keport

on
1he Goa|
A rocess of Cngo|ng Improvement


SubmlLLed Lo
rofessor Atanu Chowdhary

SubmlLLed by
Group S Sect|on L
Shallesh kumar
Shlbashlsh 8ehera
Shlshlr kumar klsan
Shreya 8haLnagar
ShrlkanL Agarwal
vlnay L




ConLenLs

arL A Learnlngs from Lhe bookages 36
denLlfylng Lhe Coal
ecldlng Lhe measuremenLs
AccounLlng for varlable change dependenL evenLs and sLaLlsLlcal flucLuaLlons
8oLLlenecks
8rlef summary of lssue dealL ln a subploL workllfe balance
Learnlngs and summary
arL 8 AppllcaLlon ln Lhe lacemenL CperaLlons rocessages 67

















Pait A - Leainings fiom the Book

1he sLory ls based around Lhe llfe of Alex 8ogo lanL Manager for unlwarea dlvlslon of
unlCo AfLer a very upseL cusLomer approaches Alex's boss 8lll each he ls glven an
ulLlmaLum Lo Lurn Lhe planL around ln Lhree monLhs ue Lo Lhe llmlLed amounL of Llme
avallable Lhere are noL many ouLslde Lools avallable Lo hlm such as consulLanLs surveys
eLc WlLh very llLLle hope Alex foresees Lhe lnevlLable unLll he remembers hls conversaLlon
wlLh !onah a physlclsL Alex knew from hls sLudenL days

lJentifyinq Tbe 6ool

Alex decldes Lo make ouL some meanlng from hls conversaLlon wlLh !onah urlng Lhe
conversaLlon !onah asks hlm several quesLlons Lo analyze hls company's slLuaLlon 1he
conversaLlon leads ulLlmaLely Lo Lhe quesLlon WhaL ls Lhe goal of any buslness? AfLer
ponderlng over hls conversaLlon Alex reallzes LhaL Lhe goal of any buslness ls Lo make
money lurLhermore lf Lhe goal ls Lo make money any acLlon Loward Lhls goal ls Lo be
counLed as producLlve and any acLlon noL movlng Lowards Lhe goal ls nonproducLlve Alex
unsure of such a slmple answer decldes Lo conLacL !onah Lo conLlnue Lhe search for more
answers

eciJinq tbe Heosurements

Cnce Alex conLacLs !onah Lhey deflne Lhe followlng measuremenLs Lo deflne Lhe success of
any planL's producLlon
O 1hroughouL ls Lhe raLe LhaL Lhe sysLem ls used Lo generaLe money Lhrough sales 1hls
measuremenL would conslsL of whaL a producL would be worLh when sold aL markeL
value afLer deducLlng operaLlonal expense and lnvenLory
O nvenLory ls all Lhe money lnvesLed ln purchaslng lLems LhaL wlll be sold 1hls could
lnclude Lhe remalns of Lhelr machlnes afLer belng used for producLlon
O CperaLlonal expense ls all Lhe money used Lo Lurn lnvenLory lnLo LhroughpuL 1hls
would lnclude such lLems as depreclaLlon of a machlne lubrlcaLlng oll scraps eLc

Goa|
Make
money

ccountinq for vorioble cbonqe - epenJent events onJ fluctuotions

!onah explalns Lo Alex LhaL a planL LhaL ls conLlnuously producLlve ln Lo be consldered

lnefflclenL Pe furLher explalns LhaL conLlnuous producLlon wlll resulL ln hlgh absenLeelsm
poor quallLy and employee Lurnover 8ased on Lhls he would need Lo reduce operaLlonal
expense and lnvenLory Lo lmprove LhroughpuL Lo demonsLraLe a balanced llne of
producLlon !onah leaves hlm Lo ponder Lhe undersLandlng of Lwo Lhlngs whaL are Lhe
dependenL evenLs and sLaLlsLlcal flucLuaLlons ln hls planL Pe comes ouL wlLh Lhe followlng
deflnlLlons

ottlenecks
urlng a hlklng Lrlp wlLh hls son and hls scouL frlends Alex develops an experlmenL wlLh
bowls and maLches Lo model a balanced planL 8owls represenL work sLaLlons maLches
represenL producL lnvenLory and one dle (from a palr of dlce) ls used Lo slmulaLe Lhe
sLaLlsLlcal flucLuaLlons (varlaLlon) ln performance aL each work sLaLlon or operaLlon 1he
bowls are seL up as a producLlon llne represenLlng dependenL evenLs where each operaLlon
has Lhe same capaclLy le slx producLs per day wlLh a range of varlaLlon from one Lo slx

