简介：TheanalysisandevaluationofanactualportsystemcontainingRdifferentship’skindsandNsubsystemseachofwhichconsistsofM.(i=1,2,…,N)portsarederived.TheexternalarrivalsofshipstothesystemcomefromRdifferentPoissonsourcesandtheinterarrivalstothequeueofeachportofasubsystemaredeterminedaccordingtoconditionprobabilities.EachporthasmultipleberthswhoseservicetimesaredistributedaccordingtoErlangprobabilitydistributionfunctions.EachportofasubsystemhasRinfinitecapacitybuffersforstoringshipsinterarrived.Thismethodprovidesagoodapproximationprocedureforobtainingsystemperfor-mancemeasuressuchaswaitingtimesofships,averagequeuelengths,etc.Optimumportcapacitycanthusbeevaluatedbyusingthisanalysis.Asimulationisalsopresentedtotestthisapproximateanalysis,andagoodagreementisobserved.Thismodelshallbeusedintheplanningofseveralactualportsystems.Throughtheapplicationofthismodel,itispossibletoanalyzeandevaluateperformancesofthesystems.

简介：Fivekindsofconesareintroduced,whichareusedtoestablishtheconstraintsqualifications,underwhichthegeneralizedKuhn-Tuckernecessaryconditionsaredevelopedforaclassofgeneralized(h,φ)-differentiablesingle-objectiveandmultiobjectiveprogrammingproblemsbyusingMotzkin'salternativetheoremandBen-Talgeneralizedalgebraicoperations.

简介：

简介：Thispaperpresentsatrustregionalgorithmwithnullspacetechniquefornonlinearequalityconstrainedoptimization.Consideringinthenullspacemethodsthat,theconvergentrateofrangespacestepisfasterthanthenullspacestepforthemostcases,theproposedalgorithmcomputesnullstepsmoreoftenthanrangespacestep.Moreover,thenewalgorithmisbasedonthereducedHessianSQPmethod.Globalconvergenceoftheproposedalgorithmisproved.Theeffectivenessofthemethodisdemonstratedbysomenumericalexamples.

简介：Usinggeneralizedunivexfunctions,anondifferentiablemultiple-objectiveoptimizationproblemisconsidered.Kuhn-Tuckertypesufficientoptimalityconditionsareobtainedforafeasiblepointtobeanefficientorproperlyefficientsolution.Mond-Weirtypedualityprogrammingisconstructed,theweakandstrongdualitytheoremsareproved.

简介：ThispaperdealswithHeniggloballyefficiencyinvectoroptimizationinvolvinggeneralizedcone-preinvexset-valuedmapping.Somepropertiesofgeneralizedcone-preinvexset-valuedmaparederived.ItalsodisclosetheclosedrelationshipsbetweenHeniggloballyefficiencyofgeneralizedconepreinvexset-valuedoptimizationproblemandHeniggloballyefficiencyofakindofvectorvariationalinequality.

简介：Generally,thetraditionaltroopsoptimizationmodeladoptsthemethodofcertainmathematicprogramming.However,itishardtosatisfytheactualrequestoftheaccuratelyquantitativeanalysiswithalargenumberofstochasticfactorsinmodernhightechniquewar.Inthispaper,thegeneralcharacteristicofthisproblemisanalyzedfully.Anewmethodofestablishingtheoptimizationmodelofjointtacticsorganizationbasedonchance-constrainedprogrammingissettedup.Intheend,anexampleisdemonstratedandsolvedbythegeneticalgorithmbasedonstochasticsimulation.

简介：

简介：这篇论文在计划的一个材料要求(MRP)处理计划的铅时间计算的问题在随机的铅时间下面的环境。目的是最小化成立和backlogging费用的和。建议途径基于库存控制决定变量在哪儿是整数的分离时间。系统的二种类型被考虑：连续生产的多水平和汇编系统。为连续生产系统(在每水平的部件的一种类型)，一个数学模型被建议。然后，这个模型等价于众所周知的分离报童模型，这被证明。这直接在计划的铅时间提供最佳的价值。为多铺平汇编系统，一个奉献模型被建议，变量和目的工作的决定的一些性质被证明。这些性质被用来在客观功能上在决定变量和更低、上面的界限上计算更低、上面的限制。获得的限制和界限打开可能性开发例如，使用一个分支和界限途径的一个有效优化算法。纸与相应证明和几个数字例子详细论述建议模型。建议模型的一些优点和这研究的观点被讨论。

简介：城市的总线网络的多客观的优化能帮助改进运输系统的操作效率并且为在中国减少城市的交通拥挤开发策略。工作使用了累积前景理论，当前为在无常下面的决定的最有影响的模型，优化城市的公共汽车网络。完成研究目的，工作开发了城市的总线网络优化，包括的优化原则，优化目的和限制的理论框架。而且，优化目的能包括地从时间，空间和价值的尺寸反映旅客和总线公司的期望。它在以前的研究独自与仅仅一个股东或尺寸相比更科学、合理。另外，为由到理想的答案TOPSIS的类似的顺序偏爱的技术被用来决定积极、否定的理想的选择。在优化选择和理想的选择之间的关联被灰关系分析同时估计。累积前景理论CPT被用来由比较每种选择的全面前景价值决定最好的选择，精确地在实际生活与期望的用途理论相比描述决策行为。最后，Xian的大小写证明方法能更好调整总线网络，并且优化解决方案是更合理的满足实际需要。

