半導(dǎo)體生產(chǎn)面積預(yù)測(cè)不確定性的量化.pdf

1、電子科技大學(xué)碩士學(xué)位論文半導(dǎo)體生產(chǎn)面積預(yù)測(cè)不確定性的量化姓名：馮文雅申請(qǐng)學(xué)位級(jí)別：碩士專(zhuān)業(yè)：機(jī)械制造及其自動(dòng)化指導(dǎo)教師：李輝20060301ABSTRACTIntheindustrymanufacturingmanagement，marketingdemand，productfequigementandequipmentperformanceefficiencyagethekeyvariablesdeterminingfactory，s

2、capacityTheuncertaintyofthesevariablesCaUSeSthevariabilityoffactory’scapacityplanningespeciallyforlongrangecapacityplanninginsemiconductormanufacturingbecauseofitsshorteningproductlifecycle，highequipmentcost，etcTheuncert

3、aintyissueoflongrangecapacityplanningprocessinsemiconductormanufacturingwasstudiedinthisthesisThekeyuncertainvariablesinthecapacityplanningprocessshouldbedefinedfirstlyThentheeffectofthesekeyuncertainvariablestOcapacitys

4、houldbequantifiedFinallyanewdecisionmakingmethodwouldbeproposedtoensurethattheoperationmanagerofthefirmcouldmakeaconeddecisiononwhetherthefactory’ScapacityshouldbechangedaswellasitsvolumeConsequentlytheuncertaintyofthelo

5、ngrangecapacityplanning1nsemiconductormanufacturingcouldbecontrolledeffectivelyThenthehighleveluncertaintyanalysisoflongrangecapacityplanning’。?！甪actorysspaceresourceplanningwasdiscussedindetailBasedonthehistoricalspac

6、eforecaStdataandcorrespondingactualdataprovidedbyaglobalsemiconductorassembivandtestcompany，theuncertaintyofspaceplanningwasdefinedDuringthisanalysisprocess，linearregression，greyprediction，neuralnetworkbackpropagationaIg

7、orithmandconfidenceintervalwereapplied，respectively，tOdefinetheuncertaintyComparedwiththosemethods，theconfidenceintervalofhistoricalspaceforecasterrorcalculatedbvmathematicalstatistics，wasthereasonablemethodtodefinethesp

8、aceforecastinguncertaintyBasedOnthedefinitionofthespaceforecastinguncertaintyValueatRiskfabbreviatedasVaR)modelwasdevelopedtoidentifytheriskcausedbytheuncertaintyofspaceforecastingThemostconcernintheVaRmethodistodefineth

9、edistributionofthepopulationofspaceforecasterrorwhilenormaldistributionwasadoptedtocale：ulatethecriticalVaRvalueaccordingtotheacquiredsamplesandexpertexperienceItwasshownthatthenewdecisionmakingmethodbasedonVaRwasimprove