Lach player (scouL) rolls Lhe dle Lo deLermlne how many maLches Lo place ln hls bowl 1hls
represenLs one days producLlon for LhaL operaLlon lor example lf Lhe flrsL player rolls a slx
Lhen he places slx maLches ln hls bowl f Lhe nexL player rolls a four he can only move four
maLches from Lhe flrsL bowl Lo hls bowl Lach operaLlon ls dependenL on Lhe upsLream
operaLlon for lnpuL f Lhe nexL scouL rolls a flve he can only move four Lo hls bowl because
Lhere are only four avallable ln Lhe prevlous bowl (upsLream operaLlon) ln Lhe process 1he
flrsL scouL Lo roll Lhe four became Lhe boLLleneck operaLlon f anoLher player down sLream
rolls less Lhan a four Lhen he becomes Lhe boLLleneck
1he scouLs each roll Lhe dle several Llmes ln sequence Lo represenL several days producLlon
and each Llme Lhe boLLleneck nearly always appears aL a dlfferenL operaLlon or scouL 1he
polnL of Lhls game or demonsLraLlon ls Lo show LhaL where each operaLlon ln a sequence of
dependenL evenLs has Lhe same amounL of capaclLy (a balanced planL) Lhe varlaLlon and
dependenL evenLs wlll cause Lhe boLLleneck Lo move from operaLlon Lo operaLlon le
floaLlng boLLlenecks occur Managers wlll noL know where Lhe boLLleneck wlll show up nexL
and wlll noL be able Lo manage Lhe sysLem

AL Lhe end of Lhe game he concluded LhaL Lhe boLLleneck's speed of producLlon ls whaL
deLermlnes Lhe speed of Lhe oLher dependenLs Pence lnvenLory moves very slow because
of sLaLlsLlcal flucLuaLlons 1hls would evenLually resulL ln huge boLLlenecks over Llme

n Lhe end Lhls ls where Alex beglns hls search lnslde Lhe planL upon arrlvlng back aL work
he and Lhe crew begln Lhelr search for Lhe boLLlenecks 1hey ldenLlfy one of Lhe roboLs nCx
10 and Lhe heaL LreaLmenL area as Lhe source of Lhe boLLlenecks 1hey also clarlfled Lhelr
noLlons abouL Lhe managemenL of Lhe planL whlch were fundamenLally flawed

nsLead of balanclng flow wlLh demand Lhey used Lo balance capaclLy wlLh demand
erformance of lndlvldual ls deLermlned noL only by hls poLenLlal buL by oLhers
consLralnLs Loo
AcLlvaLlon uLlllzaLlon of resources are noL synonymous
1hey also reallzed Lhe lmporLance of knowlng Lhe Llme facLors lnvolved ln shop floor
processes
SeLup Llme lnlLlally spenL walLlng for resource
rocess Llme spenL ln modlfylng lnLo new form
;ueue Llme spenL ln llne whlle Lhe resource ls busy
WalL 1lme parL walLs Lo geL assembled
dle Llme when Lhe machlne sLands wlLhouL work

Cnce Lhey had developed an undersLandlng on Lhe ground rules Lo proceed Lowards Lhe
goal for a planL and Lhe boLLlenecks were ldenLlfled Lhey began Lhelr search for solvlng
Lhem 1hese are Lhe sLeps Lhey proceeded wlLh Lo solve Lhe boLLleneck lssue

-Ck10

% Pave Lwo employees dedlcaLed Lo Lhe seLup of Lhe roboL Lhls wlll help ensure LhaL Lhe
machlne wlll be conLlnuously runnlng
% AddlLlonally Lhese employees Lake Lhelr breaks whlle Lhe machlne ls worklng slnce Lhey
are only needed for seLup of Lhe nexL baLch
% ConLlnue use of Lhe older machlnes LhaL Lhe nCx10 was supposed Lo replace
% o noL produce parLs for lnvenLory produce only whaL ls ln demand (vla an order from a
cusLomer)
% roduce smaller baLches Lo lncrease producLlvlLy for Lhe whole process

eat 1reatment

% erform quallLy conLrol LesLlng before golng lnLo Lhe boLLleneck
% Plre ouLslde vendors Lo asslsL ln heaL LreaLmenL
% roduce smaller baLches Lo lncrease producLlvlLy for Lhe whole process
% Comblne baLches LhaL requlre Lhe same LemperaLure

1hese sLeps Laken can be summarlzed as
nexL Loglcal SLep 8educLlon ln baLch slze Lo half leadlng Lo reducLlon ln work ln
progress whlch ulLlmaLely led Lo reduced lnvenLory
LffecL reduced pressure on cash flow
LffecL of reduced quanLlLy lncrease ln frequency from suppllers

WlLh all of Lhese changes however one Lhlng remalns consLanL an hour losL aL a boLLleneck
ls an hour losL aL Lhe whole sysLem 1herefore lL ls exLremely lmporLanL Lo keep Lhe
boLLleneck movlng aL all Llmes 8ased on Lhese few slmple sLeps Lhey slgnlflcanLly lncreased
producLlon by 12 durlng Lhe flrsL Lwo monLhs and 20 ln Lhe Lhlrd monLh 8ased on Lhese
lncreases Alex saves Lhe planL and ls promoLed Lo 8lll each's poslLlon

