
showing our current status and our training plan. So, it was easy when
someone had a little time available: First, they checked the cross-training
list to see whether any of the tasks for which they required experience
were currently taking place. If not, they checked a things to do list I kept
updated to keep people productive and put improvements in place. (An
example of the cross-training matrix is shown in Appendix H.)
I have seen many plants where a things to do list does not exist. So,
when things slow down, the operators also slow downadopt a slower
pacebecause they fear that they could be laid of or lose their job if
they do not appear busy. Tough my tracking system in which operators
recorded their actual times compared to quoted times prevented them
from slowing their pace, I fgured it was better to list the things we needed
to do so that, instead of slowing down the pace to look busy, operators
understood there was work waiting. I wanted everyone to feel there was
always something to do that added value and prepared us for busy times.
Basically, it worked although it was by no means perfect. But, as I sup-
port other companies, I ofen fnd this list is not available (maybe it is in
someones head), at least it is not out in the open for all to reference. Ofen,
these nonlean plants seem too busy frefghting to take advantage of this
excess operator time when it becomes available. I learned this technique
from Toyota, which always organized a few areas/departments in the shop
(i.e., making new tips for the welding robots) where supervisors could
send operators who had no work. And then the area would not be respon-
sible to account for those labor hours. It is simply a matter of recogniz-
ing that it is hard to guarantee eight hours of work every day for every
employee without having lots of fexibility and management organization
to deal with the varying demands placed on peoples time.
Key point: Again, it is vital to have a cross-training matrix (with a devel-
opment plan for each employee) visually displayed and up to date, along
with a list of tasks (improvements, maintenance items, etc.) that need to
be completed so you can motivate operators to switch to value-adding
tasks during slow periods. Tis is critical for every company.
9.10 IncreAsIng Process cAPAbIlItIes
to reduce outsourcIng
We also did what most others do in tough timeswe tried to stop out-
sourcing any products we had the possibility to manufacture in-house.
For a year and a half, we had been trying to quote all drawings we received.
However, we ofen did not have the in-house capability to make some of
these parts, or in some cases we could manufacture them but not with the
most efcient and competitive equipment for that type of part (i.e., we only
had manual lathes, not CNC lathes, for making certain round parts). So,
we had subcontracted those to other shops when we received the order (we
would typically charge 15 percent to 20 percent above the subcontractors
price); but, in this price-sensitive market, we stretched our capabilities in
every way possible to manufacture most of these parts in-house.
I had begun leaving the business for short periodsapproximately one
week per monthto do some lean consulting work. I knew that with more
of these standards in place, the business would run better utilizing less
tribal knowledge and rely more on established processes even for difcult
tasks like quoting. As I mentioned, one of my goals was to get the business
stable and standardized enough for it to run on well-defned processes
instead of requiring an on-site master frefghtermeto help coordi-
nate the other frefghters.
9.11 stAndArdIzIng the busIness
Processes AlloWs tIMe for
PursuIng other oPPortunItIes
It was about this time that the dot-com boom was taking place, and living
in the San Francisco Bay area put me right in the middle of iteBay and
Yahoo had their headquarters only a few miles from my business. I had
many friends who were getting swept up in the dot-com boom and later
one or two made their fortunes from it before the bottom fell out.
I came up with a concept for a Web site and got a frienda high-tech
guruinterested in working with me in the development. Te idea was
based on none other than the machining and fabrication businesses.
During this period, it was becoming typical for customers to send their
technical drawings out for quotation via the Internet. In some cases, it
was better for machine shops to receive the AutoCAD fle directly as they
could import it and build their CNC program around its geometry instead
of redrawing the complicated parts in their own sofware. Te Web site
idea came from several directions:
Customers were sometimes not sure of the diferent services, qualif-
cations, and limitations of the various shops.
Customers also wanted to ensure their drawings would be kept con-
fdential and to know which suppliers held the correct certifcations
(i.e., AWS [American Welding Society], ISO 9001:2000, etc.).
Customers were usually unaware of other suppliers who could pro-
vide competitive quotes for various types of parts.
Many CNC and sheet metal companies did not have the capability
to open electronic drawings in the various formats in which they
arrived (e.g., AutoCAD, Mechanical Desk Top, Inventor).
Te opportunity that the Web site could provide for the supply side
the various machine and fabrication shopswas to bring customers and
suppliers together. My idea was to create a type of marketplace Web site
where customers could submit their drawings to all suppliersopen
quoteor to approved suppliers in an organized and secure for-
mat. Suppliers could receive the necessary information and the draw-
ings in a format they could access. Technical questions could be posted
anonymously to a forum or comment board visible to all those quoting
on the job and the customer would only have to answer each question
once. (Keep in mind this was quite a while ago, before any similar Web
sites existed.)
My friend and I built a prototype Web site to display the principles and
some typical menus. I started meeting with some of the purchasing people
at my larger customers to get their feedback on the idea. I would show
them how to log on to the Web site and how to send a few drawings out
for bid. It was going well, and we were busy taking in the feedback we had
received and making modifcations.
However, my bubble was burst when I met with a buyer who said she
had been visited by someone with a similar idea about a week prior to my
visit. She even gave me the mans name and the name of the company.
Dismayed, I ran back to the ofce where my buddy and I did some quick
research on the Internet. Between our research and the information a few
friends could gather through the grapevine, we discovered that the con-
cept had been in development about ten weeks and the company had just
received $6 million in venture capital and now had fve people engaged full
time. We were in no position to go further as we basically had no money
and both had full-time jobs. I guess the only consolation was that my idea
was at least worth $6 million to some venture capitalists; unfortunately, I
was not on the receiving end of that moneyalways a day late and dollar
short, the story of my life.
I guess the most applicable business lesson when developing a new prod-
uct or concept is to quickly determine whether there are competitors and
what they are working on. Tats probably taught in the frst year at busi-
ness school and anyone who didnt learn it academically would say its just
common sense. I will say, however, that at the time it was not easy to know
which dot-com sites were under development as everyone was keeping
their ideas secret, and I just stumbled across the fact that somebody was
working on a similar concept.
One saving grace might be that I have monitored the situation for years
and I have never seen that Web site nor heard of anyone in the industry
using it, which either means it was not such a great idea or it was not exe-
cuted correctly. Oh well, at least I tried to earn the title entrepreneur.
I have mentioned taking this diversion since it was made possible by
standardizing the business processes to the point that I felt comfortable
enough to temporarily put my focus elsewhere.
9.12 suMMAry
Lean accounting, specifcally activity-based costing, can help to bet-
ter allocate direct, indirect, and administrative costs to the specifc
product or service that is consuming that activity my modifed
version is one way to arrive at various labor rates to account for the
ranges of activity levels involved.
A matrix utilizing various subcategories can be helpful in grouping activ-
ity levels based on certain characteristics of the part and the customer.
Most companies function with lots of costs being arbitrarily allo-
cated, making the accuracy of any specifc products proft margin
very questionable. Although my version of activity-based costing is
by no means exact, it would likely be more accurate and allow better
decision making if properly applied.
145
10
Up-Front Delays
10.1 the MeAsureMents confIrM success
We seemed to be bringing in more work and the measurements indicated
it was at least partially due to our new pricing structure. Since we had
switched our hourly rates from a fat $60 per hour to a range of $48 per
hour to $74 per hour, our hit rate on quotes was averaging over 60 percent,
compared to averaging less than 50 percent previously. It was still slow
times in the industry we predominantly supported so the improved hit
rate was helping to increase our sales, thereby increasing our profts. It
also seemed our proft margin was stabilizing, which was another mea-
surement of our new pricing structure. Te fnal measure of success I uti-
lized was randomly looking at periods where we had a few high-labor-rate
jobs running (i.e., $74/hour), and monitoring how busy (or how stressed)
the ofce staf and/or indirects in the shop were, compared to normal.
Tis was basically a subjective measure and I was probably slightly preju-
diced, wanting to prove my point. But, it certainly appeared that when we
had a few more jobs with higher labor rates, the indirects (both shop and
ofce) were busier. So, the higher labor rates truly refected a higher activ-
ity level among indirects, thereby proving we were better cost-estimating
our indirect time.
10.2 contInuously develoPIng
Protocols AlloWs More AutonoMy
froM dAy-to-dAy oPerAtIons
Standardized systems like establishing criteria for using a variable labor rate
continued to put more structure in the company, not only requiring fewer
judgment calls and less tribal knowledge when performing various tasks but
also they increasingly allowed me to be removed from the day-to-day opera-
tions. As I mentioned in the previous chapter, I had started to consult, which
kept me out of the ofce about one week per month and forced the others
to manage all aspects of the business (though I ofen had my mother watch-
ing the money a little more closely during those periods). You know that
old saying about a boss only being as good as the people under him or her?
Well, I was trying to evolve this, not only through minimizing good people
running around with heads full of tribal knowledge but further by putting
standard processes in place that drove results while not completely relying
on knowledge, experience, and/or personal judgment. In other words, mak-
ing an attempt to have an established protocol for most situations.
One of the wastes I was continuously working to eliminate in my business
was the delays, the things that kept us from cutting chips on a machine. It
may sound simple: Afer we got the order, we bought the material and any
special tooling; then, we wrote the CNC (Computer Numerical Control)
program for the machine. And when the job came up in the schedule,
we manufactured the part. Still, there were many reasons why this might
fall apart, and a good place to focus was where problems and delays were
encountered in the up-front processes. For example, imagine if we had the
ability to foresee problemsestimating problems, purchasing problems,
problems with the drawings or technical specifcations, problems with
quality specifcations, problems with the program, tooling problems, etc.
and resolve these issues before it was too late. If we could do that, we would
make large improvements on the lead time. Life would have been much
easier if I had always been able to provide the actual value-adding process
with everythinggood quality material, drawings or specifcations with no
issues, clear quality standards, an achievable schedule, a qualifed employee,
and fully functioning machines and toolingon time. Well, dream onor
maybe not. With standard parts, those we had been manufacturing for
some time (runners or A type parts), we were able to eliminate most
Up-FrontDelays 147
of these issues. We were able to minimize many of these disturbances for
most runner parts when repeat orders were received. Tough since at least
40 percent of our work at any point in time was for a part we had not pre-
viously manufactured, we needed to work toward eliminating these prob-
lems on all jobs. I found better planning was essential: If we could arrive
at a point where wed actually be performing capacity planning instead of
chasing materials, many of these issues would be taken into consideration.
10.3 PlAnnIng only bAsed on
AvAIlAble MAterIAls Is costly
We had in place many of the traditional systems to avoid the classic prob-
lems; for example, the inspection of incoming materials, the writing of the
programs at the earliest possible time to help in identifying issues with
the drawings (missing dimensions or specifcations). And we had begun
visual capacity planning afer materials were available.
Te use of capacity planning is one critical diference between my busi-
ness and many of those that I support. Some companies are so stressed
with chasing late materials that when they all fnally arrive, the job is
quickly released to manufacturing without any capacity planning. Some
of you might even be familiar with jobs being released with only enough
material to complete the frst few operations steps and the hope that the
missing materials will arrive before they are needed for the later process
steps. Tis releasing of orders only based on material availability is obvi-
ously dangerous because the work order might arrive at an overloaded pro-
cess or a bottleneck. And if nobody in production had been made aware
of the constraint, they could not take any proactive steps to minimize the
bottleneck and maintain the lead time. Consequently, this newly released
order immediately increases the size and delay of an existing bottleneck.
Naturally, in these cases the lead time through the business becomes what-
ever time is necessary instead of a planned time and is usually dictated by
the bottleneck (basically following the theory of constraints).
Te ideal situation to work toward is to have a weekly overview capacity
plan in place for all bottleneck processes based on the materials antici-
pated arrival. Afer the materials have arrived, those orders should then be
released for detailed capacity planning not to exceed twenty-four hours at
a time. Te twenty-four hour limit cuts down on the amount of time you
will have wasted if you experience problems during the dayand we all
doand your plan is thrown of and needs revisions. When this happens,
youll only need to revise the plan for those unfnished jobs that had been
previously scheduled within the twenty-four hour period instead of some
longer periodsay, rescheduling an entire week. A rule of thumb that has
worked is to only post a detailed plan for a process one day in advance (in
a three-shif operation, plan three shifs ahead). I would like to capture
this as a key point because we were doing it and receiving a competitive
advantage, but I know many others who do not perform this planning for
a variety of reasons.
Lean principle: Te theory of constraints is a management approach
that focuses on the importance of managing constraints. A constraint
(also known as a bottleneck) is something that inhibits you from getting
more out of a process. All businesses have a minimum of one constraint.
Te constraint within the fow of your business is determined by the
process or step with the smallest capacity. Since lean entails increas-
ing capacity (aided by capacity planning), the theory of constraints is a
management methodology that compliments lean thinking.
Key point: To improve on-time delivery and process efectiveness, once
per week, prior to receiving all materials, a weekly overview capacity
plan should be developed for all bottleneck processes based on the mate-
rials anticipated arrival dates. Ten, afer the materials have arrived,
those orders should be released for detailed capacity planning (not to
exceed a 24-hour plan). Tis ensures you can anticipate and minimize
the efects from these bottleneck processes.
10.4 cAPAcIty PlAnnIng bAsed on leAd tIMes
We were essentially a JIT (just-in-time) business in the sense that we only
built when we had an order and we released the job to production primarily
based on taking the customers due date and subtracting the lead times for
internal and any subcontracted process steps planning that the customers
Up-FrontDelays 149
will receive it exactly on the promised date. Tis type of planning requires
that you know your specifc lead times and task times for each process and
ensure they are maintained or risk being late to the customer. In our case,
this also entailed managing the fow through external processes like heat-
treating, plating, painting, etc., where we maintained a close relationship
and followed up with our suppliers to ensure those lead times were achieved.
Te trick with lead times is to realize when they were starting to exceed that
which was originally planned and take action to minimize this. Te prob-
lems we encountered were typically at the second, third, and fourth process
steps, when work started to queue up at that process and create a bottleneck.
We needed to either anticipate this in our planning or identify it in real time
and become proactive. Troughout the years that I owned the company, we
maintained between 98- and 99-percent on-time delivery to the promise
date, but this required hard work, the ability to identify the bottleneck early,
and a lot of fexibility within the team (i.e., overtime, cross-training, running
processes simultaneously, etc.). Tis also occasionally required we negotiate
a promise date with our customer upon receipt of the order instead of just
agreeing to their request date. Our capacity planning was accomplished
through the day-by-hour boards, as shown in Appendix D.
Lean principle: Lead time is simply the time between instructing
something to begin (i.e., placing an order) and receiving the fnished
good from the process/supplier. It is a key focus for any company work-
ing to improve proftability. Tis is a focal point in value stream map-
ping, process mapping, and reducing inventory. Remember: there are
various lead times, like customer lead time and manufacturing lead
time, and various terms in lean like throughput time, total product
cycle time (used in value stream mapping, which is slightly diferent
than lead time). Make sure you understand the start point, stop point,
and contents of what you are working to improve when dealing with
any form of lead time.
Key point: To maintain on-time delivery and keep the fow moving, you
need to frst know your lead times (both internal and external), and then
plan accordingly. Further, you need to monitor where the lead times are
not being maintained and quickly treat those processes as bottlenecks
and take appropriate actions.
10.5 IdentIfyIng AreAs Where We
Were only beIng reActIve
We knew how to be proactive in some areas although I was continuously
looking to identify where we seemed to be only reactive. Naturally, this is
critical as only reacting without solving the problem at its root cause means
it is likely to repeat. However, since there are rarely sufcient resources
available to resolve many issues at their root cause, you must have a method
to select those that will ultimately lead to the most proftability upon their
resolution. One area where it was easy to become proactive instead of reac-
tive was with suppliers that were ofen late. Instead of accepting that a part
had not been delivered by the due date, we determined that any supplier
with a history of being late more than twice would be contacted basically
each day afer placing an order to see whether it was still progressing to
plan. It may sound crazy to call each day but remember: we were a short
lead-time business; therefore, we were frequently working with materials
that were due in a few days. Even for exotic or special materials, we were
frequently paying to expedite as our customers rarely ordered with long
lead times. (Naturally, this begs the question, Was the entire industry dis-
organized? But, it appeared that our customers customers who purchased
the machines to etch the computer chips all worked in a very dynamic
and quickly changing industry.) Tat is not to say that we never experi-
enced few weeks lead time for materials, but this was the exception and
in these instances we would only call a problem supplier onceone week
before it was due. Naturally, this does not always get a part in on time
but at least nobody could claim they had not received the order, and this
constant communication would make us aware it was going to be late and
everyoneincluding our customercould plan accordingly.
10.6 AvAIlAble tIMe utIlIzed to
docuMent Processes
Whenever the indirects had enough free time, I tried to have them write
out the steps (either a standardized operations sheet or a process map) for
Up-FrontDelays 151
any process we had not previously documented. Tis may sound extreme
but I found that people would rather be busy than look busy.
Remember: the ofce is an even-more-difcult environment to provide
eight hours of value-adding work for everyone each day he or she shows
up. I had shown everyone how to draw process maps (basically fowcharts)
and we only did it by hand; therefore, it was not too complicated. Tere is
always the possibility that employeesthinking about job securitywill
not document all the process steps or will leave out some shortcuts (or
improvements) they have discovered. But, some documentation is always
better than nothing.
Process mapping (like value stream mapping) is also a great help in iden-
tifying where standards are lacking or opportunities for improvement lie.
I noticed that these types of keep busy looking for improvement tasks
never got done when I was out of the ofce, and of course I knew it wasnt
because suddenly everyone was busy. I assumed it had more to do with me
continuously pushing people to add value whenever I observed otherwise.
I found you can push people toward their best productivity if they see
the boss always working hard and adding value. And you need to provide
them the knowledge and structure to identify non-value-adding tasks, and
to create a format that encourages them to make suggestions (an example
of mapping techniques is shown in Appendix M).
Lean principle: Value stream mapping and process mapping are great
facilitation tools to focus diferent perspectives through examination
of the processes based on data gathered at the place the work is per-
formed. Value stream mapping is used to analyze the fow of materials
and information (related to customer orders and purchasing of mate-
rials) currently required to bring a product or service to a consumer.
Tis is a higher-level map (not necessarily expanding each process into
detailed steps). Process mapping (sometimes referred to as VSMi) is a
fowcharting method that uses general symbols to show the fow of a
process, typically utilized in administrative or ofce processes. Tis is
used for mapping in greater detail and takes into account processes with
parallel steps and decision points (a graphic representation is shown in
Appendix M).
Any other recurring delays that we noticed in releasing jobs would be
tracked with a simple tick list on someones desk or up on one of the boards.
Afer a few weeks, I was either convinced it happened ofen enough to
merit looking at for a more permanent solution, or it was nothing to lose
sleep over, as I already had enough items both in my professional and pri-
vate life to keep me awake at night. So, I did not want to add items to the
worrying list that only occurred infrequently.
10.7 leAd tIMe reduced through
strAtegIc Inventory
Another area that was previously mentioned in helping to shorten lead
time and avoid delays was stocking a wider selection of raw materials.
Naturally, it was critical to have the correct type of material in stock,
though in cases where the stock material was too large, we had the ability
to cut it to size (which incurred additional cost but, if cutting to size short-
ened the lead time and helped secure the order, it can be justifed). On the
other hand, stocking too small a size of a material would never help as
there was no such thing as a material stretcher in our business. A bit of
experience and data helped in choosing which materials to stock and I did
not view it as excess inventory. Instead, I considered it strategic inven-
tory, meaning I not only had it there for a reason, I also regularly reviewed
and adjusted it to ensure I had the correct type and quantity on hand.
Instead of viewing all inventories with suspicion and considering them a
negative, I tried to separate what was on hand because of poor planning
versus what had been strategically acquired to improve the businesss prof-
itability. Tis is critical, and I would like to note it as a key point because I
ofen feel lean transformations focus on reducing all inventories instead of
separating out strategic inventory.
Key point: Any inventory that is in place to reduce lead time, improve
on-time delivery, and is regularly reviewed to ensure you have the cor-
rect material and quantity on hand should be considered strategic
inventorya lower priority for reduction. On the other hand, inven-
tory that is a result of poor planning or is not helping improve lead time
Up-FrontDelays 153
or on-time delivery should be a high priority for reduction. Remember:
inventory is necessary to compensate for the diference in the customers
lead time and the suppliers lead time. So, the focus should be lead-time
reduction, thereby reducing the need for inventory.
One other item that I kept more stock of than the previous owner was
commonly used machine tooling (i.e., end mills, taps, drills). It was easy to
determine which tooling had the highest consumption frequency and to
stock levels of those tools based on their lead time while considering how
various order quantities afected price, reducing administrative costs and
lost time because the operators did not have the tooling when required.
Tere was essentially no chance of this tooling becoming obsolete and
resulting in a write-of. I even considered systems ofered by tooling sup-
pliers where they stocked and replenished a toolbox for you at their own
expense and invoiced each week based on what you consumed. I deter-
mined this method would be more costly for us as I received generous dis-
counts on tooling by going directly to the wholesaler, although the lead
time was longer (which I was usually able to accommodate with my plan-
ning method). I was able to justify the cost of the higher inventory expenses.
Many companies with whom I have worked have found that a suppliers
automatic replenishment system actually results in savings over having the
process internally managed. Tat was not my case (as my administrative
costs were probably lower than that of other companies), but it is worth
looking into. Lean people typically call the system that sets levels of con-
sumables and automatically replenishes them consumable Kanban, and
I would recommend implementing it wherever possible afer determining
whether it is less expensive to manage internally or externally.
Lean principle: Consumable Kanban is setting up an automatic replen-
ishment system for consumable items (tooling, gloves, chemicals, pack-
aging material, small hardware, etc.) to reduce all the administrate
costs, carrying costs, obsolescence costs, etc. Again, Kanban is basically
determining minimum and maximum points that trigger an automatic
replenishment in the easiest fashion. Kanban is ideal for consumables
and materials that have regular usage, especially those that are fre-
quently ordered (meaning they incur high administrative costs) and/or
have a high inventory level. Remember, many companies experience a
high administrative cost to process each purchase order. Kanban reduces
this cost by using a few blanket purchase orders.
10.8 PrIorItIzIng ProbleMs durIng busy PerIods
Many of the refnements I have been speaking of all sound nice to imple-
ment if you have the time and resources. It probably appears I had things
running so smoothly that I came into the plant, got a cup of cofee, read
my email, and then went looking for improvements to make. I wish that
would have been the case; instead, I had days I still spent frefghting from
the moment I walked in the door or times when a phone call got me out
of bed early. Other times, I found myself either flling in for someone
who was not there or jumping in to support a seriously overloaded bottle-
neck process. Tose of you familiar with small businesses know that the
fewer employees you have, the more hats you are likely to wear. And these
small businesses are ofen fghting the same problems day in and day out
because everyone is too busy to resolve the real issues. Instead of randomly
and blindly making the time available to work on problems, I always tried
to only focus on repetitive problems and periodically I had to do this on
overtime. Of course, my personal overtime hours included no cost (just as
yours might be free of charge) but, during busy times, Tom (the shop fore-
man), Jake (the programmer/estimator), and some of the operators might
even be asked to stay to work through a problem that was plaguing us or
to implement something we could not fnd the time to handle on straight
time. I am still convinced it was better to pay time and a half than to lose
time reworking parts or being forced to wait because all the material, tool-
ing, or information was not available to perform a job.
10.9 the lAck of hAnds-on
exPerIence Is very costly
I indicated that we tried to write our CNC programs as early as possible,
ofen the same day that we received the order. Tis was critical because
ofen we were working on prototype or newly released parts; therefore, the
Up-FrontDelays 155
drawings were ofen missing information or contained errors. I would like
to say that most of these problems were caught during the quoting process.
But, during quoting you do not review the drawing with the detail that goes
into writing the programs necessary to manufacture the part. However,
what the quoting process ofen demonstrated was the lack of practical
training and experience the customers design engineers had. I feel com-
fortable making this statement because I am an engineer and had previ-
ously worked as a mechanical engineer doing design work. I was now on
the receiving end of the design work, though at this point in my career the
job at hand was trying to fgure out how to manufacture the parts. Many of
the mistakes I saw had to do with tolerances. For example, sometimes it was
as simple as a sheet metal or fabricated part that had been drawn within a
template for a machine part and the general tolerances specifed in the title
block were for a high-tolerance machined part, not for a simple sheet metal
cover panel. In many cases, the part could be manufactured to the tighter
tolerance (with some difculties) but it took more time, frequently it had
to be run on a better machine, it required more quality checks, and would
likely incur some scrap. All this equated to a much higher cost for myself
or for my competitors, most of which we would identify at the quoting
stage and pass on to our customers. Te other option would be to ques-
tion the tolerance. Te two predominant reasons this was not ofen done
were that the buyer did not want to go through the work of contacting the
engineer and requesting they change the tolerances, or they did not want
the anticipated delays that would come with any changes. We also knew
that their incoming inspectors would reject any parts not within toler-
ance (without questioning why such a tight tolerance was being applied
to a simple cover panel, for example). So, we had to choose between three
options: maintain these incorrectly specifed tight tolerances, get it cor-
rected, or not quote the part. Tese problems with tolerances are just one
example. However, most engineers feel that any feature that is marginally
critical should have a tighter tolerance called out than necessary and this
will ensure the manufacturer achieves close to what they need. But, they
ofen do not consider the cost implications or that the manufacturers gen-
erally achieve or exceed any tolerance specifed.
Some think these issues are covered by DFM (design for manufacture)
or DFA (design for assembly), and naturally these are valid process steps to
go through. But they normally do not catch problems or issues with fabri-
cation drawings relating to tolerances or missing dimensions. Tolerancing
problems are representative of issues that just drive the cost up but remain
buried in the design of the part. It is typically the subcontractors who
might notice this, but they realize pointing it out will cause delays in the
best of cases. Or, it might encourage overworked purchasing agents to buy
from a competitor who has submitted a price that includes this extra cost
instead of taking on the extra work and delays to save costs by getting a
tolerance changed. Te only time buyers called me back was when my
competitors and I had all no-quoted a part because of a difcult/impos-
sible tolerance or other callout in the drawing. At this point, they were
interested in resolving the problem. During my time in the business, we
probably had more than seventy-fve of these occurrences. And every time
that it escalated to the point where it was necessary for me to speak with
