gafahp在建设工程评标中的应用研究-application of gafa hp in bid evaluation of construction projects.docx

摘要招投标制度是国际、国内建设工程项目广泛应用的工程承发包模式，是一种富有竞争性的采购方式，是市场经济的重要体现。在招投标活动的多个环节中，评标是建设工程招标采购的中心和关键环节。科学的评标办法有利于市场的有序发展。选择一个科学、合理的评标办法是关系到项目最终能否成功实现的重要问题。传统的评标办法大多主观性较大、定性分析多、定量分析少，很难全面的把握当今多元化发展的市场。因此，本文从完善评标理论出发，对建设工程项目评标办法进行了全面分析，结合智能计算理论，得到一种新的评标决策方法，即改进遗传算法的模糊层次分析法。本文以传统的层次分析法为基础，结合模糊评价理论，对影响建设工程评标的因素进行总结、分类、分层处理，建立建设工程评标指标体系和各指标的相对重要性比较判断矩阵；同时，引入改进的遗传算法，对各评标指标权重进行求解，得到较为合理、有效的评标指标权重。最后，通过实例验证评标体系及评标指标权重的合理性。利用模糊评判原则，得到相对满意的中标单位，为决策者的决策提供参考。关键词：评标；模糊评判；层次分析法；遗传算法AbstractBiddingsystemistheinternationalanddomesticconstructionprojectswidelyused in the mode ofconstruction contracts,is a full competitive procurement, is an important manifestation of the market economy. Bidding activityinanumberofareas,theevalua- tionisthecenterofconstructionbiddingand key links. Scientific evaluation methods conducivetotheorderlydevelopmentofthemarket. Selecting a scientific and rational evaluation methods is related to the ultimate successoftheprojecttoachieveanimport- ant issue.Mostofthetraditionalevaluationmethodsaremoresubjective,morequalitativeanalysisandthequantitativeanalysisissmaller,itisdifficulttograspthefulldiversity oftodaysmarket.Therefore,thisarticleisbaseedthesoundevaluationtheory conductedacomprehensiveanalysistotheevaluationmethodsofconstructionproject tothecombinedwithintelligentcomputingtheory,toobtain,anewevaluationmethod, whichimprovedgeneticalgorithmforfuzzyAnalytic hierarchy process method.Based on the basis of the traditionalAnalytichierarchy process, combining with thefuzzyevaluationtheory,theimpactofconstruction engineeringevaluationfactors weresummarized,classification, layering, processing, establish construction engineer- ing evaluation index systemand the relativeimportanceofeachindexcomparative judgment matrix, Meanwhile, theintroductionofimprovedgenetic algorithmfor solv- ing the bid evaluation indexweight,obtainarationalandeffectiveevaluationindexes weights.Finally,throughexamplesprovetherationalityofthebidevaluationindexes weights. Using fuzzy evaluation princi