Study On The Location Routing Problem With Time Windows Based On Tabu Search Algorithm

In recent years, logistics developed very fast in our country, specialization level of logistics improved greatly, and how to reduce logistocs cost become an important problem facing by companys. With the development of mornden life rhythm and in order to improve the production efficiency of company, logistics service time of every company become more strict, so, time become more and more important, especially logistics syestem using latest logistics mode, lean logistics and just in time theory. In the tradition logistics decision-making process, the location of logistics facilities and the routing of transportation vehicle received researchers' abroad attention, because of the integration of logistics activity, only facility location and transportation route consider simultaneously the cost of logistics can reduced effectively, base on this consideration, this paper study the combind problem of the above problem-location routing problem, and we add the time windows in it to make it close to the relastics. So, it is of theoretical and practical senses to some extent to study the location routing problem with time windows.The object of this paper is to research the multiple-depot location routing problem with time windows. It introduces basic knowledge of logistics, develop the meaning, classification and model of location allocation problems, vehicle routing problems and location-routing problems first. And then, based on location routing problem, this paper founds the math model of the location routing problem with time-windows by adding the time constraint on it, and checks the correctness of this math model in small scale problem using Lingo software.This paper designs tabu search algorithm to solve LRPTW(Location Routing Problem with Time Windows) for large scale problem. The characteristic of tabu search algorithm is forbid the work has been done before. In order to evite the background of local neighbors filed search immerse into the local optimization. Tabu search algorithm use a tabu list to memorize the local optimization points and the process of it, when search the point next time, it make use of the information stored in tabu list, and no more or choose these point to search, in order to jump from the local optimization. After compare the result of our method and Lingo software and other domestic researchers, we found that the designed tabu search algorithm is faster and more accurate than other methods. For the different scale date problem, we constructed 25points, 50points, 100points and 150points for the LRPTW from the international standard date of VRP. After test and analysis, we found the designed tabu search algorithm is stable, speed is quickly, it proves validaty and feasibility of designed algorithm.