رويكرد فراابتكاري تركيبي براي حل مسيله‌ي مسيريابي وسايل نقليه با محدوديت ظرفيت و تصميم‌گيري همزمان درمورد خريد يا اجاره‌ي وسايل نقليه

A Hybrid Meta-Heuristic Approach to Solve Capacitated Vehicle Routing Problems using Vehicle Hire or Purchase Decisions

بيشتر مدل هاي بررسي شده در مسيله‌ي مسيريابي وسايل حمل و نقل با هدف كاهش مسافت كل يا تعداد وسايل انجام شده است، حال آن كه در عمل نحوه‌ي تامين وسايل اثر قابل توجهي در هزينه‌ها و همچنين نحوه‌ي طراحي مسيرها دارد. در مقاله ي حاضر با در نظرگرفتن حالت خريد يا اجاره ي وسايل نقليه، مدل اين مسيله‌ توسعده داده شده است، چرا كه تصميم بر اجاره يا خريد وسايل نقليه بر نحوه ي شكل‌گيري مسيرها اثرگذار است. از آنجا كه حل بهينه‌ي اين مسيله فقط در ابعاد كوچك ميسر است، در اين مقاله يك رويكرد حل تركيبي از الگوريتم شبيه‌سازي تبريدي موازي و الگوريتم الكترومغناطيس براي حل مدل توسعه يافته ارايه شده است. در پايان، روش تركيبي پيشنهادي در تعدادي مسايل نمونه آزموده شده كه نتايج حاصله نشان مي-دهد الگوريتم پيشنهادي بهتر از شبيه‌سازي تبريدي عمل كرده و نمونه ها نيز در زمان كوتاه‌‌تري نسبت به نرم افزار GAMS به جواب رسيده‌اند.

The vehicle routing problem (VRP) plays a central role in the optimization of a distribution network. There is numerous research work considering several various assumptions and models of this problem. In most work published in this field, minimizing the total distance traveled by the vehicles or minimizing the total number of vehicles are the most popular goals. But, to the best of our knowledge, the way in which vehicles are procured, and its effect on overall cost and the routing design has not yet been investigated. This has an essential effect on overall costs and the routing design in practice. This is because making an optimal decision between purchasing or hiring a vehicle is dependent on the total distance to be traveled by the vehicle during the planning horizon. In this paper, we consider a capacitated vehicle routing problem (CVRP), in which capacitated vehicles start from a single depot simultaneously and deliver the demanded items of several customers, and where each costumer must be visited once. Each vehicle can be hired or purchased at different costs. Since aptimal vehicle procurement cost is a function of the total distance that the vehicle traveled during the planning horizon, the model is modified in such a way that the decision of purchasing or hiring of each vehicle is made simultaneously. Since some classical instances with a small number of nodes resist the best exact solution methods, most researchers concentrate on metaheuristic algorithms for solving real-life problems. Therefore, to solve the model in real-life dimensions, an electromagnetism algorithm is hybridized with a parallel simulated annealing algorithm and a hybrid solution algorithm is presented. Finally, to evaluate the efficiency of the presented algorithm, some test problems, randomly generated, are solved by the algorithm and the original simulated annealing algorithm. Experimental results show that the hybrid algorithm has far better efficiency than SA.