Artificial Bee Colony Algorithm Optimization For AGV Routing Problem With Time Windows

Vehicle routing problem (VPR) is a front and research focus in academics, as a crucial link in modern logistics and distribution processing. Moreover, vehicle routing problem with time windows (VRPTW) is the most representative issue in this field due to its wide range of application and internal complexity. More and More people pay attention to automatic guided vehicle (AGV), as it is the newest and smart method for logistics and distribution without pollution. Now it becomes a trend of future logistics and distribution between carrying workshops. So the study effort of VRPTW in which AGV is transport vechile has the vital significance in both the theoretical research and practical application. This dissertation will study it from following several parts.(1)We propose an artificial bee colony algorithm with tournament selection strategy and a tabu list in the process of abandoning poor sources. TLABC for short. In order to improve the convergence of ABC algorithm, tournament selection in the process of onlookers’ selecting employed bees is put to use. Besides, we take place of the fixed percentage of employed bees in colony to an adjustable parameter, ao as to control the number of elite individuals in the bee colony. Furter more, a tabu list is added to the process of abandoning poor sources in order to improve the convergence and avoid premature convergence, and this new algorithm can avoid scout bees searching the abandoned poor sources again. And then TLABC algorithm is verified by the numerical examples, and it is proven to be suitable for solving multi-modal problem.(2)We propose a new artificial bee colony algorithm to solve vehicle routing problem with time windows, in which the percentage of employed bees is adjusted according to the complexity of the problem automatically.We argue and divide the solution framework into several essential issues, which are the structure of the initial solutions, optimization of vehicle/path number, the improvement of total distance and the post-optimization of best results. And then their concrete implementation methods are given. We develop a TLABC-VT algorithm by combine VRPTW solution framework and TLABC algorithm, and the validity of TLABC-VT to solve VRPTW is proved effective through Solomon’s VRPTW application examples.(3)Based on theoretical study above and the practical problem in B company shop, we design an AGV scheduling platform, this work includes the working principle, the physical schema, the deployment architecture, the functional structure and requirements of software and hardware configuration. Then the implementation methods of task optimization module are presented specificly, which are corresponding to work in first and second part. Then we analyze traffic jams and vehicles collision avoidance in practice, and establish the VRPTW problem model considered traffic jams. Finally we use the platform designed this dissertation to solve the vehicle routing problems in B company and verify the effectiveness of the platform.