Collaborative Optimization Of Rider Dispatch And Route Decision On Takeout Delivery Platform

With the development and prosperity of O2O(Online to Offline)e-commerce in China,the transaction scale of takeout market is huge,and the peak quantity of delivery orders is as high as 20 million orders per day.In the face of such a huge market demand,it is very important for the delivery platform to effectively optimize the dispatching of riders.At present,various takeout platforms are engaged in all-round competition in terms of quality supply,delivery experience and software experience.Among them,delivery timeliness and on-time rate are important indicators for customers' satisfaction with orders,and are also one of the core competitiveness of takeout platforms.For the take-out platform,users'delivery timeliness and on-time rate should be improved.At the same time,the delivery platform should also consider the constraint of delivery cost and pursue the minimum delivery cost.Achieving the best balance between distribution experience and distribution cost is the foundation and key to the survival of the distribution platform,as well as the realistic background on which the research of this paper is based.Therefore,based on the comprehensive analysis of the cost factors,a mathematical model aiming at the lowest total delivery cost on the off-market platform is established.In this paper,the Problem of collaborative optimization of Delivery order and route decision is summarized as the Problem of Instant Pickup and Delivery Vehicle Routing with Time Windows(IPDVRPTW).According to the characteristics of the model and problem scale,an adaptive large-neighborhood search algorithm was designed,in which the order allocation optimization strategy of the take-out platform was designed.At the same time,the simulated annealing criterion is used to avoid the algorithm getting into the local optimum.Finally,according to the actual investigation data collation calculation example,the experimental results analysis,at the same time the use of CPLEX algorithm results for optimal verification,the efficiency and accuracy of the algorithm is proved.The research results show that the algorithm in this paper is able to efficiently dispatch orders for large volume orders at the peak period of the outbound sales platform and reasonably coordinate and optimize the routes of riders,thus effectively improving the problems of high cost,low delivery efficiency and low on-time rate of the delivery platform.The relevant methods and conclusions in this paper can provide decision support for real-time distribution scheduling optimization.