View
221
Download
2
Category
Preview:
Citation preview
5
AdvancesinProductionEngineering&Management ISSN1854‐6250
Volume10|Number1|March2015|pp5–17 Journalhome:apem‐journal.org
http://dx.doi.org/10.14743/apem2015.1.188 Originalscientificpaper
An implementation of lean scheduling in a job shop environment
Haider, A.a,*, Mirza, J.a aUniversity of Engineering and Technology, Department of Industrial Engineering, Taxila, Pakistan
A B S T R A C T A R T I C L E I N F O
Globalization has demanded innovative manufacturing and continuous im‐provement inordertostaycompetitive.Thisneedhascompelledthemanu‐facturingworldtodevisestrategiesforproducingcost‐efficientpartswithoutcompromisingquality.TheToyotaProductionSystemwasatthebeginningofsuch initiatives. It was successful in addressing cost through elimination ofnon‐value‐addedtimeandqualitybymonitoringandcontrollingtheproduc‐tionsofdefectiveparts.LeanthinkingoriginatedfromtheToyotaProductionSystemandinheriteditsconceptsandmethodology.IncontrasttotheToyotaProductionSystem,theimplementationof leanhasbeenproposedinalmosteverydomainoflife.Inthemanufacturingdomainitisacommonmisconcep‐tion that lean is suitable formass production only. This research has beenbuiltuponthebeliefthatleanisforeverythingandhaschallengedthisstereo‐type by implementing it within a job shop environment. A manufacturingindustrywasselectedthatwasrebuildingbattlefieldtanks.Theexistingsys‐temwas sufferingdelays andmissingdelivery targets due to uncertain andcostly production. The proposed and existing systems were modeled andsimulated using Arena 10.0 software. Thisworkwas successful in reducingthe manufacturing‐led time, work in process inventory and average cycletimeswithareductionincostandspaceutilization.Costbenefitanalysiswasperformedshowingthattheproposedsystemwouldbebeneficialafter1500parts. We are further expanding our proposed approach towards the toolmanufacturingshopinordertostudytheimpactofleananditssuitabilityforschedulinginjobshops.
©2015PEI,UniversityofMaribor.Allrightsreserved.
Keywords:ManufacturingToyotaProductionSystemLeanthinkingJobshopproductionOnepieceflowManufacturingsimulation
*Correspondingauthor:aftab775@yahoo.com(Haider,A.)
Articlehistory:Received30October2014Revised4February2015Accepted8February2015
1. Introduction
Manufacturingindustryhasgonethroughevolutionarychangesinlastfewdecades.Standardi‐zationandinterchangeabilityhashelpeddevelopingcountriestoearntheirshareofmanufactur‐ingduetolowerlaborcosts.Competitivenessandeconomicchallengeshavedivertedmanufac‐turing activity fromWest to East. Industries with obsoletemanufacturing systems are losingbusinessduetoexcessiveproductioncostsanduncertainproductiondeliverytimes.Thesefac‐torshavestrengthenedmanufacturingindustriesinEasternworld[1].Japanintroducedinnova‐tivemanufacturing system thatwas later knownasToyotaProduction system (TPS) [2]. Thissystembelievedinidentificationandeliminationofsevencriticalwastesinproduction,supplychainandmanagementprocesses[1].QualitywastheessenceofTPSphilosophythathelpedittorevolutionize themanufacturing industry[1,3].Leanthinking isanoffspringofTPS[3,4].Marketchallengestomanufactureatlowercostwithlesstimeandmorethroughput,pavedwayforleanmanufacturing[4].Leanmanufacturingbelievesinsystematiceliminationofwaste,re‐
Haider, Mirza
6 Advances in Production Engineering & Management 10(1) 2015
liesoncontinuousflowconceptsandcustomerpull.Thismanagementsystemoverwhelminglysucceedsinsatisfyingcustomersondelivery,qualityandpricethrougheliminationofnon‐value‐addedactivitiesandwastes[3].Thismanufacturingphilosophybelievesineliminationofwastesforentiresupplychainandaimstoprovidegoodqualityproductsthroughlowprocessingandcycle times,andmoreresponsiveness tocustomerneeds[5].WiththesuccessofTPS in Japanand a tremendous boost in Japan’s manufacturing activity in 1970’s, USmanufacturers wereforcedtoreviewtheirexistingfordsystemandanalyzethesuccessstoryofTPS[3,4].Eventual‐ly, lean manufacturing motivated by TPS appeared in American factories [3]. It became im‐portanttogetleantostaycompetitiveandsucceedinchallengingmarketplace[4].Valuestreammapping,onepieceflow,5Ssystem,quickchangeover,Kanban,cellularstrategyandtotalpro‐ductivemaintenancearetoolsofleanmanufacturingtoimprovequality,costanddelivery[6].
