Research On Takeaway Distribution Route Optimization Based On Improved Ant Colony Algorithms In Crowdsourcing Environment

With the rapid development of the Internet economy and the improvement of the pace of people's life,takeaway has become a lifestyle that people are used to.In addition to the quality and taste of the merchant's production,distribution efficiency,as one of the important factors of competitive advantage,affects and restricts customers' choice of merchants as well as the development of O2 O platforms and merchants.In the past,takeaway distribution mainly relied on self-operated distribution of merchants or O2 O platforms,which was usually in the form of merchants or platforms hiring distributors for full-time delivery.At the present stage,the distribution mode of O2 O takeaway has gradually developed from the self-operated logistics distribution of merchants or platforms in bilateral markets to the combination of self-operated logistics distribution and crowdsourcing distribution.How to improve distribution efficiency and customer satisfaction is still crucial.Firstly,this thesis introduced the distribution mode of O2 O takeaway business under the crowdsourcing environment.Secondly,the service characteristics and requirements of crowdsourcing distribution service were analyzed,and in view of empirical dependence and randomness problems of the crowdsourcing distribution,a static open vehicle routing optimization model with a one-sided soft time window and a dynamic open vehicle routing optimization model with a one-sided soft time window were established respectively.Both of them aimed at minimizing the sum of distance cost and time penalty cost.Thirdly,on the basis of analyzing the advantages and disadvantages of each classical heuristic algorithm,ant colony algorithm was selected to solve the routing optimization problem.This thesis improved the two typical problems of slow convergence speed and easy convergence to local optimal solution of ant colony algorithm in the initial search.The improved ant colony algorithm added the impact factor of the number of potential customers moving for the next step in the state transition rule,and combined deterministic search with random search to narrow the search scope of ants.Finally,in order to further verify the feasibility of the improved ant colony algorithm,the longitude and latitude information of each practical example node was obtained by using the Application Programming Interface of Amap.Then the Euclidean distance between the nodes was calculated by using the eclipse 3.4 software based on Java according to the longitude and latitude data.Compared with the standard ant colony algorithm and the standard particle swarm algorithm,the rationality of the improvement ideas is verified,and the solution results show that the improved algorithm effectively improves the solution speed and efficiency.The research of this thesis can help to solve the existing empirical dependence problem and randomness problem in the takeaway distribution business under crowdsourcing environment,provide route planning and route selection support for crowdsourcing platform and distribution personnel,and achieve the purpose of improving distribution efficiency and customer satisfaction.