Research On Multi-depot And Dynamic Vehicle Routing Problem With Time Windows Under Stochastic Demand

With the rapid development of logistics industry,the issue of urban logistics distribution has received extensive attention.For the multi-depot logistics distribution under the large scale,the cost restraint is more obvious.At the same time,due to the increasing competition in the market,the customer also has more requirements on the arriving time of vehicles.Therefore,it is necessary to consider the time window requirements of the customer in carrying out the arrangement of distribution routes.However,with the increase of vehicles,traffic failures always occur,which will undoubtedly affect the normal running of the vehicle and even lead to arriving late.Moreover,with the rapid development of the electricity supplier industry,customers'information is always instable.It is very difficult to get all information in the path arrangement stage.and new customers' request often happens in the implementation of the path.Therefore,the route designed in the planning stage cannot adapt to changes of information throughout the distribution network.Under the constraints of time windows and distribution costs,the uncertain factors have great influence on the large-scale logistics distribution network.In this paper,the research object is the multi-depot vehicle routing problem with uncertain factors and time windows.First of all,the present situation and problems of city logistics distribution have been analyzed,and the related concepts of the uncertain factors and the time windows have been introduced.And then,this paper considered the minimum total cost of the distribution as the goal,at the same time,took the carrier distribution cost and customer satisfaction into consideration,selecting the uncertain factors under some specific circumstances,and established the mixed integer programming model to solve the multi-depot vehicle routing problem with uncertainty and time windows.Meanwhile,the influence of the new demand point is discussed,and the improved genetic algorithm based on the double layers coding mode and hybrid scheduling strategy is designed.Finally,combined with the practical example,the model is solved.The distribution cost and quality of service are weighed,and the satisfy distribution path achieving the goal of minimizing distribution cost has been found.