Waste,inleanparadigm,isanon‐value‐addedactivitythatputsextraburdenonthecustomerandcustomerisdefinitelyreluctanttospendonit[3,4].Overproductionistheworstinsevenbasic wastes in production activity. Lean thinking strongly opposed the concept of “make tostock”andstressedtheneedfor“maketoorder”.Overproductionresultsfromoverengineering,misuseofautomation,poorscheduling,and just incase logic.Balancing theassembly lineandadjustingtheproductivitymayhelptoovercomethiswaste.Anotherimportantwasteinmanu‐facturing is longwaiting times.Thismayoccurdue tosaturationofwork loadonsomework‐stations.Adjustingtheprocesstimesandmakingthesystemflexibletocopewithbreakdownsmaybehelpful to eliminate thiswaste.Overproduction and longwaiting times result in largework in process inventory. One piece flow strategy is an effective way to deal with this in‐efficiencyofproductionsystem[7].Longprocessingtimesandwaitingtimescontributetolargeworkinprocess.Transportationisconsiderednon‐value‐addedactivityandmustbeminimizedtoimprovethecycletimesofpartproduction.Thiswasteisresultantofpoorfacilitydesignandlayout and large batch sizes. Thiswaste also contributes to largerwork in process inventory.Unduemotionof thework force,equipmentandmachines isanotherwaste thatcauses largerleadtimes.Thisisalsoduetopoorfacilitylayoutandimproperlocationofmachinesandequip‐ment.Avoidingtheunnecessarymovementsmaybehelpfultoreduceworkinprocessandim‐prove the lead times and reduce cycle times. Production of defective products iswell knownwasteandeffortstocurtail thiswastearecoveredundertheumbrellaofqualitymanagement.Identification and fixing the defects is not the real purpose of quality management. Effortsshouldbetoidentifythecausesofpoorqualityandadoptmethodstoeliminatere‐occurrenceofdefects.Utilizationofresourcesmustbeoptimalandproperlyplanned.Underutilizedresourcesincreasethecostofproductandmaketheworkinprocess(WIP)inventorylarger[3,7].Ideallyaproductionprocessmustbefreefromthesesevenwastes.Thesewasteshamperthebusinessperformance andmake the production activity expensive and costly. Leanmanufacturing en‐sureseliminationofwastesintheoverallmanufacturingprocessandhelpstoreducethecostofproduction.Leanisconcernedwithimprovementinentireprocessflowinsteadofoneormoreindividualprocesses[1,8].
Leanmanufacturingisknownforitssuccessstoriesinreducingcostandimprovingthemar‐ketsharethroughbetterqualityformassproductionindustries[4].Ourresearchisconcernedwithexploringthefeasibilityofleaninajobshopenvironment.
MaroofiandDeghan[9]haspresentedaconceptualframeworkforpossibleimplementationof lean in job shop environment. Proposed model uses LET project that comprises businessprocedure management, supplier management, and value systemmanagement. They devisedtwophrased solution for suppliermanagementusing fuzzy logic and ant colonyoptimization.They suggested separate value streammap for each product due to high variety of products.Value streammap can be used for better scheduling of parts and can be helpful to eliminatewastesandnon‐value‐addedactivities.Ourproposedapproachhasaddressedthesameproblemthrough modeling and simulation. Instead of proposing separate value stream map for eachproduct, we have made necessary changes in the layout and reduced the waiting and queuetimes through sequencing of the parts.Weprovided onepiece flow for process improvementandsuggestedre‐arrangementofworkstations.
An implementation of lean scheduling in a job shop environment
Advances in Production Engineering & Management 10(1) 2015 7
Eng andChing [10] claimed that lean is not suitable for all situations andpresentedquickresponsemanufacturing as an alternative for job shop environment.Theybelieved that quickresponsemanufacturing is suitable for low volumes and high variety and can be successfullyused to reduce critical path times. Our research was to challenge this stereotype and wassuccessfulinprovingthatleancanbeusedforhighvarietyenvironmentaswell.
Assaf[11]usedprogrammeevaluationandreviewtechnique(PERT)toaddressthejobshopscheduling problem. Processes were throughly studied and then author suggested newsequencing of processing using PERT. Author used parallel sequencing for independentprocesses and was successful to reduce the lead times. Our research also re‐arrnages theworkstationsafter thoroughstudyof existing system.Weusedexpert judgment to re‐arrangeour 10workstations problem. Usability of PERT for large number of workstations cannot bedenied but implementation will require formation of groups for different part families. Thismethodology is similar to already existing group technology and cellular manufacturing. WehaveusedpartandprocessesmatrixinsteadofPERTtoseparatethepartfamilies.
Modrák and Semančo [12] presented the cell design methodology to transform job shopproductionprocess to lean.Theydefineddecisionmaking rulesandprinciples toachieveOnePiece flow for job shop. Similarly,wehavedevised theOnePiece flow forour case studyandimplementationhasresultedinachievingtheWIPequaltoworkstations.
