Class7技术-2风险分析的方法体系.ppt

TheToolsandMethodologiesofRiskAnalysis,工程项目风险管理与保险,ContentsofRiskAnalysis,Analysisofriskeventsevolution;RiskEvaluationfor:occurringprobabilityofriskevents,variedpossibleriskconsequences,andtheirconditionalprobabilitiesgivenoriginaleventoccurring.,Analysisofriskeventevolution,FaultTreeAnalysis(FTA,故障树)EventTreeAnalysis（ETA，事件树）,Whatisafaulttree？,Isaspecialkindofinvertedtreelogiccausalitydiagram,itUSESeventsymbols,transferoflogicgatesignsandsymbolstodescribeacause-and-effectrelationshipbetweenvariouseventsinthesystem.Logicgateinputoutputeventsbecause,logicgateoutputistheinputeventsfruit.,Theprocedureoffaulttreeanalysis,1.Befamiliarwiththesystem2.Investigateaccidents3.Determinethetopevent4.Determinethetargetvalue5.Reasonsforinvestigation6.Drawthefaulttree,Theprocedureoffaulttreeanalysis,7.Analysis8.Theaccidentprobability9.Comparing10.Analysis,Supervisionofprojectfailure,Fromtheownersrisk,Fromthecontractorsrisk,Fromtheriskofsupervision,X1,X2,X6,X3,X4,X5,X7,X11,X10,X8,X9,ACase:Typicalrisk,X1:feasibilitystudyreportlackofseriousnessX2:macromanagement,investmentdeficiencyTheX3:blindintervention,deprivedoftherightsoftheengineeringconsultantX4:dontrespectcontract,norespectfortherightsandinterestsofthecontractorX5:projectpaymentdoesnotreachthedesignatedpositionorsettlementinatimelymanner,Fromtheownersrisk,X6:contractorbiddingnothonestBiddersX7:lackofbusinessethicsTheunder-fundedbyX8:contractorsX9:contractorqualityistoolow,Fromthecontractorsrisk,Eventtreeanalysiscase,Thewatersystemoftheaccidenteventtreeanalysis,ProbabilityMethods,UsingtheexistingdataanalysisriskeventprobabilitydistributionTheoryofprobabilitydistributionisusedtodeterminetheprobabilityofriskeventsTheprobabilityofriskeventsusingsubjectiveprobabilityanalysisAnalogyisusedtoderivetheprobabilityofriskevents,Usingtheexistingdataanalysisriskeventprobabilitydistribution,Case:theemergenceofconstructionperioddelayriskprobabilitycalculationAconstructioncompanyinthepasttocomplete72constructionprojects,duetovariousreasons,somepartoftheprojecthasdelayedtimelimitforaproject.Draggedprojecttimelimitforaproject,arrangethemgetsituationasshowninthetable.Thetimelimitforaprojectinthetablearetherelativevalueofdelaytime;Thenumberoffrequencyfortheproject.,Theconstructionperioddelaydatastatistics,Timelimitforaprojectdelaydistributionexperience,-30-25-20-15-10-5051015202530%,Frequency(probability),0.2083,Estimatetheconstructionperioddelayeventprobability,Forexample,acompanyplannedconstructionperiodof16months,ifyouwanttoknowtheprobabilityofconstructionperioddelay3months,shallbecalculatedonthefollowingsteps：3/16100%=18.8%Fromthetableorfigurecanchecktheconstructionperioddelayistheprobabilitythat3months5.56%。