the engineer, he or she said it would be no problem to make the recom-
mended change. Ofen, the engineer currently responsible for the part was
not even aware of the reason it had been designed and drawn that way.
However, in all these instances only once or twice did engineers have the
time or ambition to change the original drawingthey would only com-
mit to faxing me a red line, in which they would note the change and
sign it. Tis meant the problem would come up again when either some-
one else built the part or the incoming inspector referred to the original
drawing. Tis is just one example of where the ofcial process of chang-
ing a drawing must be so long and painful that the workaround appears
easier. But, in the long run this problem is likely to bite you in the butt a
few more times. Tis is also an example of not actually solving a prob-
lem because you have not arrived at the root cause. Identifying the cost
of each occurrence and quantifying how ofen this happens would likely
lead the respective companies to mapping the current state and identi-
fying improvements. Tis issue was brought up again as it continued to
plague us and should again be highlighted as an improvement point for
companies that hold responsibility for design work.
Lean principle (from Chapter 1): DFM (design for manufacturing)
and DFA (design for assembly) are two lean ideas that work to bring the
designers and manufacturing team together during the design process
with the aim to improve manufacturing productivity. Keep in mind that
typically 75 percent or more of the products cost is determined during
the design phase.
Up-FrontDelays 157
10.10 My PersonAl exPerIence of
generAtIng costly desIgns
When I started as a design engineer fresh out of the university (actually
designing new manufacturing process equipment), I was put straight to
work as a design engineer for production equipment. Like in many com-
panies, new employees needed to be productive right away, and there
was no time for me to be further developed or to learn the business in a
hands-on fashion. Productivity was judged by how well you handled proj-
ects, mostly on time and to budget, though all budgets are relative to how
detailed they are and how stable are the infuences.
During this period, I got to know (socially) the main fabricator this par-
ticular manufacturing company used. Tis level of comfort with a supplier
is normally looked down upon but since I had no infuence over sourcing
decisions made by the purchasing department, I dont think it hurt much.
What I learned one night over a few beers with Rex (the owner of the fabri-
cation shop) was more practical and cost-efective than a lot of years expe-
rience without this feedback. He was joking about working long hours the
past week and then indicated that most of that time had been spent on
a piece of production equipment I had recently designed. He went on to
say that he was not too bothered because he had included most of these
hours in the quotation and factored it in the cost. I was a bit surprised
because I felt it was a fairly straightforward fabrication for a shop like his.
He indicated that the tight tolerance called out was consuming the major-
ity of these additional hours. Te next words out of my mouth were, What
tight tolerance? Afer he stopped laughing (I assume because he expected
this response from me), he asked whether I had not noticed that in all the
designs title blocks, a 0.005-inch tolerance was being called out and I had
specifed all dimensions to three decimal places. Being young and nave, I
answered that I had not even noticed the tolerance in the title block. And
Im sure he repeated this over the water cooler and it received a few hearty
laughs there as well. To put it in perspective for me, Rex went on to explain
that a single human hair was roughly 0.005 inches thickand that really
sunk in. I had not really ever thought of what that tolerance meant in prac-
tical terms and now I was sitting in the bar thinking that there was nothing
in the entire fabrication that needed to be accurate to within the thickness
of a human hair. I asked if I could revise the drawings and we could receive
some type of cost savings. Rex was a good guy and told me he would work
with me. Te next day, I went to my boss and explained what I had learned,
assuring him I would not make the same mistake twice, and he encouraged
me to do what I could to save money on that job.
Looking back, either Rex was just being a nice guy or his ulterior motive
for buying the beers was that maintaining those tolerances was really
starting to make the job difcult and possibly a money loser. By ofering a
price reduction in exchange for reasonable tolerances, he would likely not
incur further scrap and losses from any rework time. Either way, I learned
a lot and it has always stuck with me. Owning my business allowed me to
refect on how many engineers had never learned the same lesson.
My belief is that if all mechanical design engineers responsible for
machined or fabricated parts spent at least a few weeks working in a
machine or fabrication shop, the company would really beneft. When
I worked at Toyota, all new employeesespecially those without expe-
riencereceived hands-on training before starting in the ofce or the
shop. Tis is the traditional approach Toyota takes and a reason for their
reduced lead times and cost. Naturally, this principle is applicable for all
areas and disciplines.
10.11 suMMAry
Standardizing protocols for all processes leading to quality problems
or additional costs should be a priority for managers trying to avoid
daily frefghting.
Value stream mapping or process mapping (for administrative pro-
cesses) should be taught as a tool to objectively evaluate current pro-
cesses and identify improvement opportunities.
Developing a schedule should not be based only on the availability of
materials but also on the lead times and capacities of the processes
involved (machines and people). Tis will necessitate that a PDCA
(Plan, Do, Check, Act) cycle is continuously updated with new pro-
cesses and lead times.
Inventory that reduces lead time to help meet customer expectations
should be considered strategic inventory and viewed as a tempo-
Up-FrontDelays 159
rary improvement until other solutions are implemented to reduce
the lead times.
Upstream or downstream processes that are infuenced by your
activities/responsibilities are areas where you should receive hands-
on training; this will allow you to better provide what is needed,
ofen leading to signifcant cost savings.
161
11
Making Money during the Good Times
11.1 IntroductIon
Afer a few years of hanging on during a signifcant industrywide sales
slump, it seemed the dreams for all of us tied to the semiconductor
equipment industry were coming true. For some time, the chip-making
equipment industry had been developing new technology around etch-
ing on 300-mm diameter silicon wafers instead of the 200-mm diameter
wafers that had been the standard. Te chip makers (e.g., AMD, Hewlett-
Packard, IBM, and Intel) were preparing to invest and would receive
many benefts, like signifcant cost savings and higher yields, from the
300-mm advances. But, there were also disadvantages, like the cost of the
new tooling, the time-consuming and complex requalifcation of existing
part numbers on the new tools, more bowing and bending with the 300-
mm wafers, and various other problems that come along with new tech-
nology. We had been producing some prototype parts for the 300-mm
wafer machines and were now starting to see a heightened level of activity
to get some of this new technology into production.
11.2 IMProvIng Processes durIng
sloW PerIods PAys off
Taking advantage of the workforces available time to make improve-
ments during slow sales periods can really pay of when things trend back
to the upside. I was not only excited about the potential proftability and
the chance to grow the business, I was also feeling a little smug, know-
ing that all the process improvements I had put in place during the slow
times were about to really pay of. I was also hoping to be rewarded for
hanging on to my key employees during the downturn as we were now
in prime position not only to take full advantage of their skills but also,
with our improved documentation and standardization, we could quickly
train any new hires.
As normal when transitioning from the prototype to production stage
with highly technical products, the order quantities started small and
there were immediate requirements for a wide range of part numbers
to build these complex machines. I am not sure how many diferent
fabricated and machined part numbers an etching chamber has but it
appears to be in the thousands. We were starting to receive small produc-
tion orders for most of the prototype parts we had previously worked on
and were also busy with various other frst-builds on new parts. It was
a period when we tried to quote almost everything including both fab-
rications and machined parts that could not be manufactured in-house
(outsourcing was required).
Te ofce was busy and I kept thinking of the payback that would
come from all the improvements to the fow of the administrative tasks.
Tese improvements freed up more time for the staf to quickly provide
additional quotations to our customers, obtain prices from more than
one subcontractor, and request more than one quotation on all mate-
rials. Naturally, this led to increased sales with signifcant cost-saving
implications.
Te simplifed lean implementation strategy of OSKKK (Observe,
Layout) does not dictate where the primary focus should be (e.g., the
shop foor or the administrative processes); instead, OSKKK dictates
that the primary focus should be on bottlenecks. (See Appendix L,
which shows the actual sequence as we went back and forth between
shop and ofce improvements based on the needs of the business or
the bottlenecks.) Since I had shifed the focus back and forth between
the shop and the ofce depending on the perceived constraint, I would
now receive the full benefts coming from the entire business having an
increased capacity.
MakingMoneyduringtheGoodTimes 163
11.3 ProbleM-solvIng skIlls develoPed durIng
the sloW PerIods begIn to PAy off
Probably the most important thing I had partially gotten in place was the
development of my employees Kaizen eyes and increasing their prob-
lem-solving skills. Tis is not to say that I had spent any signifcant time
developing the operators or setup men but at least Tom (the shop fore-
man), Jake (the programmer/estimator), Sandra (the ofce manager), and
Andrea (Sandras assistant) had begun the process of learning how to
see, so to speak. Tis lef me with signifcant resources to deal with all
the new problems that we were sure to face if the business rebounded and
grew as quickly as it appeared to be doing. It would have been great had I
also spent more time evolving the setup men and the operators, but at least
I had always tried to ask questions and let them reach their own conclu-
sions. I had also encouraged Tom to coach the operatorsthough he still
enjoyed the quick and direct solution of telling people what to dobut I
understood this would take time, as dictating was the easiest and most
comfortable for him as it is for many of us.
Another reason for our less-formal problem-resolution method was
that we were small enough for it to be almost impossible to avoid going to
where the work was done when confronting a problem. Ofen, many com-
panies are criticized for quickly coming to solutions while sitting in meet-
ing rooms or shying away from the details and only being results-driven.
In my case, I only utilized a little structure to get people to specifcally
defne the problem and identify its root causes since most of the team
was already familiar with the details of many of the processes for which
they might be working on a solution. I only needed to ensure they did
not jump to conclusions without progressing through the proper thought
processes or analysis and miss out on the root cause. I prided myself on
being able to perform any task in the ofce or the shop and made sure I
did it with just enough frequency to keep from getting rusty or forgetting
the details. As I have said, the devil is in the details, and unless you
try performing the process, it is unlikely you can guide its improvement,
especially by only demanding that the measurements are improved. Tere
may be times the problem-solving skill level of your direct reports are suf-
fcient to implement working solutions, but keep in mind how expensive it
is to implement the wrong solutions. I am always interested in managers
who only demand results and do not get involved in the details because
they must believe their staf are very skilled and they themselves do not
add any value in the problem-solving process. If they only stomp around
demanding improved results (fnancial numbers), how can they be sure
they are really adding value to the company? Although I had arrived at
the comfort and confdence level that allowed me to accept work as a con-
sultant one week per month, I still believed I was adding value the other
three weeks by guiding and developing the problem solvers and resolving
a few issues myself.
Lean principle (from Chapter 4): Problem solving in lean terms
involves a structured approach basing the solution on data gathered
through observation. Normally, this requires a minimum of fve steps
(as discussed in Chapter 4) or could involve more complex statistical
methods like Six Sigma or a structured approach to engage and develop
those involved like A3.
When working to identify opportunities in larger organizations, those
involved with the processes should be guided through value stream
mapping for processes involving a physical product or through process
mapping for administrative and ofce processes (see Appendix M for clar-
ifcation on these applications).
11.4 the leAder keePs In PrActIce
I periodically challenged those who performed daily tasks in the shop and
ofce to determine who was fasterthem or me. For example, I would
compare the time it took to set up similar jobs (similar number of tools,
measuring points, etc.) on one of the machines. We had an agreement
that any signifcant mistakes disqualifed that individual and the other
won by default. Tis ensured a high quality level but also made the point
that standardization supported productivity. I set up the challenge for
two reasons: (1) to keep my skill levels up and (2) it was motivating and
encouraging to the operators to create a little competition. I always drove
home the point that speed wasnt important if you lost it on qualitybut
it was fun. Actually, I lost more than half the time, which obviously made
my team happy. But, I did continue to earn their respect, especially on the
occasions when I won.
As the number of incoming orders increased, we immediately returned
to a forty-hour workweek. Afer months of short workweeks, now I was
ofering overtime and had plenty of volunteers. Tom was always the frst to
get overtime and the last to lose it because I always needed his skills over-
seeing the shop, preparing the machines, and watching the quality. It was
not long before Jake was also putting in some extra hours writing all the
new programs that were required. As it became obvious that we needed an
additional setup man, our concern was in fnding someone with experi-
ence with our type of machinery so the new person could quickly become
productive. As I indicated earlier, this was usually not a problem as we had
equipment that was common within our industry and many were famil-
iar with the controls. But, these were not normal timesour competitors
were also ramping up quickly and looking for qualifed people.
Tom threw out a suggestion about hiring his brother, who had previ-
ously worked therehe was familiar with the business and Tom said he
was a very hard worker. If his work ethic was anything like Toms, that
would be a real attribute, but then I remembered some strange advice from
the former owner, Bob. He had recommended not hiring Toms brother if
it was suggested, apparently Toms brother had previously worked there
and come to blows with Bob over attendance issues. Being pressed to sat-
isfy my customers and take advantage of some potential sales records, I
decided to take the chance.
Bert, Toms brother, seemed honest enough in the interview, admitting
that he had screwed up with his attendance while working for Bob. I
explained our rather demanding attendance policya combined total of
three times late or absent in a rolling ninety-day period meant immediate
dismissaland Bert said that would not be a problem for him. Also, Bert
felt he would be comfortable working for his brother, and Tom had prom-
ised me that he would recommend fring Bert immediately if he got out of
lineno hard feelings. In this case, I believed Tom since he had never put
up with any unacceptable behavior and I imagined he would not make an
exception for his brother.
Bert got to work quickly and was really adding value. He was an
extremely hard worker like his brother (I guess it runs in the family) and
appeared to be a team player. His biggest fault was that he jumped into all
tasks so quick and hard that he periodically missed a detail in the setup,
which cost a bit of scrap or extra adjusting time.
Within the next month, we took on another operator, and this growth
pattern continued for some time. I was so busy that I was starting to shy
away from committing to too much consulting work as I was continually
managing/supporting our overloaded shop and ofce processes.
11.5 dIversIfIcAtIon Would hAve been helPful
Sudden growth spurts are not ideal in most situations and we were stretch-
ing ourselves to cope. We really had our hands full with six out of our
seven largest customers being in the same industry. We would not have
gone through these wild gyrations in sales levels had we been able to
change our portfolio of customers through diversifcation. I continued
working to evolve customers in other industries but up to this point I had
only succeeded with one new customer in medical devices and two in the
microwave/radar manufacturing industry. So, about 70 percent of our sales
volume still remained in the semiconductor equipment industry. Wanting
to be a better salesman was not going to get the job done. So, I had tried a
few diferent independent sales reps who specialized in machined and/or
fabricated parts and did not have a confict of interest by also representing
one of our competitors. Although none of the experiences were negative,
they never really brought in any signifcant requests for quotes, and the few
jobs we landed through them were not highly proftable. We were scraping
the bottom of the barrel, so to speak, with these types of jobs.
11.6 the need for A second shIft
About six months into the production phase of the 300-mm etching
machines, we were running out of capacity for one shif. We had many
of the tools in place that had proved helpful in increasing the existing
shifs capacity. For example, we were now excelling at capacity planning
and monitoring against that plan. We had also made strides with setup
reductions and increased machine feeds and speeds up to the limits of new
and superior tooling. Terefore, a second shif was next in the cards.
Operating a second shif was always a point I wanted to arrive at.
Actually, I dreamed of a three-shif operation, as most owners would.
Te downside is that without competent people to manage the various
shifs, an owner could fnd himself dealing with problems at all hours. We
also found ourselves in the typical growth dilemma in which we did not
require the capacity of a full second shif but were lef to justify the extra
associated administration costs.
Ideally, Tom would manage the second shif and in the short term I
would manage the frst shif. We could immediately move a few people
and some work to the second shif to help create a balance. Te problem
was that Tom had a very large family, actually nine kids, and wanted to be
at home with them in the evenings. In the end, it was Bert who was willing
to run the second shif, and he appeared to have sufcient skills.
Te plan was that Tom would work a little overtime to make sure the
second shif had a smooth start and in the near term I would handle
phone calls during the evening and deal with any problems outside Berts
comfort level. Te plan was two fold: (1) allocate only the less-complex
jobs to the second shif, (2) schedule more work than they have capacity
to handle so if they reached a point on any job in which they could not
solve the problem, they could leave it set up on that machine and move to
another machine. Because the second shif would not be operating all the
machines, it would be possible to leave a particular machine set up when
a problem arose and move to another job. Ten, the problem would be
resolved the next morning. Naturally, the second shif had to clearly com-
municate the problem so Tom or I could quickly resolve it the next day.
Te second shif ran about as smoothly as could be expected. Bert had
two of the better operators with him in the evenings and was tackling the
easier jobs. I periodically received a phone call at home and occasionally I
went into the shop when there were enough problems simultaneously that
the operators did not have another job to work on (no use paying people
to stand around).
Te second shif began immediately afer the frst shif ended, which
was not ideal for a number of reasons: Te frst shif had no catch-up abil-
ity for problems. Te frst shif basically could not work overtime to make
up for shortfalls in completing the planned production for the shif, which
in turn could afect on-time delivery. Te main reason we went this way
was because Bert felt more comfortable with a direct handover and he and
the other operators did not want to work too late since the public buses ran
infrequently afer midnight. Also, we were not sure how long the second
shif would last based on the customers demand so we wanted to make
any transitions back and forth as painless as possible.
While I worked with Toyota I learned they normally planned a gap
between their shifs allowing them the ability to make up for any lost
production, almost always achieving the shifs planned production
goal. Terefore, if it appeared they were not going to reach it, all afected
departments were notifedtwo hours prior to the end of the shifof
the required minutes or hours of necessary overtime (always in incre-
ments of six minutes). As condition of employment, an employee had to
be available to work up to two hours of overtime each day. Tis allowed
them to maintain their schedule and on-time delivery whether adapting
to temporary increases in customer demand or recovering from various
minor production problems. (Obviously, major disruptions could not be
compensated for in the two hours between the frst and second shifs and
typically necessitated working on Saturday.)
Even though I was not able to implement a gap of time between the
shifs, I want to capture it as a key point (when either you are operating
something less than a three-shif, continuous operation or assuming you
do not incur start-up losses afer a process has been shut down).
Key point: Allowing a gap of time between shifs allows the option for
overtime accommodating for either temporary increases in customer
demand or to make up for production losses. Tis should help in main-
taining a high on-time delivery.
We were also working most Saturdays during this period and I was hit-
ting Starbucks for more cofee and donuts than I care to think about to
maintain everyones cafeine and sugar levels. You cant maintain a long-
term plan of increased working hours by keeping a workforce hyped up on
sugar and cafeine, but nobody knew for how long those 300-mm etching
machines were going to be selling. Ours was not the type of business whose
customers provided forecasts, and the one or two we received were about
as useful as that famous paper used in the bathroom. So, my plan was to
take in all the business that we could manage while maintaining our high
quality and on-time delivery levels, supplementing this with overtime and
the second shif. I was sampling my team to check if working on Saturdays
was starting to impact their private lives, but everyone was happy with the
overtime money that they had lived without for a few years. My warning
was not to get used to the extra money as it could end at any time.
Te other option was to slow sales down by raising the prices. Like any-
one else, we had the option to raise prices to the point where our supply (or
capacity) was balanced with demand. Te danger is that up to 60 percent
of the parts we were manufacturing would receive repeat orders and it
was likely these new 300-mm machines would be the standard for years
to come. So, it was in our best interest to try and stay competitively priced
while securing many parts into our portfolio. Tese repetitive purchases
also applied to the parts we were currently outsourcing, some of which
could be made in-house (though on less-than-ideal machines) during a
slowdown to help in proftably utilizing excess capacity when necessary.
Also, keep in mind that we had a higher proft margin on repetitive parts
as we then had a debugged program and saved time on the adjustment
phase of our setup with the methodologies previously discussed.
Tere was other good news during this proftable period: I was busy
paying down the balance on my bank loan. Although the term was for fve
years, I had set a goal of paying it of in four mostly due to the 11 percent
rate and the vision I had that I was paying for luxury cars for the bankers,
and with this boom, it was looking possible.
11.7 the entrePreneurIAl sPIrIt PAys further
dIvIdends durIng the busy tIMes
During the slow period, we had extended our capabilities, not really
through investment butin desperation, to stay proftablewe took on
jobs that we might not have previously considered. Some were just dif-
cult to machine, but most required additional operations that would not
have traditionally been considered our core competency. In reality, some
of them required creativity and courage in fnding a method to produce
them to the quality level required. But, since I was an engineer instead of
a machinist managing a machine shop, I had no preconceptions of what
machinists did and did not do; therefore, there were no roadblocks in my
mind and every drawing we received for quotation was fair game if I could
fgure out how to manufacture it without signifcant investment. I would
like to say that Tom and Jake supported me in this direction but, unfortu-
nately, they were more traditional machinists. And although I was mak-
ing progress at opening up their minds, their big leaps of faith only came
each time we took on a non-traditional job and were able to manufacture it
without signifcant pain. Tese were confdence builders and the next simi-
lar item that came along was not such a stretch of their imagination and
was likely more proftable. Tat is not to say that we didnt have a few losers
on which we really struggled to perform these non-traditional operations,
and we periodically would decide not to take on more work with similar
characteristics. Examples of the type of work we expanded into included
some heating and forming operations; we also got into a bit of metal rolling
(mostly into circular shapes), working in exotic plastics and rubber materi-
als, and also the fusing together of plastics utilizing chemicals. While none
of these new areas were by themselves signifcant, every bit helped during
that slow period and during the busy times the customers were funneling
more of this type of work our way. During this market upturn we could
also charge a higher rate for these specialized services as there were not
many other open-minded competitors wanting to experiment and learn
these new processes. Whether or not a higher price was justifed, our new
variable pricing structure led us to charge a higher rate for this type of work
because signifcant time (activity) was involved from the indirects in the
ofce (locating materials and tooling) and those on the shop foor (devel-
oping the necessary manufacturing processes). Considering how this new
pricing applied to these products further helped in proving to me that this
attempt at ABC costing was likely moving in the right direction.
11.8 structured ProbleM solvIng
MInIMIzes oPInIons
Unfortunately, this quick sales growth was a natural cause of tension,
especially between the employees and me. Tis was on top of the natu-
ral tension that exists between a boss and those who report to him or
her in any working relationship. In my case, I was trying to get as much
production out as possible, not only for short-term proftability but on the
chance some of those parts would become repeat orders as the new 300-
mm etching machines became the industry standard. Helping to keep the
tension in check were the overtime hours and job-security that accompa-
nied this growth. So, no matter how hard I pushed to get product out dur-
ing the day, nobody worried that if they worked harder during their shif,
they would lose any overtime. Again, we all understood overtime was not
the long-term solution but a way to deal with this increased volume while
the second shif evolved. One key component I used to resolve tension
during this period was data gathering. Tis helped remove any emotion
between myself and the team. So, when a problem was encounteredand
many were springing up during this periodif it was caused for any rea-
son other than not following standardized procedures, we waited to see if
it came up again before making a big deal out of it, which helped us in not
continuously fghting the wrong battles.
Lean principle: Discussions based around data are something I consider
a lean principle. Instead of discussions where participants indicate it
happens a lot or it happens all the time, simple data gathering (i.e.,
tick lists) can be used to help prioritize the problems and guide the dis-
cussions. One of my favorite sayings in this regard is In God we trust,
all others bring data.
Tom and I were mentally keeping track of the issues. When we noticed a
pattern, we either started a tick list to see how ofen it happened or we tried
to informally apply some simple but structured problem solving to get to
the root cause instead of just frefghting with our gut instincts. Lean prac-
titioners might prefer that we utilized a more structured problem solving
or completed an A3 (an eleven-inch-by-seventeen-inch, one-page struc-
tured problem-solving guide). We were not formally completing A3s as I
had ofen done while working for Toyota, which meant we were not going
through a comprehensive dialogue and back and forth iterations to clarify,
fnd the root causes, and plan a trial solution for one of the possible causes.
Unfortunately, we did not go through this type of A3 process, which could
have helped me evolve as a leader and my employees grow into potential
leaders and better problem solvers. But, we had a simple structure every-
one followed. Tis allowed us to keep everything based on fact.
As I mentioned earlier, we were small enough that anyone involved in
resolving problems was likely already intimately familiar with the related
processes. Also, I had driven everyone to be process focused, meaning that
when a problem occurred he or she would think about what went wrong
in the process and ofen questioning whether a process even existed for
that task. We were then inclined to either modify the existing process or
create a simple but robust process so the problem would not repeat itself.
It was difcult not to just jump in and tell someone what I thought the
solution was and what needed to be done (actually, this took place more
than I would like to admit). Being small and having worked hard to set in
motion fact gathering and process-based principles, I had a relative basis
from which to successfully resolve issues without using a more structured
method to resolve problems. But, in larger organizations my experience
has shown this structure is a necessary step for learning and development.
Although I only used some of the simple analysis tools like the fve whys,
fshbone diagrams (man, method, machine, materials), and a few times
completed a simplifed written format (fve steps) to resolve our signif-
cant/larger problems, I regret not developing myself and my team further
in this regard.
I would like to capture this as a key point. A structured method helps
an organization grow and develop employees without creating signifcant
tension between the boss and the employees.
Lean principle: A3 is a problem-solving tool (a fact-based tool that leads
to efective solutions) but can also be utilized as a management process.
Unfortunately, during this period I did not make the time available to
further my skills and manage with this tool, though in hindsight it might
have helped, especially in the development of Tom, Jake, and me.
Key point: To resolve problems in larger organizations, a structured
method is needed that allows facts to be gathered through observation
where the work is performed, and to generate more than one proposed
solution leading to trials and a follow-up method. Tis structure will
allow you to develop the team into taking ownership and responsibil-
ity while evolving everyone into logical thinkers who are detail orien-
tated. A written problem-solving format (known as an A3 at Toyota) is
recommended. OSKKK can also be utilized as a basic methodology for
resolving issues.
If you are still not convinced of the benefts of at least periodically get-
ting involved with the details, compare the fnancial results from two of
my previous employers, Toyota and General Motors. Toyota managers
generally are involved in the details, ofen using A3s to focus their eforts.
GM managers generally concentrate on the fnancial numbers and shy
away from involvement in the details. You know the rest of the story.
11.9 ProbleM solvIng requIres lookIng Into
the Process detAIls usIng A structure
Even during the busy times, when I became involved in solving a problem
(not to say I was always involved), I had to get into the details to be able
not only to realize the root causes but also to try countermeasures that