Irani [13] believes that there is no specific lean tool that is ideal for job shop. It is alwaysbetter to blend these tools andmethodologies to prepare a customized recipe, suitable for aspecific job shopenvironment.Authorhasnotprovidedany specific solution for the job shopandonlydiscussedprospectsandconsequencesofdifferentmethodsandtools.
Djassemi[14]introducedthreesteppedleanimplementationprocess.Itconstitutedtraining,Kaizan continuous improvement and implementation. Author implemented this approach onpilot projects and identified the improvements made during continuous improvement phase.Thisapproachwassuccessfultoreducetheovertimesby37%andimproveon‐timedeliveryby11%.Thisapproachisaltogetherdifferentfromourapproach.Thisapproachreliesoncontinu‐ousimprovementmethodologyandourresearchisconcernedwitheliminationofwastes.
Section1outlinesemergenceofleanfromToyotaProductionSystemandbriefreviewofleaninjobshopenvironment.Insection2,wehavebrieflyidentifiedthecommonproblemsfacedbyjobshops.Section3isabouttheperformancecriteriaandmeasuresusedtocompareleanwithexistingsystem.Insection4and5,wehaveintroducedandexplainedtheexperimentalsetupandresults.Section6isthelastbutnottheleastthatconcludesourresearchworkandexplainsfuturedirections.
2. Lean in job shop
Manufacturingisabusinessactivityaimedatproducinggoodsandprovidingservicestosatisfyhumanly needs. Through value addedphysical andmental labor, rawmaterial is transformedintousefulproduct that satisfies thedemandsof customers. Suchvalueadditionactivitiesareknownasmanufacturingprocessandoverallcombinationoftheseprocessesmakesamanufac‐turingsystem.Manufacturingsystemscanbeeitherproductorientedorprocessoriented.Pro‐cessorientedprocessesprovide continuousproductionandareknownas continuousproduc‐tion systems whereas product oriented manufacturing processes are known as discrete partmanufacturing. Discrete part manufacturing systems are further categorized as low, mediumandhighbasedonthequantityproducedbyanindustry.Therecanberangeofproductsbeingmanufacturedbyanindividualindustry.Thisrangeofproductscanbeeithersimilarordifferenttoeachother.Rangeofproductsisknownasvarietyandhighvarietylimitsthequantityofpro‐duction.Highvarietyresultsinlowvolumesofproductionsandlowvarietymayguaranteehighproduction[7,15].
Jobshopisalowvolumehighvarietymanufacturingenvironment.Inordertoproducerangeofproducts,ajobshoprequireshighlyskilledandversatileworkforceandflexiblemanufactur‐ingcapability.Automationandspecializationinsomespecifictaskarenotsupportedinjobshopenvironment. Job shops are characterized by fixed position layout,where product remains at
Haider, Mirza
8 Advances in Production Engineering & Management 10(1) 2015
single location during the entire production process. Workforce and equipment move to thefixedproductforvalueadditionactivities.Ships,submarines, locomotives,aircraftsandbattle‐fieldtanksmanufacturingaresometypicalexamplesofjobshopenvironmenthavingfixedposi‐tionlayout[7,16].
Leanmanufacturingcandealwithmissingorderdates,highproductioncosts,declineinmar‐ketshareand limitedcapacity. Is leanphilosophyequallysuccessful in jobshopenvironment?Wehaveselectedadefenseorganizationhavingjobshopenvironmentandmanufacturingandrebuildingbattlefieldtanks.WefoundthatPrecisionDefenseOrganization(PDO)isfacingprob‐lemsin:
Manufacturingandrebuildingof sub‐componentsofbattlefield tankswell in time toen‐surecommitteddeliveryoffinalproduct.
Optimalutilizationofresourceswithinabilitytoidentifybottleneckworkstations. Determinationofexactproductioncapacitybeforemakingcommitmentswithcustomers.
Middlemanagementremainsundertremendouspressuretomeetunrealistictargets.Despiteextrashiftsandundueexpenditureonovertime,targetsaremissedandlingeron.OurfocuswastodeterminethebenefitsoftheutilizationofleanthinkinginPDO,becauseitiscommonmisun‐derstandingthat leanmanufacturingissuitableformassproductionsystemsonlyandwillnotbesuccessfulinjobshopenvironment[17].Thisworkstudywasanendeavortoaddresstheseproblemsandensuresmoothproductioninjobshopenvironment.
3. Performance measures and evaluation criteria
Inordertodeterminetheusefulnessofourproposedsolutionfortheimprovementsinexistingsystem,wehaveidentifiedsomeperformancemeasures.Theseperformancemeasuresare:
1. Workinprocess(WIP)inventory2. Manufacturingleadtime3. Averagecycletime4. Throughput/Productivity5. Costreduction6. Workplacearea7. Deliverycommitments(meantardiness)
3.1 Work in process (WIP) inventory
Inprocesscomponentsinasystemforsomeperiodoftimeareknownasworkinprocess(WIP)inventory.WIPisconsideredhighlysignificantfactorinproductionsystemaslargesizeofWIPincreases production costs. Optimally, the size ofWIP should be equivalent to the number ofworkstationsinthemanufacturingsystem.
