基于ProModel仿真的半导体生产系统研究

上海交通大学硕士学位论文基于ProModel仿真的半导体生产系统研究姓名：张宴申请学位级别：硕士专业：物流工程指导教师：杨东;邓伟20070101ProModelProModelProModelABSTRACTICPRODUTIONSIMULATIONBASEONPROMODELABSTRACTThefastdevelopmentofelectronicinformationindustryenlargesthemarketanddevelopmentopportunityofsemiconductorindustry,alsoenhancethefiercecompetition.AsanimportantportionofICindustry,ICfablessdesignhousemustshortentheirdeliverycycletimetoquicklyrespondcustomer’sdemandsotosurviveatthemarket.Butcurrently,theresearchanddevelopmentofnewproductandproductiondeliveryofICdesignhouseisenslavedtoitssubcontractor’sproductionschedule.Themainproblemthatthesedesignhousesmeetisthedynamicchangeoftheproductioncycletime.ThisresearchisonproductioncycletimebaseonthespecialcharacteristicsofICproduction.ThespecialcharacteristicsofICproductionarethatthedifferentproductionstagelocatedatdifferentfactoryanditslongdynamiccycletime.Thisresearchusesdiscrete-eventsimulationmethod,andbaseonrealproductiondataandinformation,simulationwithProModelformodeling.Beconsideringthedetailproductionprocessandshippingprocedure,suchcharacteristicsasresource,differentactivitydurationnecessarytime,probabilitydistribution,yieldrate,capacityallocation.SimulationandanalysisICproductionsysteminthecasesuchasaverageproductioncycletime,usageofthequantityofresource,resourceallocation,factorycapacityneck,etc.Thesimulationresultcanbeusedforproductionplanning,shortenproductioncycletime,reduceinventory,forecastandguaranteedeliverytime.KEYWORDSIC,ProductionCycleTime,ProModel,ProductionSystemSimulation20071262007126200712611.1[1][2](WIP)2,ProModel1.2PetriPetriPetri[3]IDEF0PetriIDEF0Petri,[4]Hung3RulkensEXTEND[5],[6]Yi-FengHungLeachman,R.C.[7]A.I.Sivakumar[8]ProModelProModel1.31.3.14A(1)(2)(3)(4)(5)1.3.2ProModel51.4ProModelStatFitProModelProModel672.12.1.11959integratedcircuit1962[9]20503000300[10]45%PCPC9320008PC[10]GDPGDP[10]2.1.22070IDM2080fablessdesignhouseF&FFoundry()+Fabless()FoundryICFablessFablessFoundryFabless2090foundry21[10]191fig1StructureofICindustry87~81212~15500IPIC2.22fig2ICproductionprocess10812waferfabrication200~300Fabwafersort/probe3fig3WafermanufacturingandwafersortingprocessAssembly/Packaging11FinalTesting4fig4Assemblyandfinaltestproductionprocessfront-endback-end,12,[11]MeanCycleTimeSumofprocesstime2.510[12]Re-entry2.3AAA13IC…………ICIC…………ICIC…………IC5fig5SupplychainmodeofICdesignhouseAA2.414A1512345673.13.1.1121633.1.2,[13][14]173.1.31961G.W.Morgenthater1978Korn1982Spriet[15]1984Oren[16]()6186fig6Mainfactorsofsimulationandtherelationship3.1.4:1234519678910()3.2()(Discrete-EventSimulation)(ContinuousSimulation)(CombinedDiscrete-Continuous20Simulation)(MonteCarloSimulation)3.2.112312342153.2.2[16]1222343.323123454.1123AX200511200663004.224inNnPiii==;JMPSPSS4.2.1StatFitStatFitProModelProModelStatFitAX1AXStatFit7StatFit36StatFit68StatFit910251X()()14191232039352124522657231361024174256872619222751019287112129912393012131431714483211513331516163421753571883687StatFitfig7DemandintervaldatainputinStatFit268StatFitfig8DescriptivestatisticsofdemandintervaldatainStatFit9StatFitfig9DemandintervaldatadistributioncompareinStatFit2710StatFitfig10DemandintervaldatadistributionfitteddensityinStatFitProModelE10.894.2.2ProModelStatFit10Exponential(1.,9.89)β=9.89min=128[]≤=−−min0minexp1)()min](xxxfxββ(4.1)[]≤−=−−min0minexp1)()min](xxxFxβ(4.2))(xFU=xmin)1ln(+−−=Uxβ(4.3):U0,1β=9.89min=13011U=0.27,2U=0.89,3U=0.13Exponential(1.,9.89)11.41=x8.222=x38.23=x4.2.34.11E(10.89)dayL(85.6,51.8)K2a)300b)12c)2293ProModel100SPSS12a)3b)3c)30.0040.0050.0060.0070.0080.0090.00100.00CT_wafer010203040FrequencyMean=63.5467Std.Dev.=11.62098N=289Histogram11fig11Distributionchartofwafermanufacturingcircletime302Valid289NMissing0Mean63.5467Std.ErrorofMean.68359Median63.0000Mode59.00Std.Deviation11.62098Variance135.047Skewness.176Std.ErrorofSkewness.143Kurtosis-.028Std.ErrorofKurtosis.286Range63.00Minimum33.00Maximum96.00Sum18365.00N(64,12)daysd)2.004.006.008.0010.00CT_assy010203040506070FrequencyMean=5.9046Std.Dev.=1.51768N=241Histogram12fig12Distributionchartofassemblycircletime313Valid241NMissing0Mean5.9046Std.ErrorofMean.09776Median6.0000Mode6.00Std.Deviation1.51768Variance2.303Skewness-.003Std.ErrorofSkewness.157Kurtosis-.809Std.ErrorofKurtosis.312Range6.00Minimum3.00Maximum9.00Sum1423.00N(6,2)dayse)121~212HR324/E(10.89)day//L(85.6,51.8)K/300lotsN(64,12)day/1/3day/12lotsN(6,2)day/2/3day/2lots12hour80%80%=85%*99%*95/2days/3/3days/4.3ProModelStatFitPROMODEL33ProModelProModel6.0ProModel5.1ProModelProModelProModel5.1.1ProModelProModel(ProductionModeler)ProModelPROMODEL34,ProModel.ProModelProModel:()();what-ifProModel(GUI)WINDOWS98/2000/XPWINDOWSProModelProModelProModelPROMODEL35[17]5.1.2ProModelProModelProModel[18](1)ProModel:(Entity)(Location)(Resource)()()().PROMODEL36ProModel(Path)ProModel(2)ProModelif-then(Routing)(EntityOperation)(EntityArrival)QtyEach/Occurrences(Frequency):PROMODEL37(EntityandResourceMovement):(DowntimeandRepairs)ProModel:(Clock)(Usage)T-(Cycle)(Setup)T-(Shift)ResourceSchedulingProModel.5.2ProModelPROMODEL385.2.1AX1