were likely to work in the long term. Although it is a necessity to jump into
the details, a structure is needed to guide problem solving so those with
a technical understanding of the process do not assume they know the
solution without working to arrive at the root cause. I think I have a good
scientifc head that applies logic to resolving problems, but I know you
must dive full force into the nitty-gritty details to fnd the real solutions.
I learned the business from the bottom upmaybe too quickly for some
peoples tastebut I understood how each process worked and I either felt
confdent enough to discuss any problems that came up or I went back into
the process using a simple fve-step problem-solving approach to arrive at
a solution.
In most large businesses, managers do not see the need or do not feel
they have the time to learn the various tasks, and in complex/technical
industries this may not be feasible. But, a manager at least needs to get
into the details when resolving a problem unless the manager feels there is
nothing additional he or she can ofer by involving problem-solving skills
or coaching abilities.
Te managers at most companies that I support as a consultant are con-
tent to run from one issue to the next, usually only demanding results from
the metrics and assuming their subordinates have the abilities to fgure out
how to get the problems resolved. Ofen, I am lef to run a workshop with-
out the managers (who appear too busy managing the results), and so I
get into the details with those assigned the task of resolving the problems.
Frequently, nobody in the room has a problem facilitation methodology to
follow (hence, the need for a consultant) but most feel they already know
the cause, some have their own solutions in mind, and others feel the prob-
lem cannot be resolved. Usually it is interesting to note afer working on
the frst step (which is to specifcally defne the problem) how many do not
understand the entire process and there is rarely sufcient data to focus
the group or the analysis. Ofen the workshops require a break while those
involved go out to better understand the processes and gather the facts or
some data. No matter how ofen I request some of this data to be gathered
prior to the beginning of a workshop, it is rarely completed with enough
detail to start a thorough discussion. If the managers instead would coach
employees through a logical problem-solving approach (i.e., A3), they
would create a team of future leaders while simultaneously developing
their own skills. However, afer many years in consulting, I still fnd that
most are too busy attending meetings to sit patiently and dive into the
details and evolve their teams problem-solving skills.
Well, enough talk about how the industrial world should functionback
to life in my busy machining business. By now, we had been operating prof-
itably for more than a year in this growth market and I had achieved a
record of sales and proftability within the companys twenty-three-year
history, actually almost doubling the previous owners best year. Te loan
was now paid of, which would allow me all kinds of freedom to consider
expansion. Te second shif had actually not lasted long, and sales remained
high. But, as we were performing more additional operations like assem-
bly and some of those new bending and plastic fusing jobs I previously
mentioned, the machines were no longer the bottleneck. We went instead
to a staggered shif, meaning Bert and two operators started their day two
hours later than everybody else, giving us ten hours per day of machine
time before going into overtime. Everyone was able to stay busy either run-
ning a machine or assembling or performing various other operations. I
was not disappointed with canceling the afernoon shif as we remained
at about the same level of sales and proftability, and I was not required to
pay a shif premium. Now, my phone never rang late at night, although the
early morning calls still periodically interrupted my sleep and that oh-so-
wonderful peace of mind. I viewed those phone calls as what you get when
youre involved in the details, care about the productivity, and are not able
to implement standard processes to deal with every conceivable situation.
11.10 equAlIty In enforcIng your
oWn PolIcIes Is crItIcAl
I ended up having to let Bert go because of attendance problems. Of all
the employees I had to fre, fring Bert was the most difcult for me. He
was a great worker with sufcient machining skills, and he was basically
my second most-productive employee, next to his brother Tom. He made
a lot of money for the company, had a great attitude, and was a real hustler
at keeping things moving. However, I had adopted Toyotas strict atten-
dance policy and had to enforce my own rules to remain fair. When Bert
exceeded the limit of the combined three times late or absent within a
rolling ninety-day period, I actually considered relaxing the policy just
so I could keep him on. But, I knew that as soon as I started bending the
rules or making exceptions, it would be unfair and I would lose respect
from the entire team. I prided myself on being strict but fair, so Bert had
to go. When I called him in the ofce to tell him it was his last day, he did
not even argue. He said, I knew the rules and screwed up. He walked
out knowing the blame was his alone and Tom (his brother) supported
this decision 100 percent. Tom actually used a little stronger language
describing his brothers behavior by saying he fu___ up and needs to pay
the price. Ironically, when I later sold the business the new owner also
ignored my warning and hired Bert. I do not know how long he lasted
with the new owner, but it had the makings of a script for a soap opera.
11.11 suMMAry
Leans most lucrative improvements frequently come from proftably
selling the capacity created via improvements.
Signifcant problems should be solved with a structured approach
instead of jumping to conclusions. Either a basic problem-solving
approach or the A3 methodology should be used just like OSKKK is
followed when transforming a business.
Sudden growth is hard to cope with; but standardized processes that
minimize training time combined with a cross-trained workforce
provide the best chance to take advantage of this.
Discussions based on generalizations like it happens a lot cause
a machine-gun approach to Kaizening. Instead, focus on recurring
problems, even if this requires you to go back and quantify their
severity with simple data-gathering tools like tick lists.
177
12
Expanding into New Products
12.1 IntroductIon
As it goes, the good times dont last forever. Afer almost two years of record
sales levels resulting from the introduction of 300-mm diameter wafer-
etching machines, the industry was once again slowing down. It appeared
that a lot of the chip makers who were planning to switch from 200-mm to
300-mm diameter wafers had already done so, and there was talk around
the industry about a long slowdown. By the way, I heard most of my indus-
try gossip from the metal and aluminum sales reps that called on us. Some
days they appeared to me like old ladies running around with plenty of
time to gossip. Normally I was too busy to give them much of my time and
only really took the time to discuss any quality problems on the occasions
when they arose (out of tolerance material, late deliveries, material dam-
aged in transport, etc.), though when things were slow I found myself with
more time and suddenly was willing to listen to a bit of gossip.
12.2 the logIc behInd exPAndIng Into
sheet MetAl fAbrIcAtIons
Now that I had paid of my bank loan, I was determined to expand the busi-
ness. Tis was further driven home by the fact that I had time on my hands
and needed to fnd more work or new products to ofset my overhead costs and
remain proftable. I had previously purchased a new, larger CNC (Computer
Numerical Control) machine to expand our capabilities and capacity into
manufacturing larger parts. I had also replaced some older equipment with
newer CNC machines. Te logical expansion I now envisioned was into
sheet metal fabrications since for some years we had already been providing
these products to some of our customers (though we were outsourcing this
work). I fgured if we already had some demand in this related market with
a few of our clients and were getting away with a 15-percent markup during
the good times, with our available foor space all we needed was equipment
and the know-how to manufacture these parts in-house.
Tis is where that same nave drive that gave me the confdence to buy
the business came into play. I fgured if I had been able to quickly learn
the machining business, I could sort out sheet metal fabricationwhat
could be so difcult, right? Te big diference was going to be that instead
of having a trained staf with many years of experience to teach me the
business, I was going to have to stumble through it on my own and then
teach my team. Once again, I was going to have to convince my employees
I wasnt crazy. But, at least the employees who had been with me from the
start fgured that if I could buy a business in which I had no experience,
learn it quickly, and make continuous improvements, wouldnt I be able to
fgure out how to punch out some metal shapes, and bend and weld them
before sending them out for paint and silk-screening? No big deal, right?
I already had my strategy to go forward in this new businessnaturally it
was also OSKKK (Observe, Standardize, Kaizen Flow & Process, Kaizen
Equipment, and Kaizen Layout). However, the observation step would be
difcult since there were no existing processes to watch and our subcon-
tractors would probably take a dim view of being used in that way.
Well, I didnt have as much at risk as it might appear. My plan was to buy
some simple pieces of used equipment to get started in the business, and if
it went well I could expand things once I really understood how to manu-
facture the parts and knew what the local market demanded (i.e., the best
equipment to ft our niche). I was also planning to start only with easy parts
and work my way into the more complex parts. I could begin any time
as repeat orders were always trickling in for the previous fabrication/sheet
metal parts we had supplied. And, I could start with some of the simple
ones instead of subcontracting them. I also had the option to hire someone
with sheet metal experience but as we were not going to follow all the tra-
ditional sheet metal processes (e.g., using a punch or laser cutter to cut the
ExpandingintoNewProducts 179
geometry), this had its limitations. Te ace in the hole was that instead of
investing in a big CNC punch or laser cutter and learning to program it, I
thought we could start by cutting the geometry on our existing larger CNC
machining centers. Although not specifcally made for this, I assumed I
could develop some vacuum fxtures to hold the sheets of metal in place (if
that didnt work I could go with electro-magnetic tables) and use machine
tooling to cut the geometry. I knew it would not be as fast as a punch or
laser cutter but the batch sizes were normally small in our industry; so if we
could keep our setup times short, meaning comparatively less than that of
a punch, it should work. I also thought we were very good on setups, and if
we could develop some quick fxturing we would be well on our way.
Out I went to buy a basic book on forming sheet metal parts. My plan
was to buy a used shear and start with a manual press to form the metal.
A manual press would only allow me to bend light-gauge metal but I
fgured it was a start and, if successful, I could quickly buy a larger press
(known as a press brake). Welding equipment would also come later.
One other item that would be required was a hardware press as most
sheet metal had requirements for various small pieces of hardware (nuts,
studs, etc.) to be pressed in place. Te consumable hardware was easy to
obtain though it required a few ton press and the associated tooling.
I had some discussions with my father (my advisor and fellow investor)
and he also felt this expansion was the right step. He thought the risk was
minimal, assuming we ensured a high level of quality. Te advantage, he
noted, was that, as long as we started early and had sufcient lead time,
if we encountered a major problem and were not able to manufacture the
parts, we could always go back and ask our subcontractors to make them.
Naturally, this might afect our on-time delivery (the promise date), which
was currently at about 98 percent, but this was the biggest downside other
than possibly investing some time (which we had) and money into some
equipment that might not provide sufcient payback if we failed.
12.3 PrevIous exPAnsIon consIderAtIons resulted
In lIttle IMPAct to the current lAyout
A few weeks later I found a used equipment auction that ofered a shear
that could handle the size and gauge metal we would likely be working
with. At the auction, my father and I found the selling prices were staying
fairly low but most of the equipment was old and outdateduntil the point
when the shear went up on the auction block. As luck would have it, every-
one appeared to have been waiting for that shear. Opening bids moved up
quickly into the thousands though I persisted until we fnally walked away
as the winners. Te convenience of these types of auctions is that heavy-
equipment movers are milling around ofering quotes on transport of the
purchased equipment. Te shear was delivered a few days later and the
electrician arrived soon thereafer to complete the connections (naturally,
all connections were made fexible for any future layout changes).
When I frst decided on the expansion, we had previously planned for
our layout to include the additional space for the equipment we were
likely to purchase. Te layouts limits were the space available and the
lack of knowledge/experience of which type of equipment would really be
required. When we had moved to our new location, I already had this type
of expansion in mind and the foresight that we wanted all existing equip-
ment together to reduce travel distance. Tis had resulted in an open area
near the shippingreceiving door. It was also critical that this area be near
the door as the raw material for sheet metal was typically provided in four-
feet-by-eight-feet or four-feet-by-ten-feet fat sheets of aluminum, steel, or
stainless steel. You did not want to be moving these sheets all the way
to the back of the shop before shearing them into smaller pieces. On the
other hand, the existing business of machined parts normally started with
smaller purchased material already cut to size or bar stock that was cut
to size near the receiving door. So, it was easier to transport this smaller
material throughout the shop.
Knowing the products into which you might expand is critical when
planning any layout so you can prepare for future expansion. Many busi-
nesses seem to spread out their current equipment to fll the existing
space, feeling more comfortable using the space and not being cramped
together (though this typically leads to poor 5Sworkplace organization).
By envisioning future expansion and leaving an area open as a continuous
reminder of the possibilities, our growth was further simplifed. I would
like to capture this as a key point.
Key point: Te best layout is to keep all current processes close together
(taking into account all lean/fow considerations) and leave all open
space grouped according to where you believe future expansion will take
place. Tis minimizes current travel distances and seeing this open space
continuously reminds everyone of the growth opportunities.
A picture of the original equipment and layout we put in place for
shearing and bending of sheet metal is shown in Appendix K. Again,
within this area we took into consideration all the previous layout points
that had been reviewed when we moved to our new location (discussed
in Chapter 8). We had additional space in a separate area to set up weld-
ing equipment and hardware installation in the future. Any assembly
or further operations could be performed on our existing assembly
benches. Te only equipment purchase that would cause serious layout
concerns would be a CNC punch or laser machine to cut the metal
these machines required more space than we had available. I would need
to look into renting part of the unit next door, which was being used as
a warehouse for a neighboring business. (You might recall that when we
moved to this location, the leaseholder of that unit agreed to rent some
of the space to me if the need arose.) Tough all this was way down the
road since I fgured I could use our larger CNC machining centers to cut
the geometry for the time being and when the volume or the order quan-
tities reached a point where we were not competitive, then we would
consider investing in a punch or laser.
I was also busy designing and testing vacuum fxtures to hold the sheet
metal in place on our CNC machining centers. As previously mentioned,
a great advantage of a machine shop is many types of parts and fxtures
can be built, which allowed us to do all this in-house. My engineering
background was coming into play in designing these fxtures and speci-
fying the vacuum systems we needed to purchase. We had machine time
available to work on producing these fxture plates, and afer manufac-
turing about the third prototype we had enough vacuum pressure to hold
the aluminum parts and a few of the lighter gauge steel parts in place.
Te biggest problem was linked to where we had to drill holes and make
cutouts because once we created a hole in the sheet metal, we lost the
pressure from at least one of the vacuum points (zones) in that area and
the parts periodically started moving. We worked on some variations
utilizing diferent fxtures to cover the eventualities and separated the
vacuum pressures into diferent zones within the fxtures. Tis gave us
some fexibility but also required that we change fxtures more ofen than
I had anticipated, increasing changeover time and reducing our ability
to compete. In the end, we settled with a growing number of vacuum
plates that also incorporated some adjustable mechanical clampdown
points. Tis got us started and actually held an advantage over many
sheet metal punches; for example, when we additionally clamped a part
down, we could mill and drill through thicker and harder materials than
most CNC punches could go through, opening up a niche market within
the heavier gauge materials.
Working in these heavier gauges (this niche market) was a consider-
ation for the size of press brake we would later purchase. To manufacture
parts in this unique niche now required a press that would allow us to
also bend these thicker gauge metals. Tis would open a market niche that
the average sheet metal business did not have and could be a potentially
proftable one for our non-conventional approach. Tis is another point
worth capturing.
Key point: When developing new processes, make sure to understand
what special capabilities you can add that will allow you to serve unique
markets that support potentially higher proft margins.
Afer the shear was up and running, we found it was a very easy tool to
operate and it required almost no training. I had also purchased a small
manual bending press that could handle up to 16-gauge material. Tis
would be a starting point and later we would purchase a used press brake
with sufcient capacity for the heavier materials. Te frst bit of difculty
was in determining the overall dimension of the sheared material by tak-
ing into account the bending coefcients that were required. When
you bend metaldepending on a combination of the type of metal, its
thickness, and the bending radius of the tooling you are usingyou must
subtract a coefcient (these are provided in various tables) for each bend
afer summing the lengths of the various edges involved. It is not worth
going into any more detail, though this was the frst step that required a
bit of experience. So we pulled out some prints for our customers lighter
gauge simple sheet metal parts that had previously been outsourced and
started experimenting.
12.4 fIllIng the fIrst order
Afer a bit of playing around and throwing a lot of small bent pieces into
the scrap bin, we fgured we were ready to try flling our frst order. We had
learned that our shear held very tight tolerances, which made us extremely
happy. On the other hand, the manual bender was difcult to use and
did not hold accuracy very well. Both Tom (the shop foreman) and Jake
(the programmer/estimator) had been experimenting along with me and
were quite excited to learn these new skills and bend metal afer spend-
ing most of their lives only cutting it. Afer waiting almost two weeks,
we received a purchase order for a previously subcontracted simple sheet
metal partffeen pieces of a simple bracket. We went to work and soon
had the parts manufactured within tolerance. Other than a bit more time
and scrap than would have been competitively acceptable, we had done it
and were on our way. For the time being, neither Jake nor I were ready to
start doing any quoting for these parts until we had a bit more experience,
but we could certainly build in-house any repeat orders for which we had
the capabilities and grow into more complicated parts with more bends,
more hardware, and tougher geometry.
Regarding preparation of quotes for newly fabricated parts being
requested, we fnally ended up cheating a little. Jake and I would take our
best shot at coming up with times and costs and we simultaneously sent it
out to our subcontractors for quotation. We then compared our numbers
to theirs and tried to understand the diferences, though this was difcult
because we could only see their fnal cost in dollars and could not deter-
mine where the diference could be within our labor times, labor rates and
material costs. We adapted our standard quoting process to incorporate
the diferences for sheet metal parts (including a slightly diferent labor
rate) and tried to include all the possible tasks for a sheet metal part on a
single page, even services like welding, which we were not currently doing.
Now, we had one quote page for machined parts and one for sheet metal
fabrications. Te sheet metal was only in its infancy and likely would go
through a lot of iterations as we learned and continuously improved the
process. But, for the time being it helped guide us so we did not forget to
include any costs. We only used our subcontractors quotes as a check-
ing system in the short term, and within a few months we had enough
confdence to quote on our own. For our sheet metal quoting process,
our observation was done by comparing our quotation with those from
our subcontractors, than we were able to evolve a standardized process,
which would later be followed by continuously Kaizening the process,
always mirroring OSKKK.
We quickly arrived at the point where we needed to purchase a proper
bending press, known in the industry as a press brake. I determined that
between a twenty-fve-ton to thirty-fve-ton press would ensure we could
bend the thicker materials, many of which the local sheet metal shops
could not produce as they lacked the capabilities to punch the holes or
shapes in these materials. I located a press that would do what we needed
on eBay. Te seller was located on the East Cost, which would add time
and transportation costs to get it to California, but none of the local equip-
ment brokers had anything meeting my requirements and price range. As
with the shear, the price went a little higher on the eBay auction than I
would have hoped for, but I came out the winner. I found an equipment
transporter and things were put in motion. Once again, the electrician ran
some fexible power lines and we had already allowed sufcient space in
our layout (this is the press brake shown in Appendix K). So, the instal-
lation went smoothly. We had also purchased the hardware press and
stocked it with what we were fnding to be some of the most commonly
demanded hardware. Tis small hardware ofen had to be purchased in
500- or 1,000-piece lots (not an ideal minimum-order quantity) and some
part numbers had a few days lead time.
Since this hardware was not an expensive item, we set up a simple orga-
nization system and reordered through a basic Kanban system (frequently
referred to as consumable Kanban). As most of you have learned, these small
items should be treated as consumables and a simple replenishment system
that reduces administration costs should be utilized. Supplier-managed
replenishment systems (where the customer is not involved in managing
the inventory) are a great option, but this was not available to us.
Te press brake arrived without tooling so we started purchasing some
common dies. Each new tool required a learning curve and frequently we
had to come up with a custom set of bending coefcients for the various
materials we were using it on instead of simply using the coefcients from
the sheet metal handbooks. Te vacuum plates combined with the fex-
ible clamping were working fairly well at holding the sheets of material in
place, and we were getting quicker on our setups and faster at verifying the
measurements according to the print. While programming the parts, Jake
and I learned early on we had to indicate the bending coefcient being
utilized on the setup sheet so that whomever was verifying the dimensions
could subtract out this number (i.e., 0.045 inches/bend) when verifying
the dimensions. Other than this complication, most of the steps were the
same as on the machined parts. Te only other additional step is where
there were square the corners on the rectangular punched holes called
out. We were cutting this with a small end mill so the corners had a mini-
mum radius, which required that the operators additionally use a specially
designed fle to manually square up the corners.
12.5 InvestIng steP by steP
From the incoming requests for quotations, those that required welding
were still being outsourced and we had not yet started aggressively pursu-
ing our customers for additional sheet metal work until we acquired more
experience. Slowly, more complicated parts containing more bending work
were being manufactured in-house and our tooling for the press brake and
hardware press was growing. Now, our biggest limitation was welding. We
fnally reached the point that we required the ability to weld aluminum,
stainless, and mild steels to expand and meet the required specifcations
demanded by our customers. Welding the various types of material to the
required quality levels were not skills we could quickly learn in-house.
Terefore, my idea was to hire a part-time welder. I could also use that
person to help advise what type of welding equipment I should purchase
based on the type of work we were currently outsourcing. I knew terms
like MIG and TIG welding but was not sure which of the capabilities were
required. Also, I had yet to be confronted with a request for any welding
certifcations on the work I was doing (I assume because most of it was not
safety critical). But I was sure this was likely to come up.
I quickly learned that what I required was TIG (tungsten inert gas) weld-
ing equipment. TIG uses a shield of inert gas, usually argon, to shield the
process from the atmosphere, resulting in no slag spraying out all over the
part and almost no porosity in the welds. A TIG system using DC (direct
current) electricity can weld stainless and mild steels, and a unit using
AC (alternating current) electricity can weld aluminum. MIG (metal inert
gas), also known as wire feed welding, has two disadvantages: it produces
lots of spatter and creates porosity in the welds. Tough MIG is much
easier to use, it would not produce the type of welds required for the work
I wanted to perform. So all I needed was to fnd a used TIG with both AC
and DC electrical capacities and a welder who knew TIG welding inside
and out, could obtain the necessary quality levels, and was willing to work
part time. How difcult could that beright?
While discussing a complication with some work I had subcontracted to
a sheet metal shop down the street, I mentioned to the owner that I wanted
to buy a used TIG machine and fnd someone to help me on a part-time
basis. He told me he had just replaced one of his TIG machines with a
new super-duper machine and wanted to sell the old unit, which had both
AC and DC capabilities. It got even better when he indicated his primary
welder, who had twenty years experience and was one of the best he had
ever come across, was always looking to pick up additional work. I imme-
diately bought the TIG welding machine and took the contact details for
his welder. By the way, I did inform him that I was planning to enter the
sheet metal business with limited capabilities, and would build the smaller
and simpler parts in-house but still send out any larger or more complex
parts. I also put in a sales pitch for doing work in thicker gauges that he
did not have the capabilities to punch through. Later on, I did periodically
receive subcontracted work from him for heavier-gauge sheet metal parts,
which just goes to show that although I disliked the sales side of the busi-
ness, maybe I was developing a fair for seizing on new opportunities.
Te TIG unit arrived and Emilio, the welder, was not only interested in
working for us part time, he also volunteered to fabricate a welding table
and order the necessary start-up supplies on his own time. I could not
accept his ofer to perform some work without receiving pay, but we agreed
on a pay rate. And while he was getting my little welding shop in order,
we agreed on a simple system to notify him when I required welding on a
particular afernoon, I would leave a text message on his cell phone during
the day and at 3:00 p.m., when he fnished his shif, he would come and do
my welding. In the beginning when I was learning the welding symbols in
the drawings that I was quoting, he would tell me how long it would take
him to weld those particular parts. He had no problems reading prints
and estimating times. And just to add a little icing on the cake, he had a
friend with similar skills. who could do the welding when Emilio was not
available. As we all know, you dont want to put all your eggs in one basket,
and I could not put my business in a position where we might temporarily
be without welding capabilities in a quick-turn industry. Afer meeting
with his friend, we set up a similar arrangement and contact system. Tis
functioned very well the entire time I owned the business.
Te point to reinforce here is that you need at least two qualifed people
to perform every task within your businessanything less is too risky.
Terefore, you should have some type of cross-training matrix visible to
ensure you meet the minimum requirement of two trained employees for
each job in your shop and ofce. Although many try to keep track of it
in their head, it should be documented in a matrix that visually displays
all employees and all the tasks performed within your business, further
indicating who is qualifed (and maybe to what level, on a scale of one to
four) for each task. You can also show your future training plans in a dif-
ferent color on the same matrix. An example is shown in Appendix H.
Key point: A necessity for all businesses is a skills/cross-training matrix
showing the current training status of all employees (ofce and shop
foor) in relation to all tasks. It should further indicate any planned
training to make up for shortfalls anywhere there are fewer than two
competent people trained to perform a particular task. Tis should not
be managed in someones head but should be visible for all to see.
12.6 oskkk tAkes to lIfe WIthIn
our neW Processes
Now we were able to manufacture in-house many of the jobs that had
formerly gone out to subcontractors. We were also doing our own quot-
ing for sheet metal parts and were continuing to evolve a standardized
quoting process specifc for sheet metal work and fabrications separate
from our quote sheet for machined parts. Because we now had a little
experience with these parts, we started the same tracking of actual hours
against quoted hours that we had been utilizing for machined parts. Since
we were less experienced and encountered more problems with these
parts, we had a much larger variance between our quoted times and actual
times. I was reviewing these on Fridays with both Jake and Tom, not only
to understand what we could learn for future quoting but also what we
had to do in our processes to remove some of the waste and reduce the
time required. Again, this feedback loop (PDCA: Plan, Do, Check, Act)
in the quoting process proved critical in driving improvement, and it is
especially prevalent on new products or those for which you do not have
a lot of experience.
12.7 once AgAIn, A PdcA cycle In the
quotIng Process Proves crItIcAl
I continue to come across companies that estimate labor times for a new
project, determine the material costs plus any development or indirect
costs, and afer determining the fnal price they are periodically con-
fronted with a customer or market that demands a lower price. Instead of
just working backwards from the market price and adjusting their hourly
rates and/or proft margins to determine whether this new target price is
acceptable, they start changing their estimated hours to ft in line with
the new target price suggested by the customer. Normally, they cite antici-
pated efciency improvements as the justifcation for reducing the esti-
mated hours. But, I have found this to be a mistake. It most ofen takes
place in organizations that do not have a robust labor-quoting system,
or feel the labor is only a small and negligible part of the entire quota-
tion, or do not really trust their quoted hours. Tese are the same frms
that typically do not have a good track record of making improvements
in working methods to reduce these labor hours. Instead, I recommend
internally developing the quoting skills for labor hours by instituting a
PDCA loop for the estimators, which should result in more faith in quoted
labor hours. Ten, when faced with determining whether or not to agree
to a customers target price, work backwards to calculate the hourly rate
(in dollars) associated with the target price:
(Target price material costs
other costs required proft)
= hourly rate you will receive