3.2 Manufacturing lead time
Timefromreleaseofanordertomanufacturingoffinishedproductiscalledmanufacturingleadtime and is inclusive of processing time, wait time, inspection and transportation time.Manufacturing lead time includesvalueaddition andnon‐valueaddition times.Manufacturingleadtimescanbereducedafterexcludingallorsomepartsofnon‐valueadditionactivities.
3.3 Average cycle time
Inmanufacturingleadtimes,thetimespentonvalueadditionactivitiesiscalledcycletime.Itisprocessing time to transform raw material into finished product and excludes wait,transportationandqueuetimes.
3.4 Throughput/Productivity
Rateofproductionistermedasproductivityorthroughputformanufacturingactivity.Workinprocess (WIP), manufacturing lead times and average cycle times are primary performance
An implementation of lean scheduling in a job shop environment
Advances in Production Engineering & Management 10(1) 2015 9
measures. Productivity is dependent upon these primary measures and can be termed assecondary performancemeasure.Manufacturing lead time (MLT) is inversely proportional toproductivity.HigherMLTresultsinlowerproductivityandviceversa.
3.5 Cost reduction
Anincreaseinproductivityresultsincostfriendlymanufacturingactivity.Wehaveselectedthisperformancemeasure to compare the existingmanufacturing systemwith the proposed. Costreduction is the ultimate requirement of any business activity and its importance cannot bedenied.
3.6 Workspace reduction
Another important aspect of lean manufacturing is to optimally utilize the space formanufacturing and production. We will review and compare the workspace utilization forexistingandproposedsystem.ReductioninworkspacecanbeguaranteedthrougheliminationofsevenwastesofproductionactivityasdescribedbyJustInTime(JIT)andleanthinking.
3.7 Delivery commitments
Industries determine production capacity to commit delivery targets with customers. Thesecommitmentsmaybebasedonexpertjudgmentofoperationsmanagerormodelingtheexistingsystem. We have found that PDO is committing the targets and deliveries based on theirexpertiseandpreviousexperiences.Wepreferredtomodeltheexistingsystemtodeterminetheexact production capacity of PDO.We have identified lateness and tardiness as performancemeasurestogaugedeliveryfulfillmentperformancemeasureofPDO.
Latenessofajobisthedifferencebetweentheduetimeandactualdeliverytime.Preferably,lateness should be positive or zero. In case of late deliveries and inability to meet targetedcommitments,latenessmaybecomenegative.Ideally,occurrenceofnegativelatenessshouldbeavoided.
Tardinessofajobisthemaximumvalueoflatenessandisalwaysnegative.Anoccurrenceofdelivery commitment before the targeted deadline is called earliness and is not part of ourperformancemeasures.
4. Experimental study
Abattlefield tankcomprises threemainmechanical assemblingunits, i.e., gunbarrel,hull andturret. In our study at PDO,we have selected hull assembly for our experiment and analysis.Since PDO is busy in rebuild and manufacturing of battlefield tanks, we have selected hullrebuildandrepairshopinvolvedinrepairofsuspensionandpowerpackpartsandcomponentsasperoriginalengineeringmanual(OEM),inhullassemblysectionofPDO.OverallprocessflowforrebuildofsuspensionpartsisgiveninFig.1.
Fig.1 ProcessflowofPDOjobshop
OUT
No
Yes
IN
Process
analysis Machining Inspection
Disassembly
Degreasing
Welding
Chrome
plating
Nickel plating
Inspection
OK
Scrap
Haider, Mirza
10 Advances in Production Engineering & Management 10(1) 2015
Ourresearchisconcernedwiththemachiningphaseofsuspensionparts.Thesepartsincludebalancearm,crankarm,sprockethub,drivengear,drivenshaft,finaldrive,idlewheeldisc,leftright supports, shock absorber blade andworm gear. An individual tank assembly requires 2parts of each except balance arms and shock absorber blades. There is requirement of 10balance arms and 4 shock absorber blades for typical Chinese and Russian origin battlefieldtanks. Each part has different routing and processing requirements. These are highly andfrequently wearing out parts in suspension and power pack assemblies of hull section inbattlefield tanks.Thesepartsdelay the finalassemblyofhullsectionandcontribute tooveralldelay inbattlefield tank assembly. Shop floor involved in repairingof theseparts is equippedwithcenterlathes,gearlathes,radialdrillingmachine,broachingmachine,verticallathe,verticalanduniversalmills,broach,benchdrillingmachine,universalgrindingmachineandinspectioncumbenchfitting.Thesearethosemachinesthatareselectedaftermakingcellsandgroupsofsimilar parts. Each part has different sequence of operations, e.g. sequence of operations forsprocket hub is, vertical lathe (turning), broaching,milling, internal grinding and for balancearm, sequence is turning, milling, cylindrical grinding, heat treatment, surface treatment andinspection.