简介：

简介：在有一个cogeneration系统的一个综合熨斗和钢植物，再循环精力连续地被搬运进cogeneration系统和电连续地被产生，并且他们俩不能很长时间被存储。而且，电的产生和消费是不规则的，它可以引起更意外的不平衡。因此，由优化精力利用的操作安排全部精力系统是一个关键问题，它在这二个系统在生产系统，在cogeneration系统的产生电和再循环的精力包括未加工的精力。在这份报纸，为在有一个cogeneration系统的综合的铁和钢植物的精力优化的一个改进线性编程模型被建立。改进模型集中于控制整个精力流动并且通过与一个cogeneration系统在一个综合熨斗和钢植物优化各种精力分发和利用在在生产系统之间的全部精力系统和cogeneration系统安排整个精力消费。改进模型在更高的精力利用，更低的精力费用和更低的污染排出物为最佳的操作条件提供的案例研究表演。

简介：Duetothelongoperationcycleofmaritimetransportationandfrequentfluctuationsofthebunkerfuelprice,therefuelingexpenditureofacharteredshipatdifferenttimeorportsofcallmakesignificantdifference.Fromtheperspectiveofshippingcompany,anoptimalsetofrefuelingschemesforashipfleetoperatingondifferentvoyagecharterroutesisanimportantdecision.Toaddressthisissue,thispaperpresentsanapproachtooptimizetherefuelingschemeandtheshipdeploymentsimultaneouslywithconsideringthetrendoffuelpricefluctuations.Firstly,anARMAmodelisappliedtoforecastatimeserialsofthefuelprices.Thenamixed-integernonlinearprogrammingmodelisproposedtomaximizetotaloperatingprofitoftheshippingcompany.Finally,acasestudyonachartercompanywiththreebulkcarriersandthreevoyagecharterroutesisconducted.Theresultsshowthattheoptimalsolutionsavesthecostof437,900USDcomparedwiththetraditionalrefuelingscheme,andverifytherationalityandvalidityofthemodel.

简介：

简介：公司破产每年在损失花费了十亿美元到银行。因此承认风险预言是一个银行的贷款赞同决定过程的关键部分。为信用风险预言的传统的金融模型为describingtoday在金融健康和一个公司的潜在的破产之间的复杂关系不再是足够的。Inthis工作，一个多重分类器系统(在一个多重聪明的代理人系统嵌入)被建议预言一个公司的金融健康。在我们的模型，每个单个代理人(分类器)基于公司的仅仅偏爱的信息在信用风险的可能性上做预言。每代理人是一位专家，但是限制了知识(由特征代表了)关于公司。所有代理人的决定一起被联合形成最后的信用风险预言。实验证明我们的模型用benchmarkingCompustat美国公司数据集超过另外的存在方法。

简介：Thispaperstudiesafamilyofthelocalconvergenceoftheimprovedsecantmethodsforsolvingthenonlinearequalityconstrainedoptimizationsubjecttoboundsonvariables.TheHessianoftheLagrangianisapproximatedusingtheDFPortheBFGSsecantupdates.Theimprovedsecantmethodsareusedtogenerateasearchdirection.Combiningwithasuitablestepsize,eachiterateswitchestotrialstepofstrictinteriorfeasibility.WhentheHessianisonlypositivedefiniteinanaffinenullsubspace,oneshowsthatthealgorithmsgeneratethesequencesconvergingq-linearlyandtwo-stepq-superlinearly.Furthermore,undersomesuitableassumptions,somesequencesgeneratedbythealgorithmsconvergelocallyone-stepq-superlinearly.Finally,somenumericalresultsarepresentedtoillustratetheeffectivenessoftheproposedalgorithms.

简介：

简介：Inthispaper,wefocusonthereal-timeinteractionsamongmultipleutilitycompaniesandmultipleusersandformulatereal-timepricing(RTP)asatwo-stageoptimizationproblem.Atthefirststage,basedoncostfunction,weproposeacontinuoussupplyfunctionbiddingmechanismtomodeltheutilitycompanies’profitmaximizationproblem,bywhichtheanalyticexpressionofelectricitypriceisfurtherderived.Atthesecondstage,consideringthatindividuallyoptimalsolutionmaynotbesociallyoptimal,weemployconvexoptimizationwithlinearconstraintstomodelthepriceanticipatingusers’dailypayoffmaximum.Substitutetheanalyticexpressionofelectricitypriceobtainedatthefirststageintotheoptimizationproblematthesecondstage.Usingcustomizedproximalpointalgorithm(C-PPA),theoptimizationproblematthesecondstageissolvedandelectricitypriceisobtainedaccordingly.WealsoprovetheexistenceanduniquenessoftheNashequilibriuminthementionedtwostageoptimizationandtheconvergenceofC-PPA.Inaddition,inordertomakethealgorithmmorepractical,astatisticalapproachisusedtoobtainthefunctionofpriceonlythroughonlineinformationexchange,insteadofsolvingitdirectly.TheproposedapproachoffersRTP,powerproductionandloadschedulingformultipleutilitycompaniesandmultipleusersinsmartgrid.Statisticalapproachhelpstoprotectthecompany’sprivacyandavoidtheinterferenceofrandomfactors,andC-PPAhasanadvantageoverLagrangianalgorithmbecausetheformerneednotobtaintheobjectionfunctionofthedualoptimizationproblembysolvinganoptimizationproblemwithparameters.Simulationresultsshowthattheproposedframeworkcansignificantlyreducepeaktimeloadingandefficientlybalancesystemenergydistribution.

简介：