مدیریت بهره وری (May 2017)
A Comparative Study of Meta-heuristic Algorithms for dynamic vehicle routing problem in order to provide efficiency of transportation systems
Abstract
Vehicle Routing Problem (VRP) wasone of the mostpopularoptimization problems that hadmany usages for productivity andefficiency of transportation systems in recent decades.The VehicleRouting Problem with Simultaneous pick-up and deliveries (VRP/SPD),which considers simultaneous distribution and collection of goodsfrom/to customers (VRP/SDP/SDC) was a variant of the classical vehiclerouting problem where customers require simultaneous pick-up anddelivery at their locations to be completed within a specified time.Applications of the SPD and its related variants are commonly comeacross in every day transportation and optimizing logistic planning. Thispaper had used Meta-heuristic to this end. The proposed method wasapplied for solving capacitated vehicle routing problem (CVRP) toimprove the distribution efficiency and productivity with an objective ofminimizing the total distance covered in each route, while consideringthe capacity of different routes. This problemwas essentially an NP-Hardin nature, so there was no known optimal solution method withpolynomial time. To solve this NP-hard VRP a hybrid genetic basedalgorithm was developed. The proposed geneticalgorithm was tested onsome standardproblem with respect to computational efficiency andsolution quality. The presented method was implemented and itsperformance was further investigated by comparing it against existingheuristics for the same problem. Theresults showed that the success ofthe proposed approach in handling the difficult problem constraints anddevising simple and robust solution mechanisms that can be integratedwith routing optimization tools and used in real world applications.