quoted hours
and decide whether this is acceptable. I experienced instances where a cus-
tomer gave me a target price with the option to take it or leave it; in these
cases, I would back-calculate the new hourly rate and decide whether there
was enough proft lef in the job.
Now back to our newly acquired ability to weld in-house. Another
advantage we discovered was that when machining an expensive part and
too large a hole or cutout was made or when we removed too much mate-
rial, Emilio could ofen fll in our mistake with some of his wonderful
welding and we were able to re-machine the existing part instead of hav-
ing to purchase more material and start over. I assume this was done by
others in the industry but I am sure it is not something most companies
admit takes place.
Lean principle: Balancing machine cycle time with the operators cycle
time when specifying new equipment benefts from lean thinking. Even
within a made-to-order business where machine cycles are likely to
vary depending on the various products being built, average cycle times
can be estimated (for typical parts), and what the operator does within
that cycle should be considered before investing too much in automa-
tion. For example, I fnd many companies invest in additional auto
load robots and conveyor systems on the same machine on which they
also pay an operator to stand and wait for the machine to cycle. Tis
was also a consideration as I purchased equipment for the sheet metal
business. (See Figure 12.1.)
Auto load/unload
0
5
10
15
20
25
30
35
40
45
Average machine cycle
Average operator wait time
(could be quickly loading & unloading)
S
e
c
o
n
d
s
Periodic Quality Check
Deburr Part
fIgure 12.1
12.8 AgAIn, engIneers lAck of
hAnds-on exPerIence Is costly
Afer a period of manufacturing these sheet metal parts, I again found (like
I had discovered with machined parts) that the engineers ofen did not
design parts that were easy to manufacture, and therefore they unknow-
ingly increased the costs. I observed this in many forms. For example,
there were ofen difcult tolerances to maintain (especially when a few
bends were involved) or no place (fllets or cutouts) to allow room for the
weld along with difcult areas to bend when it would have been more
cost-efective to weld two pieces together. Again, this demonstrates the
necessity and cost justifcation for engineers to acquire hands-on experi-
ence in manufacturing if they are designing fabricated products. Or, at
least it demonstrates a need for a design for manufacturing process to
be put in place.
Over the next year, we grew the sheet metal business to somewhere
around 30 percent of our total sales volume. We slowly progressed to more
difcult parts and added some simple equipment to roll round shapes.
During my time as the business owner, we never did purchase the CNC
punch or laser cutting machine. We continuously improved cutting the
geometry on our CNC machines and, although it limited the size of mate-
rial we could work with and the batch sizes for which we could remain
competitive, it allowed us to produce some unique parts (at higher mar-
gins) that the traditional sheet metal shops were unable to manufacture.
Buying the punch or laser was the next logical expansion step, but we
never did achieve it during my tenure as owner.
Tis unique combination of supplying both machined and sheet metal
parts was an advantage to many of our customers. Many larger custom-
ers were being pushed to consolidate their supplier base, yet their pur-
chasing agents were frequently frustrated when they sent out drawings
to either a machine or sheet metal shop (because they could not dis-
tinguish between machined or sheet metal parts in the drawings) and
the suppliers would pick and choose which parts they had the capabili-
ties to quote and manufacture. We, on the other hand, ofered more of
a total solution, and although we could not manufacture some of the
parts, we attempted to quote them all (some through subcontractors).
It was preferable for the overworked buyers to receive from one supplier
a complete quote package with competitive prices andideallyshort
delivery times for all the drawings they sent out. Terefore, we continued
to work our way into that idealized preferred supplier status with some
of our customers. Again, having customers with overloaded buyers who
are not technically savvy is a situation for all suppliers to take advantage
of by adapting their services. Tese buyers also rarely have the time of
inclination to visit their suppliers; therefore they do not recognize their
in-house capabilities. Te best situation is not to have your own purchas-
ers overloaded (always having time for obtaining competitive quotes and
managing suppliers) while putting in place methods to take advantage
of your customers busy buyers. Having worked with many companies
purchasing departments, I have found incredible amounts of time lost
on non-value-added activities and little time devoted to critical strate-
gies like sourcing new suppliers. Remember: if 50 percent or more of
your cost goes into purchased materials, a lot of opportunities lie within
this department (and having the time available to obtain the best pric-
ing, lead time, and quality leads to great payback).
I think many companies expand into related products afer they have
determined an opportunity exists. Te example from my business holds
the most interest for companies that do not develop their own products.
But, even for those frms responsible for development, some of the lessons
learned here are valuable. For example, wherever possible, try to estab-
lish a presence in the market before making any signifcant investments
(like temporarily subcontracting some or all of the steps) while evolving
as an established supplier for these items. Minimize investments where
possible until you have learned a bit about the product and market. Look
for easy-entrance products to get into the market and later develop the
skills and capabilities for providing the more difcult products or ser-
vices. Tats what I learned evolving my business into a new market,
and as things slowed down in our industry, this diversifcation really
helped. To put it in perspective, over 400 similar businesses (machining
and sheet metal companies) in California alone went bankrupt during
this time period, but I feel we survived this turbulent period because of
our focus on the continuous elimination of waste along with our expan-
sion into sheet metal.
12.9 suMMAry
Expansion into similar and complementary products is a logical step.
Job shops can minimize the risk of growth into new areas by frst
outsourcing these new parts while simultaneously developing the in-
house capabilities.
Existing layouts should not be allowed to expand into all the available
space; instead, they should consume the minimum space required
(helping 5S and fow) and leave the open area free of debris to remind
everyone of the growth opportunities.
Minimize investment by balancing the average machines cycle time
with operators task times, not overinvesting in automation while an
operator is being paid to wait on the machine.
193
13
Business for Sale
13.1 IntroductIon
Although we were not experiencing a boom in the primary industry we
served, we were holding our own. Te addition of the sheet metal capabili-
ties had positioned us uniquely in the marketwe were one of the few shops
that ofered not only machined parts but smaller sheet metal and fabricated
parts. Our customers continued to fnd this attractive as we could competi-
tively supply more types of parts, reducing their supply base and the efort
involved in obtaining quotations and in the purchasing and receiving pro-
cesses. Also, the fact that our on-time delivery remained near 98 percent
kept us in the A tier of their respective supplier management programs.
Te sheet metal side of my business was in its early growth stages; we
were learning where our strengths and weaknesses resided and thinking
in terms of which direction we could further grow this area. Our non-
conventional methods of cutting/punching the part on a CNC (Computer
Numerical Control) machining center instead of punching out the geom-
etry were allowing us to work in a unique market niche for heavier-gauge
formed parts where there were fewer competitors. With all this going for
us, plus knowing that the cyclical nature of the semiconductor equipment
industry likely meant it would bounce back soon, and with our increased
capabilities we would potentially be more proftable than ever, why on
earth would I consider selling the business?
Te reason for selling lies in my personal life and not in the state of
afairs with the business environment. It came down to an agreement
between my wife and me. Bottom line: we decided it was a good time to
move to Spain, my wifes country.
Te lean consulting work I had stumbled into through my previous
employer had expanded to a point that I felt could grow into a livelihood
based out of Spain. I had been consulting almost one week per month and
just by word of mouth had expanded past having Delphi Automotive as my
sole client. I fgured if I was fnding more consulting work without looking
for it, I could likely fnd enough work in Europe if I aggressively pursued it.
With the decision made, it was time for me to learn how to sell a business.
13.2 deterMInIng the Process stePs
for sellIng A busIness
To sell the business, I thought I probably just needed to reverse the steps
I had gone through to buy it fve years earlier. Since I had not started the
business and grown it from my garage, I likely had less emotion involved
in selling than the former owner. I just needed to fgure out where to start
and the elements involved. I did a bit of brainstorming with my father,
who had previously owned and sold a larger business, and we thought that
I should frst do some research and read up on current methods for selling
a business. A few of the books I bought proved to be less helpful than I had
hoped and also functioned better than prescription sleeping pills, but I
found a few worthwhile bits of advice that helped me to rough out a plan.
I am basically opposed to using middlemen who add little value or bring
almost no practical experience to the table; therefore, putting a business
broker between me and potential buyers ofered few advantages in my
mind. I recalled the young salesman (for the business broker) who was
involved when I bought the business: he added some experience in the
process steps of buying a company but had no background in the industry
and the previous owner was the one answering most questions. Terefore,
I felt if either the Internet or advertising in other media would bring me
together with potential buyers, I could handle the rest.
Before I could list the business for sale, I had to frst get some fnancial
records in order and then come up with a method to determine an ask-
ing price. My father and I discussed using an analysis that involved a bit
BusinessforSale 195
more than just a valuation based on earnings. Since neither of us were
accountants by profession, we decided to seek professional help.
13.3 deterMInIng An AskIng PrIce
We found a local CPA (certifed public accountant) who had experience
in evaluating sales prices for companies; although most of his experience
came from service, fnancial, or retail companies, we decided he was the
best ft. We provided him with the previous fve years of tax returns along
with our fnancial records. He worked with the same fnancial sofware
that we used so it was easy for him to pull the necessary data directly from
the computer fles. Afer a few meetings to fll in the blanks, we deter-
mined to fx a selling price based on a combination of earnings and a mar-
ket valuation (weighted more toward the earnings). Although assets are
a consideration, most potential buyers are more concerned with future
proftability. In our case, the matter regarding assets has more to do with
the age and condition of the various machinery in terms of when addi-
tional investment may have been required. Remember: we were selling the
company for the value it held as an ongoing business (the real value is
in the goodwillthe customers, processes, employees skills, technology,
past earnings, patents, etc.) and we were not looking for an asset sale.
Te accountants analysis was based on taking the yearly income (add-
ing back any owners salary) and subtracting the costs to hire an expe-
rienced business manager to replace me. Ten he took an average of the
remaining income from 1999 through 2002 and divided it by his antici-
pated rates of return15 percent, 18 percent, and 20 percentto deter-
mine a price range. He also completed a projected income analysis for
the next three years and completed the same steps with the same desired
rates of return, giving us three more potential asking prices. His advice
was to choose a starting price from among the six numbers he provided.
Te idea was to look at the business as an investment with a certain fac-
tor of risk, making the assumption that it would not be owner-operated,
and then depending on the rate of return wanted on the investment (the
accountant assumed an investor would demand a rate of return between
15 percent and 20 percent), which then determined the asking price. My
father and I were more comfortable using the valuation from the last
three years of earnings as those were actual numbers and it was roughly
the same type of valuation method we used when buying the business.
Forecasting sales numbers going forward is best lef to the potential
buyer, and although we would share this part of the analysis, they could
play with the what-ifs. We simply felt more comfortable basing our ask-
ing price on the historic reality.
13.4 A MArketIng PAckAge
Once we had a starting price in mind, it was time to put together a mar-
keting package. To align with the selling process I wanted to put in place,
I put together a package that had to be released in steps, frst providing
enough detail to interest potential buyers without providing the name of
the business. Aferwards, a potential buyer could sign a non-disclosure
agreement, afer which we would provide a second package containing
more fnancial and product detail, general terms and conditions (e.g., I
would not fnance more than 10 percent). If interest continued, the poten-
tial buyer would have to send me some company or personal background
along with fnancial statementsbank account statements, stock state-
ments, or letters of credit from a bank, etc.to prove the ability to pay. If
I found this acceptable, then we would meet at a neutral location (restau-
rant, cofee shop, etc) and I would provide the buyer the third and fnal
marketing package, which contained detailed fnancials and a few years of
tax returns. During this meeting, I planned to answer questions and afer
giving the buyer time to further review the third package of information,
I could set up an afer-hours appointment for the buyer to visit the busi-
ness. Tat was my concept, and I went to work putting the packages of
information together.
I also set up a separate email account and bought another cell phone,
both of which were only to be used by potential buyers when making con-
tact. Tis would help in maintaining anonymity for the business, ensuring
that no calls or emails were intercepted by the employees and the name of
the business was not accidently disclosed. Tis is critical to minimize the
employees and the customers anxieties.
BusinessforSale 197
13.5 AdvertIsIng
Now, I was ready to start my advertising campaign. I planned to start
on the Internet, and as I recalled from when I was trying to buy the
business, some business for sale Web sites (open market style) were
available where potential sellers could list their oferings and potential
buyers could either search or periodically run ads stating the criteria for
the types of companies they were interested in purchasing. I found that
since the time when I had purchased the business, the Internet had really
grown in this area. Tere were so many choices of where to advertise
that I had to spend many hours just trying to qualify and rank the vari-
ous Web sites. I used criteria like how easy they were to fnd on search
engines, what types of companies they specialized in, the amount of
daily trafc they had, how easy the site was to use, and the cost to place
an ad. I started with running ads on two of the larger sites and clearly
stated that I wanted principals only.
What would turn out to be a long process had fnally begun. I learned
some of the tricks that brokers used to try and weasel their way into get-
ting an exclusive listing for selling a company. Tey started contacting me
immediately afer my listing was posted (although I had stated principals
only). Tey claimed that they represented a client who was interested but
wanted to meet with me before introducing the client. My answer in the
early stages was always that I was not interested in working through a bro-
ker and would see how things went for a few months before considering
any broker intervention. At that point, I was not willing to pay the 10-per-
cent commission that business brokers were looking for until I deemed it
necessary. I was also convinced that an exclusive listing was not a route
I wanted to follow as I thought it really would limit my reach and tie my
hands in terms of being proactive.
13.6 the fIrst nIbbles
Finally, inquiries did arrive from investors, from companies in related
industries looking to expand, and from private individuals. My procedure
was very simple and the frst steps could be done though email or fax if
necessary. If they were still interested afer the frst general marketing
package, I would email them a boilerplate non-disclosure form for them to
sign and send back, then I would email out the second part of my market-
ing package. It seemed to contain enough information to interest some of
those potential buyers to go the next step. However, because of the difer-
ent formats on those business for sale Web sites, I ended up answering
a lot of inquires that only clarifed for some buyers that this was not the
business for them. Terefore, as I expanded my advertising, I chose only
those Web sites that allowed enough description and details (like maxi-
mum 10-percent seller fnancing) to eliminate the casual interest of the
typical tire-kickers, which helped in minimizing the waste of my time
by taking the frst few steps with those who were not serious.
13.7 consIderAtIons of oWner fInAncIng
Owner fnancing is an interesting subjecta bit like a double-edged
sword. Obviously, the larger the percentage of the sale price a seller is
willing to fnance, the more risk they are willing to share. If you indicate
that you will fnance a fair portion of the asking price, then it can be said
you are stating confdence in the businesss longevity and certainty in the
buyers ability to manage the business. On the other hand, if you will not
become involved in the fnancing to some extent, it makes a statement
about your faith in the businesss future. When I bought the business,
Bob (the former owner) did not ofer any fnancing and I did not push
the issue. But, in retrospect I should have requested some fnancing to
secure his interest in the success of the business. Te problem I now faced
was trusting a potential buyers ability to learn and successfully manage
the business as I had done. I fgured that ofering 10-percent fnancing
showed a bit of confdence in the business. But, I felt I could not justify
ofering more as I did not have an in-depth understanding of the poten-
tial buyers ability to make the correct management decisions. All the
process standardization I put in place would also help in supporting the
businesss success and place better odds on me recovering the 10 percent
I fnanced. I was also refecting on all the security a bank could ask for
that a seller could not (with seller fnancing). For example, in my case the
bank had required me to assume personal responsibility for the value of
BusinessforSale 199
the loan by having frst rights to all business assets. I had to commit to
securing all personal assets and agree to potential future garnishments
against me, plus a life insurance policy that would cover the value of the
loan in the event of my demise. Any seller fnancing was always going to
be subrogated to the primary lender if things went belly-up and I would
likely only have rights to any business assets that remained afer all bank
debt was accounted for.
Lean principle: Process standardization should be considered an asset
when expelling the efort to develop the standards and also later when
selling the business. You have minimized reliance on tribal knowledge
and ofen decreased the skill level necessary to complete certain tasks.
As the months rolled on without any ofers, I increased my advertis-
ing with additional Web sites and started running newspaper ads. Also,
when brokers called indicating they represented a potential buyer, I began
ofering up to a 5-percent commission if they were to secure the sale. But,
I stuck to not listing with any agencies and absolutely no exclusive listings
up to this point. I had met ofsite with a few potential buyers to answer
their questions. About four months afer listing the business, I had only
hosted two afer-hours visits of the plant and neither seemed the ideal can-
didate and no ofers came forward.
13.8 loWerIng the PrIce
I did what most in my situation would do and started lowering the ask-
ing price as I was mentally ready to move on to the next stage of my life.
Te new price created a slight increase in trafc; therefore, more requests
for additional information were coming my way. Most potential buyers
appeared impressed with my three-stage organizational approach of dis-
tributing material a step at a time to help in establishing an interest level
before both sides really started sharing critical fnancial data. Back when
I was buying the business, I had determined that banks working with the
Small Business Administration basically liked to see 30 percent or more
as a down payment; so, when I looked over a prospective buyers fnancial
data it was fairly easy to approximate the cash they needed to qualify for
conventional fnancing.
As a seller, I wanted to appear confdent without appearing like I thought
I had the worlds best business mind or had pursued every possible avenue
to expand and improve the business. Remember: every buyer has their
own improvement ideas and is thinking how they can proftably grow
the business. My angle had been the implementation of lean along with a
little common sense, and although I felt most of the low-hanging fruit
had been harvested, I was salesman enough not to advertise that thought.
I did not believe all the lean avenues had been exhausted but the next
steps would require signifcantly more efort. I also felt that many growth
opportunities still existed (including many outside the scope of my idea to
further expand in the sheet metal area), and lef it to my salesmanship to
plant these ideas in the minds of the prospective buyers.
13.9 three serIous buyers
It was probably seven or eight months since I had listed the business before
some serious prospects came along. Tree were worth getting excited
about: Te frst was the owner of four separate companies including one
that sold small aluminum enclosures to house various electronic confgu-
rations that invariably required outsourced machining. Te second was
an aggressive investor who had bought and grown various businesses and
had accumulated millions in equity. And the third was an individual with
the fnancial means and a sales background who wanted to get out of cor-
porate America and run his own company.
Many long discussions followed with the frst prospect, and he even
set up a pilot program where his team was subcontracting to us some of
the machine work for his companys small electric enclosures (the entire
pilot ran without any problems). Unfortunately, afer all the discussions
and efort he was not able to put together an ofer as quickly as the other
prospects. Obviously, his desire was vertical integration through which he
could provide additional sales (and increased proftability) to an existing
machining company like mine. Plus, he could also reduce his lead time
as he would control the planning and priorities instead of relying on out-
sourcing the machining of his electronic boxes to someone else.
BusinessforSale 201
Afer about nine months, three separate ofers came in within the same
week: Te frst came from the aggressive investor. Te second came from
the individual who wanted to get out of corporate America. And surpris-
ingly, the third came from another individual who seemed a little ofeat
and did not ask the right questions or display the level of interest that the
others had. But, lo and behold, out of the blue he also submitted an ofer. I
referred to him as Mr. Lef-Field.
13.10 evAluAtIng offers
Evaluating ofers is probably another subject of which entire books are written,
and it was also one of the many areas where I lacked experience. Whenever I
require some knowledge about a process with which I am not familiar, I start
by seeing what is available on the Internet. Based on reading a few articles, I
determined the criteria to use and also a list of the pitfalls to watch out for.
Naturally, price is likely the most important starting point. However, draw-
ing an analogy to selling your house: if the buyer making the highest bid is
not ofering to put much money down in good faith or has included a lot of
stipulations in their ofer, you would likely consider more than just the price.
Tis was basically what I had in front of me:
Te highest ofer, which came from the aggressive investor, was
written with more big lawyerly words than I could make sense
of, and there were many escape clauses. I remember that it had so
many clauses that would allow his investment frm to back out that
it seemed more like a fshing expedition than a genuine ofer. One
of the many stipulations that could render the ofer null and void
was that his investment frm wanted to meet beforehand with our
two largest customers and discuss our services and their future sales
prognosis. As far as I am aware, speaking with customers prior to the
sale is uncommon. And it was too dangerous for me to contemplate.
First of all, the customers could become anxious about the sale and
start looking for more stable suppliers. And, I would not be in con-
trol of this meeting so I had no way of knowing what else could be
brought up. On top of that, the investment company had written in
other clauses for which they could back out of the deal and I would
be lef patching things up. For these reasons, along with many of the
other scary clauses all favoring the buyer, I did not accept it even
though it was the highest ofer.
Next in line in terms of price was from Mr. Lef-Field. It seemed
to be missing details and serious thought and only ofered a mere
$10,000 down as a good faith deposit upon the ofers acceptance.
Tis ofer also set of alarms inside my head. One issue was the
small deposit being ofered as good faith along with the fact that it
appeared he wrote the ofer himself without seeking any legal advice
or thinking through the details. Although he had displayed enough
in assets to qualify with the 30-percent down payment, I fgured he
would either not receive fnancing, or if I agreed to fnance 10 per-
cent, the business would not last long enough for me to recover my
money. Although in the end you may be thinking that just selling
it and getting your money out is the most important thing, I really
wanted to see the business do well. Tis is especially true in terms
of the livelihood it provided for the employees. It was important to
know they would be looked afer and I did not feel this man would
be able to do it.
Te lowest ofer was put together by the individuals lawyer, and it
was straightforward and appeared to be the most sincere. My father
commented how the third ofer appeared most similar to the ofer he
helped me construct almost six years earlier when I purchased the
business. Terefore, the lowest priced ofer was the one I accepted.
13.11 offer AccePted
I had already found a title company to handle the transaction (similar
to a title company that handles home sales but this type of frm works
with all the legalities of buying and selling a business). We agreed to use
the title company I recommended and further arranged to split this cost.
Te good faith money was put in an escrow account held by the title
company. We were both provided with a list (diferent lists for the buyer
and seller) of what needed to be prepared and completed prior to sign-
ing. Tere are considerably more details and legal requirements than those
involved in the sale of a home but its just a matter of punching your way
BusinessforSale 203
through the list. Te buyer and I agreed on a target date of approximately
one month as we each felt that was the period necessary to complete our
respective tasks. He had his lawyer draw up the sales contract based on
the terms in the signed ofer. I had my lawyer look it over and, afer the
lawyers batted the document back and forth and held conference calls at
about $300/hour, we fnally reached a sales contract with all the is dotted
and ts crossed and were on the fnal countdown toward the big day.
Te terms of the sale included a non-compete clause of at least two
years, meaning I could not open a similar business in Northern California
during that period. I had also agreed to stay on and manage the business for
six weeks during which I would train the new owner. His background was
in technical sales but his plan was to hire an experienced general manager
to run the business while he went out and beat the bushes to increase sales.
As my plan had been to implement lean to grow the business, his angle to
grow the business was through superior salesmanship (where his exper-
tise and experience resided); I wish I knew the fnal results (lean vs. sales-
manship). Te idea was that during the six weeks of training I would help
interview and hire an experienced manager. In the meantime, I was busy
building and shipping what I could prior to the signing date at which point
the buyer would own all raw material, fnished goods, and open orders lef
on the books. With certain customers I could not ship early but where pos-
sible I was going to maximize my profts with early deliveries on those few
items where longer lead times had been granted. I would then own all the
accounts receivable for everything shipped until the handover date.
Te employees were not informed until afer the papers had been signed
and, of course, the money was in the bank. It was a Friday afernoon
when we signed, and although I suggested we break the news together,
the new owner told me I could do it alone and he would not be in until
Monday morning. I was a little surprised as this is when the employees
need the most reassurance that things will continue in more or less the
same fashion. So, this pep talk usually came from the new owner detailing
how he planned to operate and usually included some improvement ideas.
I remember when buying the business that I was very concerned with how
I handled this discussion to minimize the employees anxieties. Terefore,
although this time it was not me assuming the reins, I decided to wait until
Monday morning when he was there so we could make the announcement
together. Watching the team on Monday morning, I think Tom (the shop
foreman) took the news the hardest as we had worked closely together
over the years. During the next couple of weeks, I tried to assure him that
I really respected his skills and dedication and enjoyed our time together
but explained again how it was time for me to move on. I also tried to
reassure him that the business was secure and even explained that I had
fnanced some of it so my interests were also at stake.
13.12 drAWIng to An end
It was ofen pointed out to me that the process of selling a business of this
type without previous sales experience or involving a business broker was
