3668

3468

كارشناسي ارشد

صنايع

اصفهان: دانشگاه صنعتي اصفهان، دانشكده صنايع و سيستم ها

1386

سيزده، 121، [II] ص.: جدول

ص. ع. به فارسي و انگليسي

قاسم مصلحي، علي حاج شير محمدي

30/5/86

محمد رضا زماني، مهدي بيجاري

1396/02/25

كتابنامه

صنايع و سيستم ها

مهندسي صنايع و سيستم ها

ID3468

به فارسي و انگليسي: قابل رويت در نسخه ديجيتال

ABSTRACTWhen the net present value NPV of a project is used as a measure of its economicalperformance effective management of cash flows over the duration of the project hassignificant impact on improving the project benefits for contractor and client Considering thefact that contractor and client could negotiate for about payment schedule in order tomaximize NPV in this thesis a new model and method is proposed to simultaneouslydetermine location time and amount of progress payment in the multi mode resource constrained project scheduling problem MRCPSP with the aim of agreement betweencontractor and client Generally the proposed method has two phases and each phase has two steps The first step ofphase one start with scheduling of activities in order to minimize makespan but in the firststep of phase two the goal is maximizing contactor s NPV The second step of each phase isthe same This step has two main processes First finding equitable payment schedule and thenrecovering schedules of activity with the aim of improving contractor s NPV These twoprocesses are continually repeated until they increase non dominant result like a local searchnear the schedules of activity in the first step These solutions contain schedules of paymentsand activities Hence contractor and client are able to select one of these non dominant resultsfor the agreement on the basis of their ideals and consider effective factor in decision making In fact the proposed model comprises two main sectors payment scheduling and activityscheduling Two hybrid methods are proposed because of combinatorial nature of thisproblem The first method is for payment scheduling problem name as GASA which is thehybrid of Genetic Algorithms GA and Simulated Annealing SA The other one is the twostages method for MRCPSP which in the first stage modes are fixed and then it will be relax toRCPSP For this stage a hybrid algorithm with help of SA and Tabu search TS proposed Performance testing of These two hybrid algorithm s and parameter tuning for standardproblems with 30 60 and 120 activities have been done and then they have been comparedwith the best available methods resulted from this research and final results approve theperformance of proposed methods The results of computational experiment showed that theproposed algorithm could find appropriate solutions in a sensible time Also moreover allmodel parameters like discount rate proportion of expenses recovered and the number ofpayments have an effect on contractor and client NPV and even more on the number of non dominant solution

قاسم مصلحي، علي حاج شير محمدي

محمد رضا زماني، مهدي بيجاري