Introduction To DFQR

Download as pdf or txt
Download as pdf or txt
You are on page 1of 38

601

Reference Books
1. Introduction to Statistical Quality Control- Douglas
Montgomery
2. Practical Reliability Engineering- Patrick O'Connor
Voice of Customers, Data collection techniques– QFD– System
Modeling Techniques – Measurement & Error – Gauge performance–
Gauge calibration– R&R study
 
Benchmarking– Concept Generation Process– Concept Development
Techniques ‐ TRIZ– Design Tradeoff Analysis
 
Design for Reliability – Reliability Analysis– Reliability Testing
Probability Distributions– Performance Standards – Process
Capability– Risk Assessment
 
Design of Experiments– Response Surface Models– Process
Capability Modeling, Computation of variation– probabilistic
Optimization
 
Robust Design– Tolerance Analysis– Design verification

Control charts. Central Limit Theorem for a family of reliability


measures, modeling and reliability analysis of multi‐ state
systems, Weibull data analysis for few or no failures,

Markovian performance evaluation methods, software reliability


and testing
 
In this course students will design a component using above
mentioned techniques.
Engineering
Steps for Producing Products
1. Planning and forecasting
2. Conceptualization
3. Feasibility Assessment
4. Establishing the Design Requirements
5. Preliminary Design (Embodiment Design)
6. Detailed Design
7. Process planning
8. Production Planning and Tool Design
9. Prototyping
10. Production
11. Testing
12. Inspection (Quality Assurance)
13. Packaging and shipment
Dimensions of Quality
1. Performance (how well it performs)
2. Reliability (how often it fails)
3. Durability (effective service life of a product., length of
failure free or maintenance free operation period)
4. Maintainability
5. Aesthetics
6. Features
7. Perceived Quality (the quality that customers recognize
via the look, the touch, and the feel of a product)
8. Conformance to standards

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 5


Introduction
 The production processes are not perfect!

 Which means that the output of these processes will not be perfect.

(non identical and non-deterministic)

 Successive runs of the same production process will produce non-

identical parts.

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 6


This is a traditional definition

This is a modern definition of quality

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 7


Meaning of Quality

1. FITTNESS FOR PURPOSE (JURAN)

2. CONFORMANCE TO REQUIREMENTS or SPECIFICATIONS

(Philip Crosby )

3. QUALITY is a measure of how closely a good or service conforms

to specified standard.

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 8


An equivalent definition is that quality improvement is the

elimination of waste. This is useful in service or transactional

businesses.

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 9


How to check / ensure Quality when
Number of Products are very High

Use Statistical Methods

STATISTICAL QUALITY
CONTROL

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 10


“Statistics is a way to get information from
data”

Data Statistics

Information
Data: Facts, especially
numerical facts, collected
together for reference or Information: Knowledge
information. communicated concerning
some particular fact.

Statistics is a tool for creating new


understanding from a set of numbers.

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 11


Population Sample

Subset

Statistic
Parameter

 Populations have Parameters


 Samples have Statistics

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 12


Population
A population is the group of all items that possess a certain
characteristic of interest.
Size: very large; sometimes infinite.

Sample
A sample is a set of data drawn from the population.
Size: Small (sometimes large but less than the population)

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 13


Parameter
A descriptive measure of a population.

A parameter is a characteristic of a
population, something that describes it.

Statistic
A descriptive measure of a sample.

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 14


Statistical Methods

Statistical
Methods

Descriptive Inferential
Statistics Statistics

Hypothesis
Estimation
Testing

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 15


 Descriptive Statistics describe the data set that is
being analyzed, but doesn’t allow us to draw any
conclusions or make any interferences about the data.

Hence we need another branch of statistics:


inferential statistics.

 Inferential statistics is also a set of methods, but


it is used to draw conclusions or inferences about
characteristics of populations, based on data from a
sample.
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 16
Descriptive Statistics

Descriptive Measures

 Measures of central tendency


◦ Mean, median, mode, etc.
 Quartile
 Measure of variation
◦ Range, interquartile range, variance and standard
deviation, coefficient of variation
 Shape
◦ Symmetric, skewed, using box-and-whisker plots
 Coefficient of correlation

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 17


Inferential statistics
Confidence Interval Estimation Process

I am 95%
Population Random Sample
confident
Mean that  is
Mean, , is between
unknown X = 50
40 & 60.

