13409

12213

كارشناسي ارشد

مهندسي صنايع

اصفهان: دانشگاه صنعتي اصفهان، دانشكده صنايع و سيستم ها

۱۳۹۶

دوازده، ۱۱۱ص.: مصور، جدول، نمودار

قاسم مصلحي، محمد رئيسي نافچي

مهدي بيجاري، حميد مير محمدي

1397/01/15

كتابنامه

صنايع و سيستم ها

مهندسي صنايع و سيستم ها

ID12213

112 Two agent order acceptance and scheduling in a single machine environment with tardiness penalties Mahsa Jenabi m jenabi@in iut ac ir Ghasem Moslehi supervisor Moslehi@cc iut ac ir Mohammad Reisi Nafchi supervisor reisi m@cc iut ac ir Department of Industrial and Systems Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Language Persian Degree M Sc Date 2018 01 22AbstractThe problems related to the order acceptance and scheduling is taken into consideration recently Paying attention to the order acceptance and sequencing of orders creates abalance between revenue and the cost of the existing orders on the sequence because themanufacturer can deliver the orders with the highest profit in its due date by rejecting someof the orders and accepting the rest of them Problems related to concurrent decisions onthe acceptance or rejection of an order and sequencing of them are known as orderacceptance and sequencing problems In this research we investigate the order acceptanceand sequencing problem with maximum profit considering as the target function with totalweight of tardiness as the cost function A time index model and a heuristic algorithmsbased on the relaxation of the time index model named as SODP are proposed in order tosolve this problem the proposed SODP algorithm consists of two dynamic programmingalgorithms to find tight upper bound and a heuristic algorithm to find a tight lower bound The upper bound improves by subgradiant optimization approach Also there are somedominance rules for this problem Computational results indicated the time index modelcannot solve all the problems within 20 to 100 orders but the proposed SODP algorithm isable to achieve near optimal solutions for the instances up to 200 order in size so theaverage deviation from the problem upper bound is 0 23 The two agent orderacceptance and scheduling problem with the aim of profit maximization is the secondproblem which is investigated in this research Cost function of the first and second agentconsidered as total weighted tardiness and the number of tardy jobs respectively Weproved that this problem can be converted to another problem Furthermore we proposedsome dominance rules for solving converted problem A heuristic algorithm named asTSODP is presented to solve the converted problem The results showed the time indexmodel is not able to solve the problems up to 150 orders but by using the proposedheuristic algorithm we can solve the problems up to 200 orders in size so the averagedeviation from the problem upper bound is near 0 4 in all the categories Keywords dynamic programing order acceptance Scheduling two agent weighted tardiness

قاسم مصلحي، محمد رئيسي نافچي

مهدي بيجاري، حميد مير محمدي