Research On The Scheduling Algorithm Based On Historical Data Of Delivering Takeout Logistics Orders

In recent years,with the rise of Electronic Commerce,the new mode of ordering,online ordering also appears.Because of its unique convenience and intuitive,more and more people are recognized and accepted as an increasingly important way of ordering.In the real world,a restaurant can receive a large number of takeout logistics orders in every day,and use their own delivery vehicles to deliver these orders.The delivery mode of restaurant is as follows: oncethe orders are finished,the delivery person starts off immediately if there are available vehicles,otherwise,the delivery person have to wait until there are available vehicles at restaurant;the orders are often delivered by their placing time or the locations of customers sequentially.However,the delivery mode is absolutely not applicable when the size of customers is big enough,it will not only delay the serving time of some customers,but also greatly increase the driving distance and delivery times of the delivery person,and the operation cost of a restaurant.In order to reduce the operation cost of a restaurant and satisfy customers' demand of service,this paper defines the scheduling problem of delivering takeout logistics orders in a restaurant and studies the mathematical model and solving methods.The main research work of this paper is as follows:(1)Define preconditions,constraints and objectives of the scheduling problem of delivering takeout logistics orders.Using periodic reoptimization mechanism,through solving the static vehicle routing problem corresponding to current state periodically,the scheduling problem of delivering takeout logistics orders can be solved.Establish mathematical model of static vehicle routing problem.(2)Designed and implemented the heuristic algorithm based on historical data and its extended algorithm.Proposed five methods of predicting future customer' number according to historical data.In the heuristic algorithm based on historical data,the input information includes known customers which are known in advance and a set of prospective customers which are predicted.In the extended heuristic algorithm based on historical data,the Multiple Scenario Approach is introduced,the input information includes known customers which are known in advance and several sets of prospective customers which are predicted.(3)The experimental data is obtained by processing the actual data of a restaurant.By comparing and analyzing the several sets of experimental results,it shows that,to solve the scheduling problem of delivering takeout logistics orders,the heuristic algorithm based on historical data and its extended algorithm proposed in this paper are feasible and effective.