risky. In the end, I did all right, and although I likely made some small
mistakes, I do not believe I would have found a buyer quicker or someone
willing to pay more using a broker. I would have just been out 10 percent
of the sales price. I think in this case hiring a broker is a heavy price to
pay. With the mainstreaming of the Internet, more options are available
for bringing together parties with similar interests. Terefore, I would like
to capture this as the last learning point, not necessarily a lean idea but a
basic concept.
Key point: When you are required to perform a process for the frst time
that can also be handled by an outsider, frst understand the difculty
and steps involved and compare this to the cost, time, and skills you
have available for the task. Dont get sold into thinking you cant do it
without prior experience until you at least understand what is involved
and what benefts that prior experience will likely bring. Learning new
processes makes us all better people.
As I was writing this book almost six years afer the sale of my company,
I knew that the business was still rolling along. I have not kept close tabs
on how the business has been progressing, though I am happy to say I
received every penny back from the portion I fnanced. I would like to
believe part of the ongoing success was from some of the more robust and
standardized processes we put in place during my tenure. Because things
were standardized and documented, I know it was easier for me to train
BusinessforSale 205
the new owner than it had been when I received my training from Bob..
More ofen than not, I was able to refer to standardized and document
processes as I handed the business over. And I was almost jealous when
refecting on how simple it was for the new owner compared to the anxi-
ety and late nights I had gone through until I had some of these critical
procedures in place.
Naturally, there is an emotional side to buying, improving, and then
selling a company. Overall, I am happier in life, having removed some of
the stresses associated with being an owner, especially those for which you
have minimal control or infuence over, such as the upward and downward
shifs of the primary industry you serve. On the fip side, there are things
I miss, like the feelings of excitement when we secured a good order, or
proftably growing the business, or in implementing improvements that
had an impact on the bottom line. It is also difcult to walk away from the
natural kick you get from being completely responsible.
13.13 suMMAry
When confronted with overseeing a process for which you have no
experience, be sure to learn existing process steps from various per-
spectives before deciding how you will outline and handle the spe-
cifc process.
Standardized processes should be considered a business asset as
they help ensure ongoing proftability without complete reliance on
maintaining the same workforce over the long term because of their
tribal knowledge.
207
14
Refections
14.1 IntroductIon
One of my biggest fears in writing this book was that readers might not see
how my experience of implementing lean methods in a small business can be
applied elsewhere. However, afer years of supporting lean implementations
in mostly medium-sized companies (typically made-to-order businesses or
service companies, including small subsidiaries of larger corporations), I
have found again and again that many of the strategies I utilized in my
business are imperative, and the same basic issues exist requiring similar
solutions. Although areas within my lean transformation were fairly tra-
ditional and you may have already implemented some of these steps, I also
hoped that other discoveries are likely unique to job-shop environments
and might provide some new avenues to explore in your lean journey.
Despite having had my share of failures (e.g., never implementing a suc-
cessful bonus system or structured problem solving), I learned from each
of them but will continue to draw my learnings from some of the underly-
ing principles and philosophies that proved most successful.
14.2 lessons leArned
To recap: I bought a company with no background in the industry and
implemented lean principles as they were called for by simply following
OSKKK (Observe, Standardize, Kaizen Flow & Process, Kaizen Equipment,
and Kaizen Layout). Utilizing this strategy drew about thirty-nine vari-
ous lean ideas or tools (noted in the textboxes) into the discussion to help
improve my business. It would have been difcult on day one to determine
exactly which lean ideas would be necessary and what the priority should
be. But, by following the common sense of the OSKKK methodology, I
pulled in the necessary tools as they were required and it all fell into place.
I proved, at least to myself, that implementing lean can lead to proftable
growth as well as increasing the proft margin of most existing work, even
in a job-shop environment.
You can start an OSKKK strategy in either the shop, the ofce, or both
as dictated by the needs of the business. Te priorities are likely diferent
for every company and vary over time depending on many factors. I could
have never predicted the ups and downs in our industry and the opportuni-
ties they would ofer. During busy times, most of our improvement eforts
focused on increasing capacity in both the shop and the ofce. Bottlenecks
were not limited to the shop but ofen cropped up in the ofce, also lead-
ing to opportunities. Te slow periods presented opportunities in terms
of the team members having the time available to make changes. I was
confronted with the choice of reducing costs (through downsizing) in the
short term or implementing improvements that would be most proftable
during the next sales upturn. We utilized a combination, always choosing
frst to stay busy implementing improvements during the slow periods,
and only downsizing as a last resort.
Furthermore, I found that I could not be rigid in my implementation of
OSKKK for either the shop or ofce, especially when major opportunities
presented themselves. I needed to be fexible enough to adapt my plan and
take advantage of these events. For example, the combination of a pending
rent increase and a slow business period led me to jump to the last step of
my implementation plan (Kaizen Layout) by moving the business to a new
location, improving the layouts of both the shop and ofce in the process.
I knew OSKKK was fexible enough to allow me to adapt to this change in
the business environment.
Te sequence I ended up utilizing to learn and transform the busi-
ness (shown in Appendix L) demonstrates the fexibility of the OSKKK
approach to adapt to our continuously changing business environment.
It became clear from my learning and Kaizening sequence (as well as my
mothers sequence in the ofce) that the observation and standardization
Refections 209
steps always came frst and that Kaizening the fow and processes was
where most of the improvements were concentrated. Te chart (Appendix
L) captures our fnal sequence (not our plan) and shows that Kaizening the
equipment and layout came later for the most part, generally prioritized
based on events taking place within our business environment.
Te opportunity for me to learn a business from the bottom up and
take time to familiarize myself with most processes made me a much bet-
ter change agent than most managers I encounter. Many try to manage
from the numbers, in the worst cases comparing mostly historic fnancial
results to the budget (which is only someones best guess based on recent
information). Toyota had taught me that the devil is in the details and
allowed me time to learn the processes I would later be held responsible
to oversee. Toyota was well-rewarded for this investment of my time. Tis
demonstrates the importance of taking time for observation as promoted
by the frst step within OSKKK. So, the lesson I learned is not to be too
obsessed with managing by the fnancials or too busy frefghting; nothing
beats taking proper time for observation before jumping to conclusions.
I tried to make PDCA (Plan, Do, Check, Act) the basis for most of my
actions and not just utilizing it to guide problem solving. We are all natu-
ral problem solvers though you need a method (like PDCA, structured
problem solving, suggestion systems, etc.) to bring out these natural prob-
lem solving abilities/desires; remember you can force people into perform-
ing a task, but you cannot force them to be interested. Honestly, most of
us can only say we try to utilize PDCA since it is sometimes difcult not
to just tell somebody what you think the solution is and instruct them to
implement it. With the minimal meetings we held, I was able to ask more
questions at the work areas, letting people better identify and analyze the
specifc problems or issues (Plan). Ten, afer they developed and imple-
mented solutions (Do), I could (Check) on the results to see whether they
were achieving the desired goal. Most of the time, all this took place infor-
mally but if the desired goal was reached, I made sure the full-scale solu-
tion was implemented and documented (Act). Te idea of putting some
structure in my week (standardized work for managers) and making this
auditing/review schedule public knowledge by displaying it on a board
above my desk was helpful in structuring this approach and forcing me
to lead by example. Tis was further enhanced by my focus on the details
and encouraged me to go to Gemba (the real place, usually referring to
where value is added) to observe instead of working out solutions blindly.
14.3 usIng the correct MeAsureMents
Is crItIcAl to AchIevIng results
Probably one of the most critical factors was the trial and error of learning
the correct measurements that drove the business to achieve a high level
of proftability. Oddly enough, none of the measures that helped in daily
management turned out to be the fnancial numbers; those were a result
of focusing on the correct business metrics (i.e., periodically monitoring
more timely indicators like lead time for certain customers). In my made-
to-order business (job shop), it was critical to measure the actual times and
material usage against the quoted numbers that represented the custom-
ers expectations. In addition to truly understanding where I made and
lost money, I became better at pricing, which during industry downturns
became quite a competitive advantage. Lead time is another critical mea-
surement for almost all shop-foor and ofce processes, and is even more
signifcant in quick-turn industries. Other important measurements for
machine-based processes like ours were OEE (overall equipment efective-
ness) and DSO (days sales outstanding), which helped improve our cash
fow (reducing the necessary working capital). On-time deliveryholding
to our promise datewas also extremely critical in the fast-turn industry
we supported; this was frequently more critical than the price. Although
this book might not have focused on the issue of quality, it was an under-
lying assumption in everything we did. Higher-priced companies can
still keep a dialogue going with their customers but nobody speaks with
poor-quality suppliers. For me, DSI (days supply of inventory) was not a
critical measurement as we carried little inventory (although DSI should
be the metric used instead of inventory turns). But, for many medium-
sized companies this and other measures like labor productivity, sales per
employee, and employee turnover are all relevant. In my mind, the right
measurements are critical for at least two reasons: the wrong measures
lead to the wrong behavior, and using the right measurements with an
understanding of what infuences them will guide your team to focus their
limited resources on the correct priorities.
Another point in using lean measurements and implementing lean
principles is to help bridge the gap between the owners (or executives)
focus on proftability and the operators understanding of what he or she
can do to help improve the stability of the business, thereby guaranteeing
Refections 211
job security. Examples of this: when I converted the percentage improve-
ment in OEE to dollars to clarify for Tom (the shop foreman) why he was
more valuable as a team leader. And, when I used lead-time measurements
(in days) showing how long it took us to convert some raw materials into
fnished products so all the operators could see how reducing days of lead
time could help to earn more money.
14.4 cross-trAInIng Is A crItIcAl PArt of
success In hIgh-vArIAtIon coMPAnIes
Producing a high mix of part numbers with a wide variety of tasks shared
among a few people made cross-training vital, not only in the shop but
ofen more so in the ofce. Visualizing the current status of our cross-
training (as shown in Appendix H), coupled with a training plan, allowed
us to manufacture a wide variety of products with a small team. It also
helped in maintaining our high rate of on-time delivery. Having a training
plan linked to a wage system that paid more for higher skill levels encour-
aged cross-training by ensuring that when team members had available
time they were motivated to use it productively.
Remember: Employers should ideally provide eight hours of value-
adding tasks each day for all employees. Since this is nearly impossible,
employers need to provide tasks (i.e., cross-training, task lists, etc.) to
make the best use of slow times. If they do not, and if expectations are not
set related to output, most employees will have no recourse but to spend
this time trying to appear busy. Companies also need to have at least two
people trained on each task so they can help during the peak workloads or
fll in when someone is absent.
14.5 teAM leAders IncreAse ProfIt
through hIgher oee
A related strategy that proved to have a high rate of return was making
Tom the team leader. What I learned by designating a team leader was
that, by measuring the OEE beforehand and comparing it to the improved
OEE aferward, I proved that it had been the correct decision based on the
cost versus benefts. Tis proved Toyotas concept of having specialists
(team leaders) available to handle the problems instead of encouraging
everyone to resolve their own issues was the right idea. In addition, we
should not be obsessed with the number of direct versus indirect employ-
ees but look instead at the value each position adds.
Another key point is that a team leader is the frst line of defense in
terms of problem solving. Once Tom became profcient as a team leader, I
was less involved in frefghting and had more time for the check and act
portions of PDCA. Allowing team members to struggle with problems
until they decide they need assistance and only then notify their supervi-
sors (who are already preoccupied with their responsibilities) results in a
company continuously needing extra time allowances to deal with prob-
lems haphazardly. Companies that operate this way are forcing them-
selves to accept about a 70- to 80-percent afectivity level from the entire
organization. Instead, they could be targeting and measuring a level in
the ninety percentiles while providing the support to achieve this (i.e.,
team leaders).
To encourage various improvements within your business, I recom-
mend you quantify or convert what a one-day (or 1-percent) improvement
for certain measurements will add to the bottom line. For example, you
could quantify that a 1-percent OEE improvement for machine A equates
to $X per year, or a one-day reduction in DSI equals $X per year, or a one-
day reduction in DSO equals $X per year. Tis helps put these measure-
ments into terms everyone can relate to and will increase motivation for
improvement as well as keeping the team focused on the improvements
with the largest impact.
14.6 MAnAgIng PArts dIfferently
As we were not fortunate enough to be a widget manufacturer produc-
ing only a few part numbers with subtle diferences, we could not man-
age all parts in the same manner. Terefore, the idea to classify our parts
as either runners, repeaters, or strangers (also known as an ABC
analysis or stock vs. non-stock parts) was very helpful in how we quoted
and manufactured the parts. It is true that we were essentially a job shop
Refections 213
and could not say for certain whether any particular part would receive a
repeat order (although we had a good idea depending on the type of part,
the customer, and the way in which we were asked to prepare the quota-
tion). For example, afer determining a part was likely to repeat (a runner),
we usually did not include a cost for programming (as that might not be
competitive and the programming was a one-time cost). I also required the
shop to document all the setup procedures for potential runner parts so
that if and when they repeated, we could proft from the previous experi-
ence. Quotes for parts that were unlikely to repeat (strangers) included the
cost for developing the program, and usually we would not take the time
to document the setup or prepare a setup fxture. It took some experience
to allow us to classify orders in these groups, but we were right more ofen
than not. And, by having diferent processes for these classifcations, I am
sure we were more proftable. If you design and build your own products
you would categorize and manage parts in a similar fashion, classifying
them as either stock (for runner parts) and non-stock (stranger parts) as
your lead times would difer for these categories.
14.7 ActIvIty-bAsed costIng Proves
helPful In securIng More orders
A critical discovery was the inaccuracy in our prices resulting from using
a single and simplifed shop rate of $60 per hour, thereby evenly allocat-
ing most indirect costs over all products (like spreading butter). Afer
learning the business and taking into account that we had almost as many
indirect as direct employees, the scope of this problem became apparent.
Tere were many benefts realized by understanding most of our actual
costs more accurately in direct relation to the part numbers (or product
families) that consumed these costs. At the forefront was the better under-
standing of our true proft margin, especially when very competitive pric-
ing was required. A simplifed application of activity-based costing was
utilized to reach a range of labor rates that better refected the activity
level required for the various types of parts. However, since the basis for
applying various labor rates are in most cases direct labor hours, it was
critical to frst get these times accurate. We had worked on this through
our feedback cycle with our quotations. Having a system that continued to
refne the accuracy of the direct labor hours was our basis to proceed with
this simplifed version of activity-based costing.
I did not have the time or ability to associate many indirect and overhead
costs (i.e., some indirect labor functions, rent, taxes, etc.) to a specifc part
number, but I had learned many of the characteristics of the parts and/or
customers that afected how much efort/activity was required by each.
From this basis, I was able to redistribute the $60-per-hour fgure much
more appropriately between the shop and ofce depending on the efort/
activity required by the type of product or the particular customer. Tis
breakdown (shown in Appendix J) created a new pricing rangefrom $48
per hour to $74 per hour (still based on direct labor hours)assigned to
a particular part and based on how the characteristics of the part and/or
the customer afected the activity level of our shop and ofce. Tis entire
concept of lean accounting is so critical, especially in competitive markets.
To better relate actual costs to the products consuming them in real time,
apply some of these lean accounting techniques (i.e., activity-based cost-
ing) instead of only compiling numbers each month based on government
requirements (meeting generally accepted accounting principles) and try-
ing to use only this historical data to manage your company.
A fnal element that I feel was critical to my success was basing my prob-
lem resolution on data. Although I was disappointed at not implementing a
more formal problem-solving process, I avoided continuously frefghting
by not taking on every issue and working only on those that repeated most
ofen. I further drove some structure into our discussions by stopping the
team when the conversation turned to it happens a lot and instead asked
that a few weeks of simple data gathering (ofen with a tick list) take place
before we put any more of our time or resources into resolving a particu-
lar issue. I also fought the urge to make a quick determination of what
the root cause was and then direct someone on how to implement my
solution. I cannot claim that I was always successful at guiding my team
through a structured analysis and letting them have input to the solution,
but more ofen than not we resorted to using the lean ideas. Having a team
leader with an understanding of problem solving added to our success.
I would also like to believe that many of the processes and much of
the standardization I put in place remained afer selling the company. I
believe that the documented processes coupled with their links to proft-
ability earned them the right to remain in place through two changes of
management in six years. Te Toyota Production System (TPS) that was
Refections 215
integrated into all Toyotas processes ensured that management changes
were not followed by major process changes.
We all know this is not true in many companies: Te systems and pro-
cesses are not structured with a management system, therefore they are
ofen changed based on the managers ideas and his or her personal style.
Tey are continuously searching for a winner in proftability but willing to
sacrifce other items to get there. Each management change brings with it
diferent priorities, usually diferent systems, beliefs, and defnitely a new
personal style. I wish I knew what has happened to the systems I devel-
oped but I would hope to fnd many still in place. I would also like to
compare my before and afer results having utilized lean to transform the
business versus the new owners sales focus for making improvements and
increasing proftability.
14.8 A Process to exAMIne your busIness
Afer reading this book, I hope you are inclined to verify whether some
of the opportunities presented exist within your company. One way to
start would be to select a product or service you provide that many of
the employees would judge to be difcult (a pain in the butt) and try
to account for all the direct and indirect employees time and the materi-
als and the resources consumed. Tis would include all the time involved
from the frst customer communication through the administrative pro-
cesses, all the way through packing, shipping, invoicing, and collecting the
money. Naturally, you would need to assign hourly labor rates and try to
proportionally account for all the other overheads that cannot be directly
assigned. You should also observe the direct labor hours utilized and com-
pare them to the quoted or standard hours for this product. I feel that by
playing detective with a sample of one or two products (going to Gemba),
you would get a feel for how accurate your estimates are and understand
whether the projected proft margin is realistic. Naturally, your investi-
gation should entail observation of as many of the process steps as pos-
sible, usually resulting in the identifcation of many Kaizen opportunities.
Tis should lead you into undertaking OSKKK, and provide a starting
point. Remember to prioritize actions for the processes that you consider
bottlenecks. Tis would also be a good time to review your defnition of
a bottleneck and how your team determines where they are currently
located and how to react to them.
Tis is only scraping the surface, but I would be surprised if you did
not come away identifying many opportunities and engaging fellow team
members in discussions that will also leave them considering the oppor-
tunities. You will likely come away understanding some shortfalls in your
quoting process, planning process, and value-adding processes as well
as in the accounting allocations toward determining your proft margin.
Before you go of relating these discoveries to wider implications for the
business and jumping into some program or action plan, you need to
think about how this might take focus away from current business initia-
tives and you end up with a machine gun Kaizen approach. Terefore, I
am only recommending that you consider this step as a starting point to
help convince yourself and the organization of the opportunities. Ten,
you need a process such as strategy deployment to help in prioritizing
your limited resources and aligning your improvement tasks to the cur-
rent business priorities.
Lean principle: Machine gun Kaizen is a negative term that describes
how you should not go about your Kaizen implementation. It is the idea
of individuals randomly performing many smaller Kaizens without
a plan linking them to the overall strategy or the metrics. Tis can be
detrimental in utilizing your limited resources on incorrect priorities.
OSKKK helps in avoiding machine gun Kaizen, but deciding where to
implement OSKKK must be linked to the companys strategy and posi-
tively afect the key metrics.
Another thought for examining your own business: Look at the tasks I
went about learning and improving during my lean transformation (listed
in Appendix L), and for those same tasks that exist within your business,
you should question which are standardized and have previously had their
fow and processes Kaizened. You can then prioritize which of these most
directly infuence your proftability or cause your bottlenecks, and then
begin to improve them possibly utilizing some of the improvement ideas I
discovered during my transformation.
Refections 217
Since the purpose of this book was to tell my story as it happened and not
to recommend a specifc strategy or an exact method to implement lean
in a high-mix, low-volume environment, you should use it to draw your
own conclusions. At this point, I refer you to my book, Made-to-Order
Lean, in which I share some ideas and a methodology to implement lean in
this type of environment. I also encourage you to search for publications
containing experiences of other lean practitioners implementing in these
specialized environments.
While I am out working with a variety of manufacturing and ser-
vice industry clients, I continuously fnd new ways to adapt what I
know as the Toyota Production System. As these principles are con-
tinually learned and explored, I am always amazed by the number of
new opportunities discovered and the creative solutions that are found.
My strongest recommendation is not to get hooked on the lean tools
presented in most books (as they are usually taken from high-volume
examples) but instead to explore the underlying principles and adapt
them to your situation.
I wish you success in your journey of continuous improvement.
219
Appendix A
OSKKK Methodology
1. observAtIon
For all operations
Take the time to see what is happening in the work area by watching multiple cycles
of the same process.
Watch more than one person performing the process and note where
standardization is lacking, especially where it afects quality or productivity.
Document by writing the individual process steps in the sequence they occur.
Identify the origins of variation in both fow of information and fow of materials.
220 AppendixA
2. stAndArdIzAtIon
For materials, motions, tasks, and management; defned as the current optimized
sequence of the process steps followed by the team members to ensure quality,
safety, and productivity
Prioritize where standardization is most critical to the organization based on
observations or data.
Observe all team members various methods for performing the task and decide on
the current standard (based on that which is best for quality, safety, and
productivity).
Have all team members work to the current standard while improvement ideas are
being discussed.
Input 5S (workplace organization) to promote simplifed and productive
standardization to be put in place.
Work to minimize/eliminate problems and interruptions to the process using
problem-solving techniques.
Ensure all fows and decision points in the process have a standard methodology.
3. kAIzen of floW & Process
Information and materials frst; least cost to implement ($)
Understand and map process fow (process mapping).
Understand and map material fow (value stream mapping).
Improve fow of material and information to the work area.
Identify all non-value adding time in both information and process fows, then
work to eliminate or minimize it.
In business processes and material fows, work to reduce throughput time.
AppendixA 221
4. kAIzen of equIPMent
Medium cost to implement ($$)
Look at setup time (SMED; single minute exchange of dies) and work to reduce it.
Look to improve feeds and speeds (reduce machine cycles).
Use OEE (overall equipment efectiveness) as the measurement to drive
improvements.
Understand operator workload in comparison to machine cycle times, eliminate
forced waiting.
Look for total predictive maintenance (TPM) improvements.
Look to simplify machines.
5. kAIzen of lAyout
Most cost to implement ($$$)
Ensure previous steps of OSKKK have been worked on before rearranging layout.
Collect data: process fows, capabilities, new products, bottlenecks, OEE, value
stream maps, information fow, etc.
Minimum of three layout proposals, all showing fow arrows and complete
correlation matrixes (details found in Made-to-Order Lean).
New layouts to consider all lean principles and remove non-value added work to
justify costs.
Should work to improve the manmachinematerials ratio.
Any new machinery should be considered only afer working to improve existing
machines and incorporating all learnings into specifcations for the new machinery.
Simulate the new layout either marking the foor (for existing layouts) or with
cardboard mock-ups of the equipment (for new processes or products).
223
Appendix B
Documentation Techniques
Figure B.1 Always start with observation.
224 AppendixB
Figure B.2 Next, use digital camera to document shop foor tasks.
Figure B.3 Videos with a running commentary are an easy backup plan for both the
shop foor and ofce.
Figure B.4 Process maps utilized for ofce tasks.
AppendixB 225
Figure B.5 For computer tasks, use screenshots with clear instructions.
227
Appendix C
My First Attempt at Standardizing
the Quoting Process
Quote & Tracking Sheet: Machined Parts Quote #: 3341
Customer: ABC Inc. Part: 713-224981-B Qty: 16 Due date: 3/7/02
MATERIAL CNC PROGRAMMING
Material estimate 1 $7.02
ea.
Supplier
name
Reliance Est.
(hrs.)
Act. Charge/
overhead
Material estimate 2 $6.80
ea.
Supplier
name
Jorgenson Op. 1 0.33
Material estimate 3 Supplier
name
Op. 2 0.25
Op. 3 0.25
Enter hourly rate: $60 Op. 4
Any add. markup: $0.00 Op. 5
OPERATION 1 Total 0
Est. Act. Qty. Cost/pc. Problems
Setup time (hrs.) 0.33 16 $1.24
Run (minutes) 3 16 $3.00
Extra measure,
deburr, fxture, etc.
OPERATION 2
Setup time (hrs.) 0.25 16 $0.94
Extra measure,
2 16 $2.00
228 AppendixC
OPERATION 3
Setup time (hrs.) 0.5 16 $1.88
Extra measure,
deburr, fxture, etc
2 $2.00
OPERATION 4
Setup time (hrs.)
Run (minutes)
Extra measure,
OPERATION 5
Setup time (hrs.)
Run (minutes)
Extra measure,
Subcontract Lot
price
per pc. Total (w/15%
markup)
Plating estimate $55.00 $3.44 $3.95 Total per part: $31.81
Painting estimate Any 1x setup
Silkscreen Expedite charge
Clean room pkg.
Stress relieve Note: Actual times noted during mfg. in areas
229
Appendix D
Example of the First Day-by-hour
Board for the Bottleneck Machine
VF-4 ScheduIe
Figure D.1
231
Appendix E
Real-time Pareto Chart
MachineBBBBBBB Week BBBBBBBBB
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
No Tool
Available
TPM or
cleaning
chips
Replacing
xtures
Machine
breakdown
Missing info.
works order
package
Program
loading
problem
Program
(code)
problem
Program
required
modication
Long
adjustment
time
Wait in excess
5 minutes
quality chk.
N
u
m
b
e
r

