Application Of Multi-objective Optimization Algorithm In Urban Logistics Vehicle Routing Problem

Urban logistics serves the needs of urban economic development,and plays an irreplaceable important role in the process of urban development and construction.It also benefits the further development and improvement of urban infrastructure.Urban logistics distribution is a business process in which logistics companies deliver goods to designated locations in accordance with customer requirements,and is a key link in urban logistics activities.In the actual urban logistics distribution process,it is necessary to consider the least number of vehicles,the shortest mileage,and the shortest delivery time.This is a typical vehicle routing problem(VRP).Since VRP is an NPHard problem,the scale of the VRP problem is large,and the calculation amount of traditional accurate algorithms increases sharply as the problem size increases.It is difficult to obtain an effective optimal path using traditional methods.In recent years,with the emergence of a large number of multi-objective optimization algorithms,such NP-Hard problems can obtain feasible solutions with high convergence quality in a short time through the optimization algorithm,which greatly improves the efficiency of the solution.From the perspective of multi-objective optimization problems,this paper summarizes the relevant theories and basic models of vehicle routing optimization problems by consulting the literature,describes the basic structure of the urban logistics distribution system in detail,and analyzes and builds to meet the multi-objective characteristics according to the characteristics of urban logistics distribution.The mathematical model of the urban logistics vehicle routing problem under the premise;however,in the actual urban logistics distribution process,the vehicle routing planning problem is large in scale.For large-scale problems,an algorithm with fast convergence speed and high solution quality is required.In response to this requirement,the problem-solving algorithm is studied,and a large-scale multi-objective optimization algorithm based on interactive identification of decision variables is proposed.By identifying the potential interaction structure of decision variables and forming subcomponents,it minimizes interdependence and reduces adaptation.The number of evaluation times;at the same time,the neighborhood range of each sub-component is divided according to the inter-individual angle in the decision space,so that each subcomponent is assigned a different neighborhood,and the diversity of the solution set is improved.Several examples of large-scale multi-objective optimization problems are used to prove the effectiveness of the algorithm.Finally,the algorithm is applied to the urban logistics vehicle path planning problem for example solution,which verifies the feasibility of the proposed algorithm in solving large-scale multi-objective problems,and provides a decision basis for optimizing urban logistics route planning.