Ourproposedapproach(Fig.2)comprisesthreemainphases.Firstphaseisconcernedwithstudyofprocessesandlayoutsandidentificationofpartfamiliesandformationofcells.Secondphase iscontinuous improvementphase. Inthisphase,were‐arrangedtheworkstationsusingthesequencingchartofpartsingroupsuchthattherewasnobackwardmovementofthepartduringtheprocessing.Apartentersfromonesideofthecell,movesaheadanddepartsfromtheother side after value addition.We identified delays through analysis ofwaiting times, queuetimes,arrivaltimesandprocessingtimes.Thirdphaseimplementsleanthinkingandusesquickchangeover, total preventive maintenance, elimination of wastes and Kaizan methodologies.UsingKaizancontinuousimprovementmethodology,were‐adjustedarrivaltimesandreducedwaiting times through increase in resources and provision of quick change over.We reducedprocessingtimethroughimprovementintimetofailureandreductionofbreakdowntimes.ThishelpedustoestablishOnePieceflowinjobshop.
We have used manufacturing simulation software Arena 10.0 [18] to model existing andproposed scenario of the PDO case study. Existing system has spaghetti layout with wovenroutingofparts for valueadditionprocesses.Wehave generatedpart families for thesepartsandre‐arrangedtheplacementofworkstationstoprovideaU‐shapedcellularmanufacturing.
Fig.2 Proposedmethodology
Job shop selection
Process and layout
study
Generation of part‐
process matrix
Identification of part
families
Formation of cells
Re‐arrangement of
workstations
Modeling in Arena
10.0
Simulation and
analysis of results
Identification of wastes & bottlenecks
One piece flow
Process improvement
Elimination of non‐
value added times
Re‐estimation of
resources
Simulation
PHASE 1 PHASE 2 PHASE 3
An implementation of lean scheduling in a job shop environment
Advances in Production Engineering & Management 10(1) 2015 11
5. Experimental results
WemodeledandsimulatedourproposedandexistingmanufacturingsystemsusingArena10.0andwehavecomparedthesethroughpredefinedperformancemeasuresdiscussedinSection3.
5.1 Work in process (WIP) inventory
HigherWIPareneitherpreferrednorwelcomed.LowerWIPresultsinlowermanufacturingleadtimes.Inexistingsystem,WIPwas22.85,butleanschedulinghelpedustoreduceitto10.OnePiece flow states that number of parts in process should not be more than the number ofworkstations.Thereare10workstationsinPDOsuspensionrebuildjobshop.Wehaveobservedthat lean scheduling has provided ‘One piece flow’ and has comparatively reduced the WIP,makingitequaltothenumberofworkstationsintheproposedsystem(Fig.3).
Fig.3 WIPinventoryforleanandexistingsystem
5.2 Average cycle time
ValueadditiontimeorprocessingtimeofMLTisknownascycletime.Itistimespenttoconvertarawmaterial into finishedpart. Ifweanalyzetheresults,wecanfoundadrasticdecrease incycletimesforproposedscenario,whereleanschedulinghasbeenimplemented.However,finaldriveandidlerwheelarehavingalmostsameprocessingtimesinexistingandproposedsystems(Fig.4).Thesepartsvisitfewworkstationsascomparedtoothersanddonotundergomilling,drillingandbroachingprocess.
Available time forproductionof theseparts is12000to12120 for5days,8hoursshift. Inexistingsystem,finaldriveandidlerwheeldiscsaretheonlypartsthatcanberebuiltwithinthestipulated time. In proposed lean scheduling, balance arm is consuming longer time than theavailabletime.Otherpartscanberebuiltwithinthescopedtimefortheseparts(Table1).
Fig.4 Averagecycletimesforleanandexistingsystem
ExistingSystem LeanScheduling
WIP 22,85 10
0
5
10
15
20
25
QtyParts
0
500
1000
1500
2000
2500
BalanceArm
CrankArm
DrivenGear
DrivenShaft
SprocketHub
FinalDriveCover
Worm IdlerWheelDisk
ShockAbsorberBlade
Left RightSupports
Lean Scheduling
Existing System
Average cycletime [m
in]
Haider, Mirza
12 Advances in Production Engineering & Management 10(1) 2015
Table1Averagecycletimesforleanandexistingsystem
S.No. Part Qty Existingsystem Leanscheduling1 BalanceArm 100 1034.8 193.242 CrankArm 20 1115 97.753 DrivenGear 20 1507.7 178.654 DrivenShaft 20 1358.3 284.895 SprocketHub 20 757.7 751.116 FinalDriveCover 20 328.57 294.477 Worm 20 1785.4 181.818 IdlerWheelDisc 20 377.91 271.119 ShockAbsorber 40 1065 160.7610 Left,rightSupport 20 1950.9 282.3
5.3 Manufacturing lead time
Manufacturingleadtimesincludetimespentonvalueaddedandnon‐value‐addedactivities.Wehaveanalyzedtheexistingsituationwiththeintenttodiscovertheeffectofnon‐valueadditionon MLT. Unfortunately, the time spent on wastes is too large in comparison to the actualprocessingtimes.Thisscenarioclearlyindicatesthatthereexistsroomforimprovementintheexistingsystemsandnon‐valueadditiontimesmustbedecreasedtoincreasetheproductivityofthesystem(Fig.5). Inorder todecrease thesenon‐processing times,we implementedKaizan,quickchangeoverandOnePieceflowforourproposedsystem.