o
f

O
c
c
u
r
r
e
n
c
e
s
Short holder 30 min. good pt.
45 min. good pt.
Electronics 1 hr.VF-4
Feb 3-7
Figure E.1
233
Appendix F
Overall Equipment Effectiveness
(OEE) for Bottlenecks
PArt I
OEE is a useful measurement because it is based on the premise that all
losses on machines can be quantifed. It is not possible to obtain an OEE
of 100 percent because all losses are counted (includes considering setup
time as a loss). Te idea of looking at this cumulatively is that, even if
all the individual numbers (availability, performance, and quality) are 90
percent (which might seem acceptable):
90% 90% 90% = 73%
therefore, a good part is coming of this machine only 73 percent of the
time.
OEE = availability (%) performance (%) quality (%)
(ideal cycle time
= operating time total parts) good parts produced
planned production
operating time
total parts produced

time
234 AppendixF
When substituting, this can be mathematically simplifed to:
OEE = (Good pieces ideal cycle time)/planned production time
I recommend using this simplifed equation to start with; in addition, you
should separately calculate a quality number (which is likely already being
measured for other purposes). Part of the reason for the simplifed calcula-
tion is that some organizations struggle with how to separate downtime
into either availability or performance. An easy rule of thumb: the longer
downtimes (i.e., greater than fve minutes) are availability issues and the
shorter downtimes (less than fve minutes) are performance. Tough
ofen there are few benefts derived from this efort, more importantly is
for each bottleneck machine to develop specifc categories for downtime
and track those while calculating OEE with the simplifed equation.
Here is an example of how I have gone about doing it with the operators
using the simplifed calculation each day:
OEE: Calculation Week: March 2-6 Machine: VF-4
Date Shif
Part
number/
operation
Ideal cycle
time
(CNC-
prog.)
(minutes)
Qty. Good
pieces
(quantity)
Productive
time =
(good
pieces
ideal cycle
time)
(minutes) Quality P
l
a
n
n
e
d

p
r
o
d
u
c
t
i
o
n

t
i
m
e

(
m
i
n
u
t
e
s
)
Total
OEE (%)
2-Mar First
716-
774112
op 1
3.5 16 56
76/81 =
93.8%
7.5 hours
X 60 min
= 450
min
273.8/450
= 60.8%
716-
774112
op2
4.8 16 76.8
716-
774112
op 3
108 16 28.8
304557 op
1
4.3 8 34.4
304557 op
2
2.6 8 20.8
440-
322109
op 1
5.1 6 30.6
440-
322109
op 2
4.4 6 26.4
Total 76 273.8
AppendixF 235
PArt II
In addition to calculating the OEE each day (directly at the machine), the
operator also flls out a real-time Pareto chart (categories developed spe-
cifcally for each bottleneck machine). Tis should provide a baseline for
where to target your improvements.
3-Mar First
554690 op
1
9 12 108
58/60 =
96.6%
7.5 hours
X 60 min
= 450
min
344.8/450
= 77%
554690 op
2
4 12 48
713-
770321
op 1
4 19 76
713-
770321
op 2
4 19 76
988431 op
1
4.6 8 36.8
Total 70 344.8
236 AppendixF
Su .
250 25
240 24
230 23
220 22
210 21
200 20
190 19
180 18
170 17
160 16
150 15
140 14
130 13
120 12
110 11
100 10
90 9
80 8
70 7
60 6
50 5
40 4
30 3
20 2
10 1
M
a
c
h
i
n
e

b
r
e
a
k
d
o
w
n
Tuesday
good/to-
tal parts
stopped
> 5 min.
Availability
stopped
5 min.
<
Availability Performance Performance Quality
OEE Machine :
Monday
T
i
m
e

l
o
s
t

i
n

m
i
n
u
t
e
s
Q
u
a
n
t
i
t
y

o
f

p
i
e
c
e
s

f
o
r

Q
u
a
l
i
t
y
stopped
> 5 min.
stopped <
5 min.
good/to-
tal parts
Quality
Week: Mar. 2-6 VF-4
S
e
t
-
u
p

t
i
m
e

p
l
a
n
n
e
d

(
q
u
o
t
e
)
S
e
t
-
u
p

t
i
m
e

p
l
a
n
n
e
d

(
q
u
o
t
e
)
T
o
t
a
l

a
t
t
e
m
p
t
e
d

p
a
r
t
s
R
e
w
o
r
k

p
a
r
t
s
S
c
r
a
p

p
a
r
t
s
R
u
n
n
i
n
g

a
t

s
l
o
w
e
r

c
y
c
l
e
A
d
j
u
s
t
m
e
n
t
s
S
m
a
l
l

s
t
o
p
s
C
l
e
a
n
i
n
g

m
a
c
h
i
n
e
/
p
a
r
t
T
o
l
e
r
a
n
c
e

p
r
o
b
l
e
m
F
i
x
t
u
r
i
n
g

p
r
o
b
l
e
m
M
a
c
h
i
n
e

b
r
e
a
k
d
o
w
n
T
o
o
l
i
n
g

p
r
o
b
l
e
m
s
W
a
i
t
i
n
g

f
o
r

t
o
o
l
i
n
g
/
p
r
o
g
r
a
m
T
P
M
T
o
t
a
l

a
t
t
e
m
p
t
e
d

p
a
r
t
s
R
e
w
o
r
k

p
a
r
t
s
S
c
r
a
p

p
a
r
t
s
R
u
n
n
i
n
g

a
t

s
l
o
w
e
r

c
y
c
l
e
A
d
j
u
s
t
m
e
n
t
s
S
m
a
l
l

s
t
o
p
s
C
l
e
a
n
i
n
g

m
a
c
h
i
n
e
/
p
a
r
t
T
o
l
e
r
a
n
c
e

p
r
o
b
l
e
m
F
i
x
t
u
r
i
n
g

p
r
o
b
l
e
m
T
o
o
l
i
n
g

p
r
o
b
l
e
m
s
W
a
i
t
i
n
g

f
o
r

t
o
o
l
i
n
g
/
p
r
o
g
r
a
m
T
P
M
S
e
t
-
u
p

t
i
m
e

a
c
t
u
a
l
S
e
t
-
u
p

t
i
m
e

a
c
t
u
a
l
Figure F.1
237
Appendix G
Typical Team Leader Responsibilities
A team leadergenerally promoted from within the teamis an hourly
employee who has many of the following responsibilities:
Provides assistance to primary operator
Answers calls for assistance
Troubleshoots tooling problems
Determines when maintenance is necessary and who should
respond (whether the team leader can handle or more technical
support is necessary)
Performs changeovers (prepares tools, material, etc.)
Ensures material including consumables is available and reor-
dered as required
Checks quality
Handles defective material
Measures and tracks results
Provides emergency relief
Provides coverage for operator absenteeism and vacation
Follows and procures material when needed
Available for overtime and helping administer overtime
Assists area supervisor to ensure proper rotation concepts
Performs training and maintains training documentation (updates
cross-training/skills matrix)
Supports monitoring and improving OEE (overall equipment
efectiveness)
Performs quality audits
238 AppendixG
Oversees TPM (total predictive maintenance) and monitors confor-
mance to TPM procedures/schedules
Provides input to root cause analysis
Provides documentation for machine problems/downtime causes
Assists supervisor to ensure proper scheduling of production (sched-
ule day-by-hour boards)
Assists supervisor to maintain department supplies and indirect
materials
239
Appendix H
Cross-Training or Skills Matrix
Manual & CNC Machining Matix
G G G G G G G G G G Y Y G G G Y Y R R Y R Y
L
o
a
d
/
u
n
l
o
a
d

a
n
d

b
a
s
i
c

d
e
b
u
r
r
i
n
g
B
a
s
i
c

m
e
a
s
u
r
i
n
g

w
i
t
h

c
a
l
i
p
a
r

&

m
i
c
r
o
m
e
t
e
r

U
n
d
e
r
s
t
a
n
d
i
n
g

o
f

t
o
l
e
r
a
n
c
i
n
g

&

d
e
t
e
r
.

o
u
t

o
f

t
o
l
e
r
a
n
c
e
A
b
l
e

t
o

m
a
k
e

m
i
n
o
r

o
s
e
t

a
d
j
u
s
t
m
e
n
t
s

i
n

m
a
c
h
i
n
e
S
e
t
-
u
p

o
f

t
o
o
l
i
n
g
S
e
t
-
u
p

o
f

o
s
e
t
s

f
o
r

t
o
o
l
i
n
g

i
n

m
a
c
h
i
n
e

S
e
t
t
i
n
g

u
p

x
t
u
r
e
s
P
i
c
k
i
n
g

u
p

0

o
s
e
t
s
L
o
a
d

p
r
o
g
r
a
m

f
r
o
m

c
o
m
p
u
t
e
r

l
i
b
r
a
r
y

t
o

m
a
c
h
i
n
e
U
n
d
e
r
s
t
a
n
d

a
n
d

c
o
r
r
e
c
t

G
-
c
o
d
e

i
n

p
r
o
g
r
a
m

(
l
i
n
e
-
e
d
i
t
)
A
b
l
e

t
o

c
h
a
n
g
e

d
i
a
m
e
t
e
r

(
g
e
o
m
e
t
r
y
)

o
s
e
t
s
U
t
i
l
i
z
e

a
l
l

m
e
a
s
u
r
i
n
g

e
q
u
i
p
.

i
n

Q
.
C
.

a
r
e
a
M
a
n
u
a
l

&

o
-
l
i
n
e

t
a
p
p
i
n
g
M
a
n
u
a
l

l
a
t
h
e

-

s
e
t
-
u
p
B
a
s
i
c

p
r
o
g
r
a
m
m
i
n
g
O
p
e
r
a
t
e

m
a
n
u
a
l

m
i
l
l
i
n
g

m
a
c
h
i
n
e
S
e
t
-
u
p

m
a
n
u
a
l

m
i
l
l
i
n
g

m
a
c
h
i
n
e
S
e
t

&

r
u
n

2
-
D

a
n
g
l
e
s

o
n

m
a
n
u
a
l

m
i
l
l
i
n
g

m
a
c
h
i
n
e
C
o
n
c
e
p
t
u
a
l
i
z
e

a
n
d

d
e
v
e
l
o
p

3
-
D

p
r
o
g
r
a
m
m
i
n
g
E
s
t
i
m
a
t
e

s
e
t
-
u
p

t
i
m
e

E
s
t
i
m
a
t
e

r
u
n

t
i
m
e

(
m
a
c
h
i
n
e

c
y
c
l
e

t
i
m
e
)
E
n
g
r
a
v
i
n
g

o
f

p
a
r
t

n
u
m
b
e
r
s
Employee
Dates for training/notes
1 Tom
2 John
3 Jake
4 Bert By August 2001
5 Julian TBD
6 Alfred By June 2001
7 Conney
8 Kenny
9
Greg on-going
G
Y
R
Low Risk
Medium Risk
High Risk
Status of cross training
Documented Procedure
Key
-Trainee, requires
assistance & vigilance to
perform this task
-Able to perform majority of
task alone
-Able to perform task and
update documents
-Able to train others, take
measurements, and modify
standardized procedures
- Planning to
train to this
level in the
future
G G G G G G G G G G Y Y G G G Y Y R R R Y Y
Figure H.1
241
Appendix I
New Layout with Some CNCs
Set Up as an L Cell
Figure I.1
243
A
p
p
e
n
d
i
x