rief summory of issue Jeolt in o sub-plot - work-life bolonce

8esldes Lhe ongolng Lheme of savlng Lhe planL Lhere ls an underlylng sLory abouL Alex's
personal llfe n Lhe beglnnlng Alex and hls wlfe !ulle are consLanLly argulng abouL Alex
worklng laLe ln aLLempLs of savlng Lhe planL whlch flows over lnLo Lhelr llves 1hls resulLs
lnLo hlm noL spendlng any Llme wlLh hls famlly Soon hls wlfe leaves hlm afLer belng
exhausLed of all aLLempLs Lo save Lhelr marrlage Powever durlng hls search for reclalmlng
Lhe planL he does Lhe same wlLh hls marrlage We suppose LhaL Lhe auLhor ls sLaLlng Lo Lhe
audlence LhaL Lhere musL be a balance beLween Lhese Lwo llves L ls lmporLanL for boLh Lo
be ln harmony because ulLlmaLely Lhey wlll flow ln and ouL of each envlronmenL

leorninqs onJ Summory

1hls book presenLs slmpllfylng ways Lo lmprove any process wheLher lL ls manufacLurlng or
servlce orlenLed Ls approach of back Lo baslcs helps breakdown whaL has become
common pracLlce 1he source of Lhe problem Lhe whole Llme was followlng Lhe rules LhaL
have come Lo be Laken as granLed And Lhe sLeps Lo lmprove conLlnuously can be
summarlsed as

(l) ldenLlfy Lhe sysLem's boLLlenecks
(ll) declde how Lo explolL Lhose boLLlenecks
(lll) subordlnaLe everyLhlng else Lo sLep Lwo declslons
(lv) evaluaLe Lhe sysLems boLLlenecks
(v) lf ln a prevlous sLep a boLLleneck has been broken go Lo sLep one
(vl) producLlon ls an ongolng process of lmprovemenL and when new problems arlse
Lhey need Lo be dealL wlLh accordlngly


Tbe Plocement Process
Slnce Lhe placemenL cell can only work Lo Lhe end Lo ensure maxlmum number of lnLervlews
per candldaLe Lo lncrease hls or her chance of selecLlon we have Laken Lhe goal Lo be
maxlmum number of lnLervlews per candldaLe Slnce a candldaLe's flnal selecLlon greaLly
depends on hls/her ablllLles whlch Lhough can be lmproved wlLh Lralnlng are sLlll very much
unconLrollable varlables we have noL consldered Lhe same as our flnal goal

1he process beglns wlLh submlsslon of Cvs and/or forms Lo Lhe companles by Lhe sLudenLs
Slnce Lhe Cv submlsslon ls done onllne whlle Lhe forms are submlLLed onllne and/or ln hard
coples over a perlod of Llme much before Lhe flnal lnLervlews lL does noL consLlLuLe Lhe
boLLleneck for Lhe overall process 1he oLher selecLlon sLages such as onllne LesLs
(psychomeLrlc/analyLlcal or oLherwlse) also are planned over a couple of weeks or so and
hence do noL consLrlcL Lhe ouLpuL

L ls durlng Lhe placemenL week amldsL Lhe mulLlple recrulLers Cs and lnLervlews LhaL
opLlmlzaLlon of Lhe operaLlons really comes ln Slnce Lhe sLudenLs shorLllsLed for flnal C
and rounds ofLen are common and also sLudenLs ofLen have mulLlple shorLllsLs schedullng
of Lhem Lo ensure 0 absence/ mlssed chances ls cruclal Slnce Lhe recrulLers are sLaLloned
ln dlfferenL locaLlons around Lhe campus (guesL house Chanakya Llbrary eLc) sLudenLs
have Lo rush Lo aLLend Cs/lnLervlews whlch are ofLen rescheduled aL Lhe lasL momenL
Agaln more ofLen Lhan noL Lhe lnLervlews have Lo be posLponed aL Lhe lasL momenL due Lo
nonavallablllLy of Lhe sLudenLs who are busy aLLendlng a dlfferenL lnLervlew/C So
alLhough Lhe sysLem does work and Lhe sLudenLs are ulLlmaLely aLLend all Lhe lnLervlews
Lhey are ellglble for Lhere ls cerLalnly a scope for lmprovemenL

We belleve Lhe Lhree measurable quanLlLles ln Lhls process

1hroughpuL no of lnLervlews/candldaLe
nvenLory 8ecrulLers vlslLlng Lhe campus candldaLes
CperaLlonal Lxpense 1he work done by Lhe sLudenLs placemenL cell
8oLLleneck 1he lnLervlews (Slow rocess)

Lach lnLervlew handled by a sLudenL can Lake as much as an hour 1hls llmlLs Lhe number of
lnLervlews he can aLLend ln a slngle day 1herefore Lo opLlmlze Lhe enLlre process Lhe
placemenL cell musL Lry Lo maxlmlze Lhe Llme a sLudenL ls aLLendlng lnLervlews and reduce
ldle Llme when he ls dolng noLhlng AnoLher aspecL LhaL reduces Lhe speed ls Lhe dlspersed
locaLlons of Lhe lnLervlew panels of Lhe companles 1he placemenL cell musL Lry Lo geL all Lhe
companles ln a slngle bulldlng and allocaLe Lhem dlfferenL classrooms 1hls wlll reduce Lhe
lag Llme a sLudenL has Lo wasLe when runnlng from one place Lo anoLher (llke from
8odhlgrlha Lo CuesL Pouse)