Fig.5 MLTforleanandexistingsystem
5.4 Lateness in delivery commitments
AfteranalysisoftheresultsinFig.6,wefoundthatproposedsystemisalsohavinglatenessinproductionofonepart,i.e.balancearmswithlatenessofabout4days.
Fig.6 Tardinessforleanandexistingsystem
0
500
1000
1500
2000
BalanceArm
Crank Arm DrivenGear
DrivenShaft
SprocketHub
Final DriveCover
Worm IdlerWheelDisk
ShockAbsorberBlade
Left RightSupports
Value Added Time
Manufacturinglead
tim
e [m
in]
‐100000
‐80000
‐60000
‐40000
‐20000
0
20000
Lean Scheduling
Existing System
Tardines [min]
An implementation of lean scheduling in a job shop environment
Advances in Production Engineering & Management 10(1) 2015 13
Wereviewedthereasonsfordelayinprocessingofbalancearmsandfoundthatthesepartsare being delayed due to extensive load on turning work stations and can be improved byreducing queue times at these workstations. However, existing system was facing severetardinessandwasunabletomeettheproductiontargets. Lateness in the existing system can be related with resource utilization given in section 5.7. Available time for rebuild of these suspension parts approximates from 12000 to 12120 based on 5 days, 8 hours shift. Results also reflect the need to review the planned targets for the job shop (Table 2).
Either the target of balance arms should be re-evaluated or efforts should be made to improve the process time on centre lathes for these parts. This can be done through use of tungsten carbide tooling to avoid unnecessary delays and reviewing the NC program for these parts.
Table2Tardinessforleanandexistingsystem
S.No. Part Existing system Leanscheduling1 Balancearm ‐91360 ‐72042 Crankarm ‐10180 03 Drivengear ‐18034 04 Drivenshaft ‐15046 05 Sprockethub ‐3034 06 Finaldrivecover 0 07 Worm ‐23588 08 Idlerwheeldisc 0 09 Shockabsorber ‐30480 010 Left,rightsupport ‐26898 0
5.5 Throughput
Wehaveplottedachievedthroughputagainstmonthlytargetforexistingsystemandproposedlean scheduling (Fig. 7).We have again noticed that sprocket hubs and balance arms are notmeeting the targeteddeadlines for theproposed system.Wehave found that excessivequeuetimes at turning work station are the reason for this delay and can be further improved ifprocessingtimeandsetuptimecanbereduced for theseparts.Thesepartshavehighersetuptimesandthiscanbereducedthroughuseofsomespecializedfixturestoaccommodatespeedychangeoverofpartsduringmachining.Existingsystem is capable toproduce finaldrivecoverandsprockethubsindesiredtargetedquantity.Itseverelylacksinproductionofbalancearmsandleftrightsupports(Table3).
Thisjobshopmostlyseeksthesupportofsisterjobshopstohelpmeetthetargetedquantity.Our proposed system lacks in sprocket hubs mainly due to non‐availability of broachingmachineandbalancearmsduetoheavyloadoncentrelathes.
Table3Throughputforleanandexistingsystem
S.No. Part Targets Existingsystem Leanscheduling1 Balancearm 100 12 622 Crankarm 20 11 203 Drivengear 20 8 204 Drivenshaft 20 9 205 Sprockethub 20 16 176 Finaldrivecover 20 20 207 Worm 20 7 208 Idlerwheeldisc 20 20 209 Shockabsorber 40 11 4010 Left,rightsupport 20 6 20
Haider, Mirza
14 Advances in Production Engineering & Management 10(1) 2015
Fig.7Productivityforleanandexistingsystem
5.6 Workplace utilization
Lean scheduling has been helpful to reduce the space requirement for parts waiting forprocessing at next stations (Fig. 8). Secondly, it has fairly reduced the work stations withprovisionofsophisticatedworkstationsthatarecapabletoperformmultiplejobs.CNCmillingcenterscanbeusedtoreplacelathes,milling,anddrillingstations.