J
I
n
i
t
i
a
l

A
B
C

C
o
s
t
i
n
g

M
a
t
r
i
x

f
o
r

S
h
o
p

a
n
d

O
f
f
c
e
P
A
r
t

I
S
h
o
p

A
c
t
i
v
i
t
y

R
a
t
i
n
g
(
+
1
)

S
o
f

M
a
t
e
r
i
a
l

(
a
l
u
m
i
n
u
m
,

p
l
a
s
t
i
c
,

e
t
c
.
)
(
+
2
)

H
a
r
d

M
a
t
e
r
i
a
l

(
s
t
e
e
l
,

s
t
a
i
n
l
e
s
s
,

t
u
n
g
s
t
e
n
,

e
t
c
.
)
E
a
s
y

D
e
b
u
r
r

(
+
1
)
H
e
a
v
y

D
e
b
u
r
r

(
+
2
)
E
a
s
y

D
e
b
u
r
r

(
+
1
)
H
e
a
v
y

D
e
b
u
r
r

(
+
2
)
N o t i g h t
t o l e r a n c e s
( + 0 )
F e w t i g h t
t o l e r a n c e s
( + 1 )
S i g n i f c a n t
t i g h t t o l e r -
a n c e s ( + 2 )
N o t i g h t
t o l e r a n c e s
( + 0 )
F e w t i g h t
t o l e r a n c e s
( + 1 )
S i g n i f c a n t
a n c e s ( + 2 )
N o t i g h t
t o l e r a n c e s
( + 0 )
F e w t i g h t
t o l e r a n c e s
( + 1 )
S i g n i f c a n t
a n c e s ( + 2 )
N o t i g h t
t o l e r a n c e s
( + 0 )
F e w t i g h t
t o l e r a n c e s
( + 1 )
S i g n i f c a n t
a n c e s ( + 2 )
(
+
1
)

E
a
s
y

m
e
a
s
u
r
e
m
e
n
t
(
1
3
0

m
e
a
s
u
r
e

p
o
i
n
t
s
)
E
a
s
y

t
o

p
r
o
g
r
a
m

(
+
1
)
1

+

1

+

1

+

1

+

0

=

4
1

+

1

+

1

+

1

+

1

=

5
1

+

1

+

1

+

1

+

2

=

6
1

+

1

+

1

+

2

+

1

=

5
1

+

1

+

1

+

2

+

2

=

6
1

+

1

+

1

+

2

+

1

=

7
1

+

1

+

2

+

1

+

0

=

5
1

+

1

+

2

+

1

+

1

=

6
1

+

1

+

2

+

2

+

0

=

7
1

+

1

+

2

+

2

+

0

=

6
1

+

1

+

2

+

2

+

1

=

7
1

+

1

+

2

+

2

+

2

=

8
M
e
d
i
u
m

t
o

p
r
o
g
r
a
m

(
+
2
)
5
6
7
6
7
8
6
7
8
7
8
9
D
i
f
c
u
l
t

t
o

p
r
o
g
r
a
m

(
+
3
)
6
7
8
7
8
9
7
8
9
8
9
1
0
244 AppendixJ
(
+
2
)

M
e
d
i
u
m

m
e
a
s
u
r
e
m
e
n
t
(
3
1
5
0

m
e
a
s
u
r
e

p
o
i
n
t
s
)
E
a
s
y

t
o

p
r
o
g
r
a
m

(
+
1
)
5
6
7
6
7
8
6
7
8
7
8
9
M
e
d
i
u
m

t
o

p
r
o
g
r
a
m

(
+
2
)
6
7
8
7
8
9
7
8
9
8
9
1
0
D
i
f
c
u
l
t

t
o

p
r
o
g
r
a
m

(
+
3
)
7
8
9
8
9
1
0
8
9
1
0
9
1
0
1
1
(
+
3
)

D
i
f
c
u
l
t

m
e
a
s
u
r
e
m
e
n
t

(
5
1
+

m
e
a
s
u
r
e

p
o
i
n
t
s
)
E
a
s
y

t
o

p
r
o
g
r
a
m

(
+
1
)
6
7
8
7
8
9
7
8
9
8
9
1
0
M
e
d
i
u
m

t
o

p
r
o
g
r
a
m

(
+
2
)
7
8
9
8
9
1
0
8
9
1
0
9
1
0
1
1
D
i
f
c
u
l
t

t
o

p
r
o
g
r
a
m

(
+
3
)
8
9
1
0
9
1
0
1
1
9
1
0
1
1
1
0
1
1
1
2
c
o
n
v
e
r
s
i
o
n

o
f

s
h
o
p

A
c
t
i
v
i
t
y

r
a
t
i
n
g

t
o

h
o
u
r
l
y

c
o
s
t
S
h
o
p

A
c
t
i
v
i
t
y

R
a
t
i
n
g
4
5
6
7
8
9
1
0
1
1
1
2
H
o
u
r
l
y

S
h
o
p

C
o
s
t
$
1
6
$
1
8
$
2
0
$
2
2
$
2
4
$
2
6
$
2
8
$
3
0
$
3
2
AppendixJ 245
P
A
r
t

I
I
O
f
c
e

A
c
t
i
v
i
t
y

R
a
t
i
n
g
(
+
1
)

E
a
s
y

C
u
s
t
o
m
e
r

(
o
n
-
t
i
m
e

p
a
y
,

g
o
o
d

t
e
r
m
s
,

e
t
c
.
)
(
+
2
)

D
i
f
c
u
l
t

C
u
s
t
o
m
e
r

(
n
o
t

o
n
-
t
i
m
e

p
a
y
,

p
o
o
r

t
e
r
m
s
,

e
t
c
.
)
(
+
1
)

E
a
s
y

s
h
i
p

&

p
a
c
k

r
e
q
u
i
r
e
m
e
n
t
s
(
+
2
)

D
i
f
c
u
l
t

s
h
i
p

&

p
a
c
k

r
e
q
u
i
r
e
m
e
n
t
s
(
+
1
)

E
a
s
y

s
h
i
p

&

p
a
c
k

r
e
q
u
i
r
e
m
e
n
t
s
(
+
2
)

D
i
f
c
u
l
t

s
h
i
p

&

p
a
c
k

r
e
q
u
i
r
e
m
e
n
t
s
(
+
1
)

M
i
n
i
m
a
l

o
r

s
t
a
n
d
a
r
d

c
u
s
t
o
m
e
r

s
u
p
p
o
r
t

r
e
q
u
i
r
e
m
e
n
t
s

(
f
e
w

c
h
a
n
g
e
s
,

e
a
s
y

s
a
l
e
s

s
u
p
p
o
r
t
,

s
t
a
n
d
a
r
d

s
p
e
c
i
f
c
a
t
i
o
n
s
,

q
u
a
n
t
i
t
y
,

e
t
c
.
)
(
+
1
)

S
h
o
r
t

b
i
l
l

o
f

m
a
t
e
r
i
a
l

(
1
5

i
t
e
m
s
)

+

n
o

o
u
t
s
o
u
r
c
i
n
g
1

+

1

+

1

+

1

=

4
1

+

1

+

1

+

2

=

5
1

+

1

+

2

+

1

=

5
1

+

1

+

2

+

2

=

6
(
+
2
)

M
e
d
i
u
m

b
i
l
l

o
f

m
a
t
e
r
i
a
l

(
6
1
5

i
t
e
m
s
)

+

o
u
t
s
o
u
r
c
i
n
g

1
2

s
e
r
v
i
c
e
s
5
6
6
7
(
+
3
)

L
o
n
g

b
i
l
l

o
f

m
a
t
e
r
i
a
l

(
1
6
+

i
t
e
m
s
)

+

o
u
t
s
o
u
r
c
i
n
g

3

o
r

m
o
r
e

s
e
r
v
i
c
e
s
6
7
7
8
(
+
2
)

I
n
c
r
e
a
s
e
d

c
u
s
t
o
m
e
r

s
u
p
p
o
r
t

r
e
q
u
i
r
e
m
e
n
t
s

(
f
r
e
q
u
e
n
t

c
h
a
n
g
e
s
,

e
x
t
r
a

s
a
l
e
s

s
u
p
p
o
r
t
,

a
d
d
i
t
i
o
n
a
l

s
p
e
c
i
f
c
a
t
i
o
n
s
,

q
u
a
n
t
i
t
y
,

e
t
c
.
)
(
+
1
)

S
h
o
r
t

b
i
l
l

o
f

m
a
t
e
r
i
a
l

(
1
5

i
t
e
m
s
)

+

n
o

o
u
t
s
o
u
r
c
i
n
g
5
6
6
7
(
+
2
)

M
e
d
i
u
m

b
i
l
l

o
f

m
a
t
e
r
i
a
l

(
6
1
5

i
t
e
m
s
)

+

o
u
t
s
o
u
r
c
i
n
g

1
2

s
e
r
v
i
c
e
s
6
7
7
8
(
+
3
)

L
o
n
g

b
i
l
l

o
f

m
a
t
e
r
i
a
l

(
1
6
+

i
t
e
m
s
)

+

o
u
t
s
o
u
r
c
i
n
g

3

o
r

m
o
r
e

s
e
r
v
i
c
e
s
7
8
8
9
c
o
n
v
e
r
s
i
o
n

o
f

o
f
f
c
e

A
c
t
i
v
i
t
y

r
a
t
i
n
g

t
o

h
o
u
r
l
y

c
o
s
t
O
f
c
e

A
c
t
i
v
i
t
y

R
a
t
i
n
g
4
5
6
7
8
9
H
o
u
r
l
y

O
f
c
e

C
o
s
t
$
1
2
$
1
4
$
1
6
$
1
8
$
2
0
$
2
2
247
Appendix K
Initial Equipment and Area Layout
for Expansion into Sheet Metal
Figure K.1
249
Appendix L
Sequence Utilized for Learning the
Business and for Implementing Lean
s
h
o
P
OSKKK
Tasks
O
b
s
e
r
v
e