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 18


Summary Measures

Central Tendency Quartile Variation

Mean Mode
Range Coefficient of
Median Variation

Variance

Standard Deviation

6/30/2020 11:23 B. Shahul Hamid Khan,


AM IIITDM 19
Mean (Arithmetic Mean)

Mean (arithmetic mean) of data


values Sample Size
n
Sample mean
 X i
X  X   X
X  i 1
 1 2 n

n n

Population
N mean Population Size

X i
X1  X 2   X N
  i 1

N N
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 20
Mean (Arithmetic Mean)
 The most common measure of central
tendency
 Acts as ‘Balance Point’
 Affected by extreme values (outliers)

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 8
8 9 10 9 10 12 14
Mean = 5 Mean = 6

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 21


◦ Robust measure of central tendency
◦ Not affected by extreme values

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 8
8 9 10 9 10 12 14
Median = 5 Median = 5

 The value of Middle when the observations are


ranked.
Property:
50% of the values are Less than or equal
to it. 6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 22
Mode
 A measure of central tendency
 Value that occurs most often
 Not affected by extreme values
 Used for either numerical or categorical data
 There may be no mode or several modes

Mode = 9 No Mode

0 1 2 3 4
0 1 2 3 4 5 6 7 8 9 10
5 6
11 12 13 14

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 23


◦ Measure of variation
◦ Difference between the largest and the smallest
observations:
Ignore the way in which data are distributed

Range  X Largest  X Smallest

Range = 12 - 7 = 5 Range = 12 - 7 = 5

7 8 9 10 7 8 9 10
11 12 11 12
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 24
Variation

Standard Coefficient of
Variance Deviation Variation

Range Population
Population
Variance
Standard
Deviation
Sample
Variance Sample
Standard
Interquartile Range Deviation
Variance
 Important measure of variation
 Shows variation about the mean

 X X
2
i
Sample variance: S 2 i 1

n 1
N

 X 
2
i
Population variance:
2  i 1

N
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 26
Standard Deviation
 Most important measure of variation
 Shows variation about the mean
 Has the same units as the original data

 X X
2
i

Sample standard deviation: S i 1

n 1
N

 X 
2
Population standard deviation: i
 i 1

N
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 27
Confidence interval for the Average

What is the 90% confidence interval for the average, where


sample size n = 15, S= 1.2 and Sample average X = 25.

(Note: When sample size is less than 30; use t- distribution. If


greater than or equal to 30; use normal distribution)

(1 –
α - α)
Risk
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 28
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 29
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 30
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 31
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 32
33
Confidence Interval for the variance

Chi-square distribution

Confidence Interval for the Ratio of two


variances

F distribution

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 34


To control or improve a process we need information or Data.

Data on quality characteristics is described by a random


variable

Random variable
1. Discrete variable
2. Continuous variable

Discrete variable
No of defective rivets in assembly
No of satisfied customers in a retail shop

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 35


Continuous variable

Thickness of a plate
Viscosity of an oil
Diameter of a shaft

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 36


1. Discrete Distributions
2. Continuous Distributions

Discrete Distributions
a. Binomial Distribution
b. Hyper geometric Distribution
c. Poisson Distribution

Continuous Distributions
1. Normal Distribution (quality control)
2. Exponential Distribution (to model the failure
rate during life cycle of product)
3. Weibull Distribution (failure rate in mech and
electrical components)

6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 37


Data A
Mean = 15.5

3.338
s =
11 12 13 14 15 16 17
18 19 20 21
Data B
Mean = 15.5
s
11 12 13 14 15 16 17 = .9258
18 19 20 21
Data C
Mean = 15.5

11 12 13 14 15 16 17 s = 4.57
18 19 20 21
6/30/2020 11:23 AM B. Shahul Hamid Khan, IIITDM 38

You might also like