Fig.8 Spaceutilizationforleanandexistingsystem
5.7 Resource utilization
Resourceutilizationistheratioofavailabletimeandresourceutilizedtime.Ourresearchfounditthatmostoftheworkstationsareunder‐utilized.Someworkstations,i.e.,centrelatheandgearlathe,arecausingunnecessarydelaysandcontributetolargerwaitingtimesforthepartsinthequeue. We identified improvements for these workstations. We suggested fixtures for theseworkstations to reduce setup times. These workstations were suffering lack of tooling formachiningpurpose. Secondly, therewas longer time to replace the faultypartsdue to lackofnecessary inventoryofcapacitors, servomotors,belts,andgears.These improvementshelpedustoimprovethemeantimetofailureandbreakdowntimes.AcomparativestudyofexistingandproposedsystemisgiveninFig.9.Weimprovedutilizationthroughbalancingofprocessingtimesandaddressingthesauratedandstarvingworkstations.
0
20
40
60
80
100
120
BalanceArm
CrankArm DrivenGear
DrivenShaft
SprocketHub
FinalDriveCover
Worm IdlerWheelDisc
ShockAbsorber
Left,rightSupport
PartsQty
Targets
ExistingSystem
LeanScheduling
0
0,5
1
1,5
2
2,5
3
3,5
ExistingSystem LeanScheduling
WorkplaceUtilization
Are
a
An implementation of lean scheduling in a job shop environment
Advances in Production Engineering & Management 10(1) 2015 15
Fig.9 Resourceutilization
5.8 Cost analysis
Manufacturing cost for a product in production setup consists of fixed and variable costs.Variable costs change with the change in level of production activity. However, fixed costsremain constant and are not influenced by production activity. Manufacturing cost ismathematicallyrepresentedas:TotalCost=FixedCost+VariableCost(QuantityofParts)[7,16].
Fig.10 Costanalysisforleanandexistingsystem
Fixedandvariablecostsforexistingsystemare0.42and0.00074MillionPKR(1US$=46
PKR.). Similarly, for proposed system these costs are 1.24 and0.00015MillionPKR.Wehaveperformed the cost benefit analysis forboth systemsand found thatproposed systemwill bebeneficialaftertheproductionof1500parts(Fig.10).PDOisproducing300partsinonemonth(Fig. 11). It can be inferred from this cost analysis that lean system will be beneficial afterpassageoffirstfivemonths.
Theseresultsshowedthat leanschedulingcanbehelpful in improvementofdeliverytimesfor job shop environment (Table 4). In problem statement, we have identified three majorobjectives for our study. These include on‐time delivery, improved resource utilization anddetermination of exact production capacity of job shop.Wewere able to determine the exacttargetsforjobshop.Itwasnotpossibletoproducebalancearmsandsprockethubsaccordingtoplannedcommitmentswithinavailableresources.Thisworkhelpedtoproposesuggestionsforenhancementinresourcestomeetthetargeteddeliveriesofbalancearmsandsprockethubs.
0
0,5
1
1,5
2
2,5
CenterLathe
RadialDrillingMachine
UniversalCylindricalGrinder
BroachingMachine
GearLathe
VerticalLathe
VerticalMill
UniversalMill
BenchDrillingMachine
InspectionCumBenchFitting
Utilization
=ProcessTime\AvailableTime
Existing System
Lean Scheduling
0
500000
1000000
1500000
2000000
2500000
3000000
3500000
4000000
0 300 1000 2000 3000 4000
Cos
t [P
KR
]
Parts Qty
Lean Scheduling
Existing System
Existing System
Lean Scheduling
Haider, Mirza
16 Advances in Production Engineering & Management 10(1) 2015
Fig.11 Productiontargetsofjobshop
Table4Comparisonofproposedandexistingsystem
S.No. Parameter Unit Existingsystem Leanscheduling1 WIP no. 22.85 102 MLTavg Time(min) 2400.12 715.53 Spaceutilization area 3 14 Cycletimeavg Time (min) 1128.13 269.425 Throughput no. 46.45 94.76 Latenessavg Time (min) ‐20851 +4864.987 Utilizationavg ratio 0.53 1.0
6. Conclusion and future work
Leanphilosophy is preferred inmanufacturing organizations due to its ability to produce theproducts at competitive prices. Lean scheduling is conceptually similar to leanmanufacturingand revolves around elimination of wastes, continuous improvement, total preventivemaintenanceandquickchangeover.Ourimplementationof leaninjobshopreflectedthatleanscheduling is possible in job shop as well and can bring positive impact on manufacturingactivity.Itmaybehelpfultoreducethelongleadtimeswithreductioninnon‐processingtimesandimplementationof ‘OnePiece flow’.Scalabilityofourproposedapproachfor largersetupsneedsvalidation.Wehaveimplementedourapproachononeprocess,i.e.machininginjobshop.Thereexistsalotofroomtofurtherexpanditandimplementitonthecompleteprocessflowofsuspension parts. We have made few assumptions about the arrival times of the parts.Sometimes,predecessoractivitiesmayundergodelaysandcannotbecompletedasdesired.Suchdelays will effect our proposed approach. Another drawback of our proposed approach isrequirement toalter the layoutandmake it feasible forcellularmanufacturing.Alteration isacostlyactivityandorganizationsmaynotoptforit.Webelievedthatinventoryofpartstorepairworkstationswill remain replenished all times. Last but not the least aspect of our proposedsystemisabout the fixtures toreducesetuptimesonworkstations.Feasibility tomanufacturethesefixturesneedstobevalidatedforourproposedapproach.Despitetheseconsequences,wehavebeensuccessful toprovidea framework tomake lean jobshops.Ourproposedapproachhas been successful to challenge the stereotype that lean is for mass production and is notfeasibleforsmallersetups.