,Theoryofprobabilitydistributionisusedtodeterminetheprobabilityofriskevents,Casestudy:Qualitydataarevolatile,thefactoriscausedbymanyfactors,andundertheconditionofnormalproduction,theroleofeachfactorisnotobvious,andthereisnoonefactorplaysaleadingrole.SothefluctuationsNormaldistribution.Intheconstructionofwaterconservancyprojects,manyhydrologyworkersstudyofriverfloodpeakflowofrandomness,foundthatmostoftheriversobeyedP-typedistribution.,Theprobabilityofriskeventsusingsubjectiveprobabilityanalysis,Casestudy:Aprojectisdividedintofivestandard,itemizedbid,abidderintendstoparticipateinoneofthebid,assumethatabidderthewinningprobabilityisdifferentforeachtarget,namelynotthebiddingriskisdifferent.Bidderspleasefourexpertofbiddingriskanalysis,evaluationforwhichstandardcanbidmore,lessriskyormissthemark.Expertsontheriskofnotwinningfromsmalltolarge,theresultofthedischargesequenceasshownintabletenderingdecisionmakersfortheexpertstrustdegreebycoefficienta,were0.25,0.28,0.22,0.25,andfourNumbersand1.,Expertstotheevaluationorderofdifferentprojects,：10.25+20.22+30.28+40.25=2.53：10.22+20.28+20.25+40.25=2.28：10.28+10.25+20.25+30.22=1.69：40.22+50.25+50.28+50.25=4.78：30.25+30.25+40.28+50.22=3.72,Differenttargetscannotbethewinningprobabilitycalculation：,Thetotalscore：2.53+2.28+1.69+4.78+3.72=15Cantthewinningprobability,respectively：2.53/15=0.172.28/15=0.151.69/15=0.114.78/15=0.323.72/15=0.25,Conclusion:Formostatriskwhenthefourthstandard,forthethirdwhentheriskisminimal,Analogymethodtocalculatetheriskprobability,hydrologicanalogymethod,Casestudy:TobuildAport,tocomputetheprobabilityofAonce-in-a-centurystormsurge,butlackofMarinedisastersstatistics.AndBislocatedinhasnotfarfromthedetailedhistoricalMarinemeteorologicalstatisticaldata,socantheregistrydataisusedtocalculatetheprobabilityofoccurrenceofstormsurge,andcomparewithAriskofprobability:Pa/Pb=(Ha/La)/(Hb/Lb)Amongthem:H-significantwaveheight;L-wavelength;G-comparedifferencecoefficient.,BayesMethods-基本步骤,确定先验概率-根据历史资料或主观判断，未经实验证实所确定的概率。进行预后验分析，决定是否值得搜集补充资料以及从补充资料可能得到的结果和如何决定最优对策。,BayesMethods-基本步骤,搜集补充资料，取得条件概率，包括历史概率和逻辑概率，对历史概率要加以检验，辨明其是否适合计算后验概率。用概率的乘法定理计算联合概率，用概率的加法定理计算边际概率,用贝叶斯定理计算后验概率。,发生的条件概率公式为：,表示在一个样本空间中的两个事件，给定,和,下，,和,的联合概率公式为：,BayesMethods-贝叶斯定理,中的一个出现是事件,构成互斥和完整的两事件，,和,概率的边际概率公式为：,和,若,发生的必要条件，那么事件,回总目录,回本章目录,BayesMethods-贝叶斯定理,中的一个出现是事件,构成互斥和完整的两个事件，,和,事件的贝叶斯公式为：,和,若,发生的必要条件，那么两个,两个事件的贝叶斯定理为：,回总目录,回本章目录,Ai中的某一个出现是事件B,个事件的贝叶斯公式为：,假定存在一个完整的和互斥的事件,发生的必要条件，那么n,n个事件的贝叶斯定理为：,回总目录,回本章目录,优点：（1）贝叶斯决策能对信息的价值或是否需要采集新的信息做出科学的判断。（2）它能对调查结果的可能性加以数量化的评价，而不是像一般的决策方法那样，对调查结果或者是完全相信，或者是完全不相信。,回总目录,回本章目录,BayesMethods-优缺点,（3）如果说任何调查结果都不可能完全准确，先验知识或主观概率也不是完全可以相信的，那么贝叶斯决策则巧妙地将这两种信息有机地结合起来了。（4）它可以在决策过程中根据具体情况下不断地使用，使决策逐步完善和更加科学。,回总目录,回本章目录,局限性：（1）它需要的数据多，分析计算比较复杂，特别在解决复杂问题时，这个矛盾就更为突出。（2）有些数据必须使用主观概率，有些人不太相信，这也妨碍了贝叶斯决策方法的推广使用。,回总目录,回本章目录,为了提高某产品的质量，企业决策人考虑增加投资来改进生产设备，预计需投资90万元。但从投资效果看，下属部门有两种意见：一是认为改进设备后高质量产品可占90%；二是认为改进设备后高质量产品