工程管理专业,外文翻译

-1-外文资料翻译资料来源：文章名:PredictingEffectivenessofConstructionProjectManagement:Decision-SupportToolforCompetitiveBidding书刊名：AnInternationalJournal作者：RasaApanaviciene,ArvydasJuodis出版社：国际杂志，2006章节：Vol.6,No.3/September-December2006页码：P347~P360文章译名：建设工程项目管理的预测功效：用于决策支持工具竞争性招标姓名：学号：指导教师(职称)：专业：班级：所在学院：-2-外文原文PredictingEffectivenessofConstructionProjectManagement:Decision-SupportToolforCompetitiveBidding1.IntroductionConstructionprojectsaredeliveredunderconditionsofriskinthecompetitivemarketenvironment.Theoriginofriskistheuncertaintyinherenttoanyproject,andeveryriskisassociatedwithacause,aconsequenceandtheprobabilityorlikelihoodoftheeventoccurring.Thereareexternalrisks(economic,political,financialandenvironmental)andinternalrisksbasedonprojectmanagementissues,i.e.projectsmanager'sandhisteamcompetency,experience,strategicandtacticdecisionsmadeduringconstructionprojectdelivery.Theopportunitytoimproveorganizationalperformancethroughmoreeffectiveprojectmanagementcouldprovidesubstantialsavingsforconstructionmanagementcompany.Projectmanagementeffectivenessdependsoncertainfactorsofprojectmanagementsystem.Theliteraturereviewrevealedasubstantialvolumeofworkonmeasuringoridentifyingthefactorsorconditionscontributingtotheeffectivenessofconstructionprojects.Therearethreemaintrendsofpreviousresearchonconstructionprojectsuccessfactors:keyfactorsidentificationforconstructionprojectsuccess[Jaselskiset.A1.(1991);Sanvidoet.A1.(1992);Chuaet.A1.(1997)];identificationofkeysuccessfactorsforaparticulargroupofconstructionprojects,e.g.BOT,design-build,public-privatepartnerships[Tiong(1996);Molenaaret.A1.(2001);Chanet.AI.(2001),Zhang(2005),Shenet.A1.(2005)];analysisofaparticularfactorimpactonconstructionprojectsuccess[Chenget.A1.(2000);Boweret.A1.(2002);Ford(2002)].Somewriterswereattemptingtodeveloppredictivemodelswhileothersfocusedongeneratingalistofpractices.Predictivemodelsdevelopedtoidentifythekeyfactorsandtomeasuretheirimpactonoverallprojectsuccesswereusingregressionandcorrelationtechniques,factoranalysis,Monte-Carlosimulation,expertsandmulticriteriadecision-makingsupportmethods.Essentiallyintheseapproachesthefunctionalrelationshipsbetweentheinputfactorsandprojectoutcomeisassumedandtestedagainstthedata.Therelationshipsaremodifiedandretesteduntilthemodelsthatbestfitthedataarefound.Whendevelopingconstructionprojectmanagementeffectivenessmodel(CPMEM)referredtohere,thewritersattemptedtocullthebestaspectsofartificialneuralnetworks(ANN)methodology.Theneuralnetworkapproachdoesnotrequireanaprioriassumptionofthefunctionalrelationship.Artificialneuralnetworksareveryusefulbecauseoftheirfunctionalmappingpropertiesandtheabilitytolearnfromexamples.Networkshavebeencomparedwithmanyotherfunctionalapproximationsystemsandfoundtobecompetitiveintermsofaccuracy[Haykin1999].Thisandtheabilitytolearnfromexamplesallowmodellingthecomplexconstructionprojectmanagementsystemwherebehaviouralrulesarenotknownindetailandaredifficulttoanalyzecorrectly.2.MethodologyofArtificialNeuralNetworksThefoundationoftheartificialneuralnetworks(ANN)paradigmwaslaidinthe1950s,and-3-ANNhasgainedsignificantattentioninthepastdecadebecauseofthedevelopmentofmorepowerfulhardwareandneuralalgorithms[Haykin(1999)].Artificialneuralnetworkshavebeenstudiedandexploredbymanyresearcherswheretheyhavebeenused,applied,andmanipulatedinalmosteveryfield.Forexample,theyhavebeenusedinsystemmodellingandidentification,control,patternrecognition,speechpronunciation,systemclassifications,medicaldiagnosisaswellasinprediction,computervision,andhardwareimplementations.Asincivilengineeringandmanagementapplications,neuralnetworkshavebeenemployedindifferentstudies.Someofthesestudiescoverthemathematicalmodellingofnon-linearstructuralmaterials,damagedetection,non-destructiveanalysis,earthquakeclassification,dynamicalsystemmodelling,systemidentifications,andstructuralcontroloflinearandnon-linearsystems,constructionproductivitymodelling,constructiontechnologyevaluation,costestimation,organisationaleffectivenessmodellingandothers[Adeliet.A1.(1998),Sinhaet.A1.(2000)].Aneuralnetworkcanbedefinedasamodelofreasoningbasedonhumanbrain[Wasserman(1993)].Learningisafundamentalandessentialcharacteristicofbiologicalneuralnetworks.Theeasewithwhichtheycanlearnledtoattemptstoemulateabiologicalnetworkinacomputer.2.1ModelofArtificialNeuralNetworkAnartificialneuralnetworkconsistsofanumberofverysimpleandhighlyinterconnectedprocessors,alsocalledneurons,whichareanalogoustothebiologicalneuronsinthebrain.Theneuronsareconnectedbyweightedlinkspassingsignalsfromoneneurontoanother.Eachneuronreceivesanumberofinputsignalsthroughitsconnections;however,itneverproducesmorethanasingleoutputsignal.Theoutputsignalistransmittedthroughtheneuron'soutgoingconnection(correspondingtothebiologicalaxon).Theoutgoingconnection,inturn,splitsintoanumberofbranchesthattransmitthesamesignal(thesignalisnotdividedamongthesebranchesinanyway).Theoutgoingbranchesterminateattheincomingconnectionsofotherneuronsinthenetwork.Figure1representsconnectionsofatypicalANN.-4-AsshowninFig