Multi-Objective Evolutionary Algorithm for Path Optimization of Urban Express Vehicles

In view of the actual situation of delivering express delivery in the city. Put forward five objectives: the number of vehicles, total distance, maximum working time, early delivery time, delayed delivery time. Also designed three-stage multi-objective genetic algorithm. A large number of solutions were randomly generated in the first stage of the algorithm, and the approximate Pareto front was quickly found from the solution set with the extreme solution and elite strategy, while the local optimization algorithm is used to optimize the extreme solution. The second stage is divided into multi-objective problems according to the importance of the target, and the path of the unified distribution station is optimized by local optimization algorithm to obtain the relative optimal solution. The third stage corrects the obtained solution with mixed neighborhood, which makes the final solution meet the requirements of express delivery, and solves the occasional local optimal solution problem. Experiments show that this algorithm is superior to the two commonly used algorithms in multi-objective express delivery scenarios.