(
D
o
c
u
m
e
n
t
)
S
t
a
n
d
a
r
d
i
z
a
t
i
o
n
K
a
i
z
e
n

o
f

F
l
o
w

&

P
r
o
c
e
s
s
K
a
i
z
e
n

o
f

E
q
u
i
p
m
e
n
t
K
a
i
z
e
n

o
f

L
a
y
o
u
t
CNC machine operation 1 6 42, 84 66, 85 78, 86
CNC machine setup 3 7 45, 55, 61 66, 86
CNC programming 7 10 11, 83, 84
Workplace organization (5S) 4 5 44 78
Interviewing 8 9
Machine TPM 59 60
Identify shortage of skills 14 15
Actual vs. estimated times 19 20 21
Problem-solving
development
48
Team leader development 49 50, 57 64, 67
Overtime management 53 54
Bonus system/suggestion
system
56 64
Machine productivity
improvements
63 78
External quality accreditation 77
Stocking of tooling,
consumables
82
Shearing/bending sheet
metal
87 88 90 92
Welding
250 AppendixL
Key: 1 through 92the sequential order in which I learned and implemented changes; A through
Vthe sequential order in which my mother learned and documented the ofce tasks.
o
f
f
I
c
e
Quoting machined parts 2 12 13, 28,
52, 58,
74, 79
Quoting sheet metal parts 89 91
Input customer orders C, 22 G 23, 79
Setting hourly rates for
customers
16 17 18, 43
Pricing for materials F, 29 72 73
Purchase orders A, 30 69
Purchasing/receiving job
materials
B, 31 I 32, 70,
71, 81
Payroll E
Invoicing D
Planning (capacity) 25 26 27, 34,
51, 80
Planning (materials) 24 33, 80
Bookkeeping H
Monthly closing of the books M, 38 O, 39 P, 40, 65
Packaging K N 78
Organizing shipment J 78
Producing the work order
package
35 36 37
Collecting customer
payments
Q, 41 R
Standardization through
visualization
75 78
Workplace organization (5S) S T U 78
Computer system/network 68, 76, V
251
Appendix M
Different Applications of Value Stream
Mapping and Process Mapping
A high level snapshot from door to door of a business
Follows material ow & some information ow (order entry)
Helps to identify improvement opportunities
Does not incorporate decision loops or parallel processes
One high-level step can become
a complete process map
Verify all new orders
in MRP system
PT = 5 min.
WT = 1 day
C&A = 46%
s the item a
kanban part ?
Determine why it is
still showing in MRP
system and not
being managed as a
kanban part
PT = 15 min.
WT = 5 day
Change to kanban
part
Yes
Kaizen idea: Ensure all
kanban parts are removed
from the MRP system
Process Mapping
C&A = 70%
Verify any
questionable
delivery dates
PT = 3 min.
WT = 3 day
C&A = 60%
s date past due or
impossible to meet the
delivery?
Yes
Contact sales to
reschedule with
customer
PT = 5 min.
WT = 4 day
List possible orders
to be rescheduled
PT = 20 min.
WT = 1 day
C&A = N/A
Kaizen idea: Sales to
check system for available
capacity during requested
period.
A detailed map showing all
activities in either a business
process or manufacturing process
Usually follows information ow
Helps to identify improvement
y
C&A = 82%
Enter order into
production schedule
based on delivery
date & material lead
times.
PT = 2 min.
WT = 0 day
C&A = N/A
Transfer "bill of
material items into
purchase requisition
system
PT = 2 min.
WT = 10 day
C&A = 75%
Does the new
production
schedule exceed
capacity ?
Yes
Can capacity be
increased ?
Yes
Request production
to increase capacity.
PT = 10 min.
WT = 1 day
C&A = N/A
C&A = N/A
ssue new schedule.
PT = 20 min.
WT = 0 day
C&A =85%
opportunities
Encourages quantifying the
following: % of time incoming
information is Correct and Accurate,
Waiting Time and Process Time
Incorporates decision loops and
parallel processes
Value Stream Mapping
N
o
N
o
N
o
N
o
Figure M.1
253
Index
A
ABC, see Activity-based costing (ABC)
methodology
Abnormal activities identifcation
initial take over, 5254
visualizing, 35, 9697
Absenteeism enforcement, 74, 9899, 175,
see also Termination
Accents, customer impact, 34
Access sofware, 102
Accounts receivable, 4344
Accreditation, 101, 110111
Accrual system, 42, see also Financial
processes
Acronyms use, xv
Act, see PDCA (Plan, Do, Check, Act)
Activity-based costing (ABC)
methodology
accurate allocations, slow time,
129130
analysis, 61
hourly costs, 136138
lessons learned, 212
matrix, 132138, 243245
pricing impact, 130132
refections, 213215
setup reductions, 83
Adjustment phase, changeover, 8384
Administrative processes, see also Ofce
processes
computer drives and fles, 109110
5S method application, 109110
focusing on, 100102
fundamentals, 9596, 111
historical data issues, 9697
late start impact, 9798
purchasing, 102107
quality accreditation, 110111
quotations, 107109
recruiting process, 98100
visual management, 9697
warm-up preparation, 9798
Advertising, selling the business, 197
Asking price, selling the business,
195196
Assembly rates, 19
Assumptions, 2425
A3 method
normal vs. abnormal, 53
patterns, tracking issues, 171
problem solving, 55
Attendance, see also Absenteeism
enforcement
bonus system, 87
paycheck incentive, 41
Auctions, 179180, 184
Automatic replenishing systems, 153
Autonomy, see Daily operations
independence
Awareness, 51
b
Background developments, xiiixxiv, xvii
Baseline improvements, 13, 15
Bingo (acronyms use), xv
Blanket purchase orders, 104, 154
Boat life, xviii
Bonus system
announcement, 68
difculties creating, 8688
Bottlenecks, see also Overall equipment
efectiveness (OEE)
available materials planning, 147148
band saws, 8990
day-by-hour board example, 228
lead times, 149
new layout, 116117
ofce improvements, 101
254 Index
OSKKK, 162, 215216
sequence, 249250
Business, initial take over
abnormal activities, 5254
extra capacity creation, 4951
frefghting, 55
implementing lean, 249250
learning sequence, 249250
normal activities, 5254
pricing decision, 4849
problem-solving, 55
productivity, proft impact, 5860
proft impact, 5860
proft margins, challenging, 6062
5S organization, 4951
visuals, shop-foor, 5558, 229
Business, moving location
changes, introducing, 121125
design, new layout, 116118
downturn time, 113114
fundamentals, 125
justifcation, 114116
layout, 116118, 120121
non-disruption in business, 118120
ofce changes, 123125
procedural changes, 121123
simulation, new layout, 120121
transition phase, 118120
Business, purchase of
background developments, xvii
boat life, xviii
closing, xxiv
CNC machining, xxii
entrepreneurial spirit, xviixviii
fnancing, xviii, xxiii
goodwill, xxiv
incorporation, xxiiixxiv
location importance, xx
minority-owned quota, xx
possible business considerations, xix
selected business, xxixxii
Small Business Administration (SBA),
xviii, xxiii
terms of sale, xxii
valuation methods, xx, xxiii
Business, selling of
advertising, 197
asking price determination, 195196
buyers and ofers, 200202
epilogue, 204205
evaluating ofers, 201202
fundamentals, 193194
inquiries, 197198
lower price, 199200
marketing package, 196
ofer accepted, 202204
owner fnancing, 198199
process steps for, 194195
valuation, 195196
Buyers and ofers, 200202
c
Cameras, digital, 13, 224
Capabilities
cost allocation pricing, 141
expansion, 169170
outsourcing due to lack, 75
Capacity, creating and selling
continuous measurements, 6770, 231,
233236
estimation, 7377
full-absorption costing thinking, 6364
fundamentals, 77
initial estimation alone, 7477
overtime, 6465
PDCA, estimating process, 7377
productivity, 7173
reductions, 6567
setup times, 6570
team leader position, 7071, 237238
Capacity planning
adding a second shif, 166169
available materials, 147148
materials planning contrast, 38
pilot board, 58
short-term planning, 39
upfront delays, 148149
Casual interest, avoiding, 198
Challenging the team, 164166
Changeovers
adjustment phase, machine output,
8384
Index 255
nonvalue added time, 51
pilot board, 57229
Changes
ofce, 123125
procedural, 121123
understanding processes frst, 5, 11
Check, see PDCA (Plan, Do, Check, Act)
Checklists, see Tick lists
Closing, business purchase, xxiv
Closing of books, 4143, 9394, see also
Financial processes
CNC (Computer Numerical Control)
machining
defned, xvi
delay waste, 146
documentation, 1314
dry erasable boards for, 5658, 229
estimating, 1719
labor estimate, 19
L cell, 19
new layout, 117
outdated and specialized equipment
impact, 20
purchase background, xxii
purchase of larger machine, 178
quality, 8, 85
setup preparation, 34
skill levels, 910
startup, 82
time wasted, 50
U cell, 1920
work order package, 40
writing, 154
Collection methods, 4344
Competitors, 143
Compliance vs. accredited, 111
Complications, larger organizations, 2
Computer Numerical Control, see
CNC (Computer Numerical
Control) machining
Computers, see also Sofware
administrative processes, 109110
naming standardization, 11, 109
ofce improvements, 101102
Connections, utility, 180
Consulting work, 81, 194
Consumable Kanban, 153, 184
Continuous measurements, 6770, 231,
233236
Conventional wisdom, OSKKK method,
23
Cost allocation, pricing
activity-based costing, 130138,
243245
capabilities, increasing, 141
cost linking to matrix, 134, 136138
daily operation independence, 142143
matrix, 132136, 243245
other methods of allocation, 138140
outsourcing reduction, 141
pursuing opportunities, 142143
slowdown impact, 129130, 140141
standardization benefts, 142143
team leader position, 140141
Costs
customer focus, 23
design experience, 157158
parts charge, 32
product design impact, 7, 156
tolerance designs, 157158
tooling estimates, 20
Cross-training, see also Skill sets
bonus system, 88
daily operations independence, 81
documenting ofce processes, 16
future plans, 187
matrix, 239
purchasing, 106
refections, 211
during slow times, 140141
tasks, two qualifed persons, 186187
value-added employees, 113114
Culture, larger organizations, 2
Customer focus, 2224
d
Daily operations independence
machine output, 8182
256 Index
Daily warm-up preparation, 9798
Data
gathered through observation, 66
impact on pricing, 33
in God we trust, all others bring
data, 171
quick tick lists, 35
to resolve tension, 171
speaking based on, 35
Day-by-hour boards
capacity planning, 149
example, 228
fundamentals, 4
Day sales outstanding (DSO), 43, 106
Days supply of inventory (DSI), 105106,
210
Delays, upfront
capacity planning, lead-time based,
148149
costly design experience, 157158
daily operations independence,
146147
documentation, 150152
hands-on experience, lack-of, 154156
materials-based planning, 147148
measurements confrmation, 145
planning, 147148
proactive response development, 150
problem prioritization, 154
protocol development, 146147
reactive areas, 150
standardization beneft, 146147
strategic inventory, 152154
success confrmation, 145
Design, new layout, 116118, see also
Layout
Design experience, costly, 155, 157158,
190
Design for assembly (DFA), 7, 155
Design for manufacturing (DFM), 7, 155,
190
DFA, see Design for assembly (DFA)
DFM, see Design for manufacturing
(DFM)
Difcult processes, 1314
Digital cameras, 13, 224
Direct labor
closing the books, 42
cost justifcation, 9093
distracted focus, 139
full-absorption costing, 63
role and responsibilities, 54
traditional separation, 69
Diversifcation, see also Expansion, new
products
dangling carrot, 23
market upturn, 166
slowdowns, 191
Dividing focus, observation, 1415
Do, see PDCA (Plan, Do, Check, Act)
Documentation
adjustment phase changeovers, 8384
best time for, 106, 150152
costs, 1516
infractions, 72
observation, 89
outdated/specialized equipment
priority, 20
setup times, 1213
techniques, 223225
training new owner, 205
vacation time buy-in, 103104
variables, machine marking, 8586
Downturn time, see Slowdown impact
Dry-erasable board, 5658, 229
DSI, see Days supply of inventory (DSI)
DSO, see Days ales outstanding (DSO)
e
eBay, 184
Enterprise resources planning (ERP)
systems, 37
Entrepreneurial spirit
purchase background, xviixviii
Web site opportunity, 142143
Entrepreneur magazine, xviixviii
Equipment, see also OSKKK
manufacturers, 2021
Index 257
outdated or specialized, documenta-
tion priority, 20
proft-sharing plan issues, 8687
return on investment, 90, 115
setup-time reductions, 65
travel distance, 180181
ERP, see Enterprise resources planning
(ERP) systems
Estimating, see also Quoting process
assumptions, 24
capacity, creating and selling, 7377
lack of feedback, 32
scheduling job visually, 80
Estimating, process standardization
assumptions as starting measurement,
2425
customers, 2224
equipment manufacturer, 2021
frst draf, 1819, 227228
fundamentals, 28
hiring programmer/estimator, 1718
PDCA, 2628
programming, frst draf, 1819
quoting complications, 1920
skill sets, 2122
sofware, 2021
visualizing measurements, 2526
Events
changes, introducing, 1112, 121125
PDCA as weekly, 49, 75
Excel sofware, 61, 102, 107, 227228
Exclusive listing, 197
Exotic materials, 39
Expansion, new products, see also
Diversifcation
current layout impact, 179182
frst order, flling, 183185
hands-on experience, lack of, 190191
investing steps, 185187
logical direction, 177179
OSKKK impact, 187188
PDCA, quoting process, 188189
Experience, comparison with lean, xv
Extra capacity creation, 4951
f
Fast-tracking, 56
Feedback
estimating, impact on, 32
fnance departments, 5960
measurement monitoring, 7677
PDCA, 26, 31, 73
quoting process, 28
FIFO process, 57, 229
Files, 11, 109110
Filling frst order, 183185
Financial processes, see also Costs
accrual system, 42
closing of books, 4144, 9394
Financing
balance paid of, 174
paying down balance, 169
purchase background, xviii, xxiii
Firefghting, 55, 9697
Firing employees, see Termination
First days alone, see Initial take over
First order, flling, 183185
Fishbone diagrams, 54, 172
5S method application
computer drives and fles, 109110
defned, 10
flling existing space, 180
observation, 1012
Five steps problem-solving method, 172
Five whys method, 54, 172
Flexibility, workers, 114
Flow, see OSKKK
Flowcharts, 15, 21, see also Process maps
Flow lines, 117
Forced waiting time, 51
Formal suggestion system, 68, 87, 209
Front ofce, see Ofce processes
Full-absorption costing thinking, 6364
g
GAAP, see Generally accepted accounting
practices
258 Index
Gap of time, shifs, 168
Gemba
fundamentals, 3
impact on product design, 7
Generally accepted accounting practices,
130
Goodwill
defned, xiii
purchase background, xxiv
valuation, 195
Growth, market upturn, 165, 166, see also
Expansion, new products
h
Hands-on approach
benefts, 8890
experience, lack of, 154156, 190191
Hard drives, computers
new rules, 11
Heavier gauge materials, 182
Hiring, see Recruiting
Historical data issues, 9697
Hit rates
accurate allocations, slow time, 129,
132
quotations follow-up, 108
Hour-by-hour planning, 98
I
Implementation sequence
fundamentals, xv
ofce, 250
shop, 249
Improvements
baseline, 13, 15
forward movement, 1112
Incorporation, purchase background,
xxiiixxiv
Incremental costs, 63
Independence, daily operations
Indirect labor and other costs
better allocation, 132
closing the books, 42
distracted focus, 139
standardization of operations, 150151
traditional separation, 69
Information, gathering as encountered, 3
In God we trust, all others bring data,
171
Initial estimation, alone, 7477
Initial take over
abnormal activities, 5254
frefghting, 55
implementing lean, 249250
learning sequence, 249250
normal activities, 5254
pricing decision, 4849
problem-solving, 55
productivity, proft impact, 5860
proft impact, 5860
proft margins, challenging, 6062
5S organization, 4951
visuals, shop-foor, 5558, 229
Inputting orders, 3031
Inquiries, selling the business, 197198
Internet, business for sale, 197, 204
Interviewing, see Recruiting
Inventory
just-in-time strategy, 38
strategic, 104105, 152154
turns, 105106
Investing, sheet metals fabrication
expansion, 185187
Invoices, see Accounts receivable
Ishikawa diagrams, see Fishbone diagrams
It happens a lot/all the time, 171, 214
J
Job security, 171, see also Longevity
Job-shop activities
ABC costing matrix, 243244
Index 259
planning, ofce processes, 3738
Justifcation
physical location change, 114116
Just-in-time (JIT) inventory strategy, 38
k
Kaizen, 216, see also OSKKK
Kanban systems
consumable, 153, 184
purchasing workload monitoring, 77
standard materials, 31
stock parts, 83
Key points
adjustment phase, 8586
boss priorities, 68
bottlenecks, 118, 149
capabilities for unique markets, 182
capacity plan, 148
cross-training, 141
data usage, 66
direct labor, 92
documentation, 9
gap between shifs, 168
Gemba, 5
hiring, 100
indirect costs, 138
indirect labor, 92
labor hours feedback, 32
layouts, 118, 121, 180181
lead times, 149
learning new processes, 204
measurement monitoring, 26, 7677,
8586
new processes, 182
outdated/specialized equipment, 20
overtime, 65
pilot/prototype, 58
productivity, 92
quoting process, 28
real-time Pareto charts, 66
root cause problem solving, 55
slowdowns, 115
standardization, 9, 13
structured problem-solving, 172
technology for standardization, 13
termination, 100
tick list for data, 35
training, new technology, 14
value-added measures, 92
verifcation system, labor hours, 32
visualizing abnormalities, 35
workstation preparation, 98
Knowledge silos, 6, see also Tribal
knowledge
l
Labor costs, see also Direct labor; Indirect
labor and other costs
assumptions, 24
skill sets, 21
Larger organizations, issues, 12
Late starts, 4, 9798
Layout, see also OSKKK
bottlenecks, 116118
design, 116118
L cell, 241
markings, 117
sheet metals fabrication expansion,
179182, 247
simulation, 120121
Lead times
job-shop planning, 3738
purchasing, 103
strategic inventory impact, 152154
Lean bingo, xv
Lean principles
abnormal events, 5253
activity-based costing, 61, 131
A3 method, 164, 172
asset, process standardization, 199
baseline, 13
bonus system, 87
changeover time, quality verifcation,
85
consumable Kanban, 153
customer focus, 23
data-based discussions, 171
day-by-hour board, 4
260 Index
design for assembly, 7, 156
design for manufacturing, 7, 156
DSO, 43
FIFO process, 5758
fow lines, 117
formal suggestion system, 68
Gemba, 3
generally accepted accounting
practices, 130
it happens a lot/all the time, 171
just-in-time inventory strategy, 38
Kaizen eyes, 33
Kaizening of equipment, 66
Kanban systems, 104, 153
layout applications, 119120, 124
lead times, 149
machine gun Kaizen, 216
machine vs. operator cycle, 189
motion Kaizen, 33
MRP systems, 31
new layouts, 117
non-stock parts, 61
normal events, 5253
OSKKK applications, 101, 115, 119120
overall equipment efectiveness, 70
PDCA, 27, 31
problem solving steps, 5455
process mapping, 151, 251
product design impact, 156
return on investment, 92, 115
runners, repeaters, strangers, 61
Six Sigma, 164
skills matrix, 88
standardization, 13, 52, 109, 199
stock parts, 61
suggestion system, 68
takt time, 4
theory of constraints, 148
total productive maintenance, 97
value stream mapping, 151, 251
visual scheduling, 80
work calendar standardization, 109
workplace organization (5S), 10, 51
Lessons learned, 207209, see also
Refections
Lifetime of tools, 5253
Lists, xvi, see also Tick lists
Location change
fundamentals, 125
layout, 116118, 120121
Location importance, xx
Logical direction, 177179
Longevity (job security)
defned, 7273
securing, 2
setup times, continuous measuring,
6768
M
Machine gun Kaizen, 216
Machine operator training, 67, see also
Cross-training
Machine output
changeover adjustment phase, 8384
closing of books, 9394
continuously running, 122123
cost justifcation, team leader, 9093
fnancial processes, 9394
hands-on approach, 8890
independence goal, 8182
Kaizen of equipment, 8384
proft-sharing plan, 8688
quality confrmation, 8486
Made-to-Order Lean, 217
Manufacturing overhead, 63, see also
Indirect labor and other costs
Market downturn
capacity planning, lead-time based,
148149
Index 261
cost allocation pricing, 129130,
140141
costly design experience, 157158
daily operations independence,
146147
hands-on experience, lack-of, 154156
materials-based planning, 147148
measurements confrmation, 145
planning, 147148
proactive response development, 150
problem prioritization, 154
protocol development, 146147
reactive areas, 150
standardization beneft, 146147
Marketing package, 196
Market upturn
capabilities expansion, 169170
challenging the team, 164166
diversifcation, 166
growth, 165
nontraditional operations, 169170
opinion minimization, 170173
policy enforcement, 175
problem-solving skills, 163164
process details, 173175
process improvements, 161162
recruiting, 165166
second shif creation, 166169, 174
Material requirements planning (MRP)
systems, 3031, 37
Materials
costs, skill sets, 21
heavier gauge, 182
purchasing, 103
Materials-based planning, 147148
Matrix, see Activity-based costing (ABC)
methodology
Measurements
confrmation, upfront delays, 145
correct, refections, 210211
focus and follow-up, 67
visualizing, 2526
Microsof sofware, see specifc package
MIG welding equipment, 185186
Minority-owned quota, xx
Misallocations, 139
Mockup, layouts, 121
Monitoring, 80, see also OSKKK; PDCA
Moving business location
fundamentals, 125
layout, 116118, 120121
MRP, see Material requirements planning
(MRP) systems
n
Negative variances, 42
Network upgrades, 101102
New product expansion
current layout impact, 179182
Noncompete clause, 203
Nondisclosure agreement, 196
Nondisruptive move, 118120
Non-stock parts and tools
proft margins, 61
Nontraditional operations, 169170
Nonvalue added processes
ofce processes, 29
purchasing departments, 191
workplace organization, 51
262 Index
Normal activities identifcation, 5254,
see also Abnormal activities
identifcation
o
Observation, see also OSKKK
dividing focus, 1415
documentation, lack, 89
fundamentals, 15, 16, 219
lack of training, 67
machine operator training, 67
processes, understanding, 56
scheduling jobs visually, 80
skill set identifcation, 910
5S opportunities, 1012
standardization, 1214
training, 67
using subcontractors quotes, 183184
workplace organization, 1012, 51
OEE, see Overall equipment efectiveness
(OEE)
Ofers, 201204
Ofce processes, see also Administrative
processes
ABC costing matrix, 245
accounts receivable, 4344
closing of books, 4143
data for pricing, 33
estimating lack of feedback, 32
fnancial processes, 4144
job-shop planning, 3738
orders input, 3031
physical location change, 123125
planning, 3739
purchasing, 3435
receivables, 4344
scheduling, 36
setup time, 41
simplifed short-term planning, 3839
tribal knowledge, 36
On-time delivery, lessons learned, 210
Opinion minimization, 170173
Order entry, 3031
Orientation process, 56
OSKKK (Observe, Standardize, Kaizen
Flow & Process, Kaizen
Equipment, Kaizen Layout)
bottlenecks, 162, 215216
documentation, 9
endless loop, 122
impact, sheet metals fabrication
expansion, 187188
implementation, xiv
Kaizen of equipment, 221
Kaizen of fow and process, 106, 220
Kaizen of layout, 221
observation, 2, 219
ofce processes, 101, 128
proft margins, 61
scheduling jobs visually, 80
shop application, 101
Outdated equipment, 20, see also
Computers; Sofware
Outsourcing, see also Subcontracted
workers
complicated parts, 80
in-house capability, 187188
lack of internal capabilities, 75
Overall equipment efectiveness (OEE)
continuously running machines, 123
cost justifcation, team leaders, 91
direct vs. indirect labor, 69
measurement technique, 233236
refections, 211212
return on investment, 92
total productive maintenance, 97
Overtime
dependence on, 93
as positive, 6465
quick growth, 171
repetitive problems, focus, 154
transition to second shif, 167
workstation preparation, 98
Owner fnancing, 198199
Index 263
P
Parrot chart, see Real-time Pareto chart
Part numbers, 31
Parts management, 212213
Paychecks, issue day, 41
Payment terms, 43
PDCA (Plan, Do, Check, Act)
bonus system, 88
boss priorities, 68
defned, xvi
estimating process, 7377, 188189
estimating standardization, 2628
feedback, 31
fnance departments, 59
in-house capability, 188
processes changed without, 129
scheduling boards, 7172
weekly event, 49, 75
Perks, 48
Personal life development
fnancial stability, 48
moving to Spain, 193194
proposal to girlfriend, 79
wedding, 95
Physical location change
fundamentals, 125
layout, 116118, 120121
Plan, see PDCA (Plan, Do, Check, Act)
Planning
capacity, lead-time based, 148149
job-shop, 3738
lead-time based, 148149
ofce processes, 3738
simplifed short-term, 3839
workforce fexibility, 114
Planning boards
materials, 39
overtime, 64, 93
recurring delays, 152
Politics, larger organizations, 2
Positive variances, 42
Possible business considerations, xix
Preferred supplier status, 191
Press brake, 184, 247
Pricing, see also Rate development
development, 22
lowering, selling the business, 199200
purchasing, 103
slowing down sales, 169
valuation, selling the business, 195196
Pricing, cost allocation
activity-based costing, 130138,
243245
cost linking to matrix, 134, 136138
matrix, 132136, 243245
other methods of allocation, 138140
outsourcing reduction, 141
pursuing opportunities, 142143
slowdown impact, 129130, 140141
standardization benefts, 142143
Print screen function, 13, 15
Prioritization, 51
Proactive response development, 150
Problem prioritization, 154
Problem-solving
market upturn, 163164
steps, 5455
structured, 170175
Procedural changes, 121123
Processes
business application, 215217
change, 5
focus, 72
improvements, market upturn, 161162
selling the business, 194195
264 Index
understanding, 56
Process maps, see also Value stream
mapping (VSM)
applications, 251
documentation techniques, 224
Productivity
bonus system, 68
proft impact, initial take over, 5860
standardization impact, 164165
Proft impact
productivity, initial take over, 5860
results of lean implementation, 174
Proft margins
lack of accurate data impact, 32
part number inaccuracy, 59
PDCA, estimating process, 7374
Proft-sharing plan, 8688
Programmer/estimator, hiring, 1718
Programming, frst draf, 1819
Protocol development, 146147
Purchase of business
background developments, xvii
boat life, xviii
closing, xxiv
CNC machining, xxii
entrepreneurial spirit, xviixviii
fnancing, xviii, xxiii
goodwill, xxiv
incorporation, xxiiixxiv
location importance, xx
minority-owned quota, xx
possible business considerations, xix
selected business, xxixxii
xviii, xxiii
terms of sale, xxii
valuation methods, xx, xxiii
Purchasing
q
Quality
accreditation, 101, 110111
competitive environment, 83
confrmation, 8, 8486
Quoting process, see also Estimating
complications, 1920
feedback, 28
training and experience, lack of, 155
using subcontractors, 183184
r
Rate development, 2122, see also Pricing
Reactive areas, 150
Real-time Pareto chart
example, 231
setup time, 66, 69
Receivable, see Accounts receivable
Recruiting
daily operations independence, 8182
experience in large companies, 98100
market upturn, 165166
programmer/estimator, 1718
Redline drawings, 7, 156
Reduction, setup times, 6567, 8283
Reference points, marking on machines,
8586
Refections
activity-based costing, 213215
cross-training, 211
fundamentals, 207
managing parts diferently, 212213
measurements, correct, 210211
OEE, 211212
process for businesses, 215217
team leaders, 211212
Reorder system, purchasing, 103104
Repeaters
customer focus, 2324
standardized setup sheet, 84
Respect, 175
Root cause analysis
Index 265
normal vs. abnormal, 5354
visual measurement, 26
Runners
adjustment phase changeovers, 8384
delay wastes, 146147
tools, setup-time reductions, 66
Rush jobs, 2324
s
Safety stock, 105106, see also Inventory
Scheduling
boards, 7172
tribal knowledge, 36
visually, 80
workforce fexibility, 114
Screenshot documentation, 13, 15, 225
Second shif creation, 166169
Selected business, xxixxii
Selling the business
advertising, 197
asking price determination, 195196
buyers and ofers, 200202
epilogue, 204205
evaluating ofers, 201202
inquiries, 197198
lower price, 199200
marketing package, 196
ofer accepted, 202204
owner fnancing, 198199
process steps for, 194195
valuation, 195196
Seniority, 113
Sequence of learning, 208209, 216,
249250
Setup activities, externalizing, 71
Setup times
continuous measurements, 6770, 231,
233236
reductions, 6567, 8283
runner focus, 83
Sheet metals fabrication expansion
layout impact, 179182, 247
Shif work
creation, 166169
gap of time between, 168
proftability, 174
Shine, 5S method application, 10
Shingo, Shigeo, 69
Shop, see Job-shop activities
Short-term planning, 3839
Silos of knowledge, 6
Simulation, new layout, 120121
Single minute exchange of dies (SMED),
12, 67
Six Sigma, 53, 54
Skill sets, see also Cross-training
estimating process standardization,
2122
labor estimates, 19
observation, 910
value-added, 113114
Slowdown impact
cost allocation pricing, 127130,
140141
focus on improvements, 113115
xviii, xxiii, 199
SMED (single minute exchange of dies),
12, 67
Sofware, see also Computers; Outdated
equipment
estimating process standardization,
2021
outdated, 14, 102
Sort, 5S method application, 10, 6566
Specialized equipment, 20
Standardization, see also OSKKK
benefts, 7477, 142143
difcult processes, 1314
documentation, 9
266 Index
fundamentals, 220
lack of, 1213
observation, 1214
productivity support, 164
quoting process, 227228
technology used in, 13
training new owner, 205
vacation time buy-in, 103104
work calendars, 108109
Standardization, estimating process
assumptions as starting measurement,
2425
customers, 2224
equipment manufacturer, 2021
estimating, frst draf, 1819
fundamentals, 28
hiring programmer/estimator, 1718
PDCA, 2628
programming, frst draf, 1819
quoting complications, 1920
skill sets, 2122
sofware, 2021
visualizing measurements, 2526
Stock parts, 61, 83, 103
Straighten, 5S method application, 10
Strangers, 83, 212
Strategic inventory
lead time reduction, 152154
purchasing, 104105
Structured problem-solving, see also
Problem-solving
opinion minimization, 170173
process details, 173175
Structured process orientation, 56
Subcontracted workers, 114, see also
Outsourcing
Success confrmation, 145
Suggestion system, 68, 87, 209
Suppliers
automatic replenishing systems, 153
documentation, 44
proactive vs. reactive areas, 150
Sustain, 5S method application, 10
t
Take over, see Initial take over
Takt time, 4
Tasks, see also Cross-training
two qualifed persons, 186187
Team leader position
capacity, creating and selling, 7071,
237238
refections, 211212
role and responsibilities, 54, 92,
237238
Technology upgrades, 101102, see also
Computers
Temporary workers, 114
Tensions, 170172
Termination, see also Absenteeism
enforcement
documentation, 72, 100
enforcing to be fair, 175
policy, 74
Terms of sale, xxii
Teory of constraints, 147148
Tings to do, slowtime list, 140141
Tick lists
quickly gathering data, 35
recurring delays, 152
TIG welding equipment, 185186
Tolerances
costly mistakes, 7, 157158
question tight, 155156
Tooling
cost estimates, 20
lifetime, 5253
purchasing, 184
time wasted looking for, 5051
Total productive maintenance (TPM), 97
Toyota Production System (TPS)
adapting, 217
A3 method, 171, 172
closing of books contrast, 4142
Index 267
cost justifcation, team leaders, 91
defned, xvi
documenting ofce processes, 16
gap between shifs, 168
Gemba, 3
hands-on training, 158, 209
implementation, xiv
layouts, 117118
orientation process, 6
TPM, see Total productive maintenance
(TPM)
Training, see also Cross-training
cost, new hires, 1516
observation, 67
updating, 14
Transition phase, 118120
Travel distance, equipment, 180181
Tribal knowledge
defned, xvi
documentation, 89
implementing changes, 11
learning, 2728
process standardization asset, 199
scheduling, 36
u
Upfront delays, see Delays, upfront
v
Vacation, 103104
Vacuum fxtures, 181182
Valuation methods
purchase background, xx, xxiii
selling the business, 195196
Value-added measures
employees ability, 113114
productivity, 92
Value stream mapping (VSM), see also
Process maps
applications, 251
problem-solving, 164
Variable costs, 63
Video cameras, 13, 15, 224
Visual management
machine output, 80
measurements, 2526
productivity, 7173
purchasing, 3435
scheduling, 80
shop-foor, 5558, 229
VSM, see Value stream mapping (VSM)
W
Wage rates
bonus system, 8788
Warm-up preparation, 9798
Waste identifcation
delays, 146
eyes for, xvi, 33
observation step, 11
Web site opportunity, 142143
Welding, limitation, 185187
Work calendars, standardized, 108109
Work ethics, 69, 99100
Work-in-progress (WIP), 83, 91
Work order package
early contents, 3940
notations for settings, 8586
proft margins, 61, 227228
setup time impact, 41
Workplace organization
nonvalue added time, 51
observation, 1012
269
About the Author
Greg Lane began his lean journey afer earning his bachelor of science in
mechanical engineering from the University of Wisconsin in 1986 and his
master of business administration from the California State University in
1989. He was one of the few selected for
a one-year training program conducted
by Toyotas masters and centered in
Japan. He became certifed as a Toyota
Production System (TPS) Key Person
and returned to the United States, where
he trained others in the NUMMI (New
United Motor Manufacturing, Inc.)
plant, a ToyotaGeneral Motors joint
venture in Fremont, California.
Since his training in 1992, Greg has
continued his lean learning by work-
ing on implementing lean around the world, supporting large and small
companies alike. In 1998, he began to focus his lean endeavors on meet-
ing the specifc needs of high-mix, low-volume manufacturing and service
providers. Although his work is geared to companywide improvements,
Greg has also focused on individual departmentspurchasing, engineer-
ing, fnance, and planningto improve efciencies, reduce lead times,
and reduce costs.
While working as an independent consultant, Greg purchased and oper-
ated (for six years) his own manufacturing company, which specialized in
fast turnaround on high-mix, low-volume parts. He used TPS to grow the
business and nearly double its sales, and sold it at a proft to concentrate
on supporting others in lean implementation.
Tough his recent concentration has been with high variation and low-
volume applications, Greg started in high-volume environments and his
experience includes everything from test, repair, defense, and overhaul
businesses as well as many non-manufacturing companies such as ser-
vice providers. His previous book, Made-to-Order Lean: Excelling in a
High-Mix, Low-Volume Environment, explained how to adapt certain lean
270 AbouttheAuthor
principles for made-to-order businesses. He has also developed training
manuals and materials in use by lean practitioners as well as publishing
articles in various engineering periodicals.
Greg is a faculty member of the Lean Enterprise Institute in the United
States and a senior advisor to the Instituto Lean Management in Spain. He
has worked in twenty-four countries supporting lean transformations and
speaks English, German, and some Spanish.
Greg and his associates are always interested in helping others in their
lean journeys, especially when challenged with nontraditional applica-
tions. Tey can be reached at [email protected] or more infor-
mation can be found at www.lowvolumelean.com.

Lean Manufacturing

Uploaded by

Copyright:

Available Formats

Lean Manufacturing

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lean Manufacturing

Uploaded by

Copyright:

Available Formats

What is the book about?

What challenges does the author face in transforming the manufacturing company?

Mr.

Lean Buys and

Tese same ideas should be

total parts produced

You might also like