Our proposed approach can also be used for capacity planning of job shops and provideaccurate targets for production activity. We are working on this concept to determine theproductioncapacityofnewlycommissionedtoolshop.Secondly,weareworkingonaproposaltoprovideidealproductionlayoutforassemblyofrebuiltpartsforT‐80UDbattletanks.
BalanceArm;100
CrankArm;20
DrivenGear;20DrivenShaft;20SprocketHub;20
FinalDriveCover;20
Worm;20
IdlerWheelDisk;20
ShockAbsorberBlade;40
LeftRightSupports;20
An implementation of lean scheduling in a job shop environment
Acknowledgement We are thankful to senior management of PDO defense organization for providing us technical manuals and resources to understand the process, layout and operations of battlefield tank’s manufacturing and access to rebuild and assembly shops. We are also thankful to lab incharge of Mechanical Engineering Department of UET, Taxila in helping us to procure Arena 10.0 for experimental purpose.
References [1] Thornton, P. (2010). The global manufacturing sector: Current issues, Chartered Institute of Management Ac-
countants, London, UK, from http://www.cimaglobal.com/Documents/Thought_leadership_docs/Global_ manufac-turing_report.pdf, accessed February 4, 2015.
[2] Ohno, T. (1988). Toyota production system: Beyond large-scale production, Productivity Press, Portland, USA. [3] Kovacheva, A.V. (2010). Challenges in lean implementation: Successful transformation towards lean enterprise,
Master thesis, Aarhus School of Business, University of Aarhus, Denmark, from http://pure.au.dk/portal-asb-student/files/9093/ak83188.pdf, accessed February 4, 2015.
[4] Fujimoto, T. (2012). The evolution of production systems: Exploring the sources of Toyota’s competitiveness, Annals of Business Administrative Science, Vol. 11, 25-44, doi: 10.7880/abas.11.25.
[5] Kachru, U. (2009). Production & Operations Management, Excel Books, India, New Delhi. [6] Georgescu, D.D. (2011). Lean thinking and transferring lean management – The best defence against an econom-
ic recession?, European Journal of Interdisciplinary Studies, Vol. 3, No. 1, 4-20. [7] Haider, A.A, Jahanzaib, M., Akhtar, K. (2007). Experimental comparison of one piece flow production: A simula-
tion based approach, European Journal of Scientific Research, Vol. 19, No. 1, 97-104. [8] Imai, M. (1986). Kaizen: The key to Japan’s competitive success, Random House Business Division, New York, USA. [9] Maroofi, F., Dehghan, S. (2012). Performing lean manufacturing system in small and medium enterprises, Inter-
national Journal of Academic Research in Accounting, Finance and Management Sciences, Vol. 2, No. 3, 156-163. [10] Eng, C.K., Ching, H.W. (2014). Tool for mapping manufacturing critical-path time in job shop environment, Sci-
ence International, Vol. 26, No. 4, 1585-1589. [11] Assaf, R. (2014). Job shop lean production implementation using program evaluation and review technique
(Pert), In: Proceedings of the 2014 International Conference on Industrial Engineering and Operations Manage-ment, Bali, Indonesia, 1554-1560.
[12] Modrák, V., Semančo, P. (2014). Cell design for transforming the job shop production process to lean, In: Hand-book of Research on Design and Management of Lean Production Systems, IGI Global, 130-147, doi: 10.4018/978-1-4666-5039-8.ch007.
[13] Irani, S.A. (2011). Job shop lean – Continuous improvement: No one solution for all, The Fabricator, Vol. 40, No. 2, 44-45.
[14] Djassemi, M. (2014). Lean adoption in small manufacturing shops: Attributes and challenges, The Journal of Technology, Management, and Applied Engineering, Vol. 30, No. 1, 2-10.
[15] Groover, M.P. (2007). Automation, production systems, and computer integrated manufacturing, 3rd edition, Pren-tice Hall, New Jersey, USA.
[16] Haider, A.A. Lean scheduling in high variety low volume environment, Master thesis, University of Engineering and Technology, Taxila, Pakitan.
[17] Bozzone, V. (2001). Speed to market: Lean manufacturing for job shops, 2nd edition, AMACOM, New York, USA. [18] Kelton, W.D., Sadowski, D.A., Sadowski, R.P. (2001). Simulation with Arena, 2nd edition, McGraw-Hill Science, New
York, USA.
Advances in Production Engineering & Management 10(1) 2015 17
Recommended