Research On Intelligent Optimization Algorithm For The Vehicle Routing Problem

With the development of artificial intelligence,the problem of logistics distribution and vehicle routing has a broad application prospect in large shopping malls,express delivery and other fields.Therefore,the in-depth study of vehicle routing problem has high scientific significance and engineering application value.It is found from the existing research that the intelligent optimization algorithm for vehicle routing problem has the following shortcomings: First,the optimization algorithm fails to make full use of local search information to guide the algorithm’s solving process,thus affecting the algorithm’s solving efficiency;Second,the similarity features of the problems fail to effectively achieve the synergy effect among the optimization problems through the knowledge transfer mechanism;Third,the effectiveness of intelligent optimization algorithm to solve vehicle routing problems in real life has not been fully verified.In view of the above deficiencies,the specific research work of this paper has the following three aspects:First,in order to effectively use local search information,a multiobjective memetic algorithm is proposed to solve the vehicle routing problem with time window.In this algorithm,on the one hand,using the multi-directional local search strategy,the multi-directional search is performed according to the different local searches under the specific problem knowledge.On the other hand,by using the enhanced local search chain technique,the potential solutions obtained in the process of evolution are adaptively selected for subsequent local search operators.In this way,the multiobjective memetic algorithm can not only effectively explore solution space from multiple directions,but also make full use of potential solutions in a chain-based way.Second,in order to effectively transfer the similarity characteristics of the problem,an evolutionary multitasking algorithm is proposed to solve the vehicle routing problem with routing balance,In this algorithm,on the one hand,the multitasking optimization strategy is used to realize the knowledge transfer between tasks alongthe search process.On the other hand,the multiobjective optimization strategy is used to optimize the non-dominant solutions of the population.Thus,the evolutionary multitasking algorithm can not only enhance the performance of the algorithm by information transfer between similar tasks,but also maintain the diversity of the population in the optimization process.Third,In order to solve the actual vehicle routing problem,an intelligent order allocation algorithm for booking trips online between cities is proposed.The algorithm uses the time and space information of the order to construct the initial order allocation scheme set,then uses local search to optimize the order allocation scheme,and dynamically allocates the new orders according to the appointment time,and finally provides different order allocation schemes for decision makers.By constructing the order allocation scheme and optimizing the order allocation scheme,the algorithm can obtain high-quality solutions in a short time.To sum up,this paper explores and studies the solution method and application scenario of vehicle routing problem from three different levels,from the generation mechanism of solution,to the search framework of algorithm,to the practical engineering application.At the same time,through a lot of experimental evaluation,it is verified that the proposed algorithm has good optimization performance,which provides an effective reference for scientific research and engineering fields.