Research On Multi-objective Material Scheduling Problem Based On Improved Differential Evolution Algorithm

In recent years,human society has been faced with threats from all aspects,with a variety of sudden natural disasters and public health events occurring frequently,which have brought great losses to people’s lives and property.Reasonable material allocation and efficient material dispatch can quickly and efficiently provide relief materials to the disaster sites when disasters occur,reduce casualties and property losses,and prevent and reduce the damage caused by emergencies to the greatest extent,which is one of the important guarantees for safeguarding national security and people’s well-being.However,the emergency scheduling problem is a multi-objective optimization problem,and there are often mutual constraints and limitations among multiple objectives,so it is difficult to find the global optimal solution or converge to the Pareto set.In view of this problem,this paper,aiming at the emergency material scheduling problem,takes the shortest time and the minimum cost as the core objectives,analyzes and studies the problem from the perspective of multi-objective optimization,establishes models so that it can realize fast response and efficient rescue in practical problems,and improves the ability of Differential Evolution algorithm in discrete application problems by improving the algorithm.Specific research contents are as follows:Firstly,starting from the theory of multi-objective problem,this paper introduces the solution methods of multi-objective problem and the related concepts of Pareto front,and expounds the related concepts of emergency material scheduling,analyzes its characteristics and the difference between it and general logistics,so as to extend the three-level dispatch transportation network of emergency material scheduling.This paper also introduces the Differential Evolution algorithm used to solve the problem,and explains in detail the specific steps of the standard Differential Evolution algorithm and the common variations of mutation strategy,which lays a theoretical foundation for the subsequent algorithm improvement.Secondly,in order to further improve the solution efficiency,this paper improves the traditional Differential Evolution algorithm to solve problems such as easy to fall into the local optimal solution and slow convergence rate.A Dynamic Adaptive Weighted Differential Evolution(DAWDE)algorithm is proposed,which is as follows: The algorithm adjusts the scaling factor and crossover factor by adaptive weighting.Then,based on the theory of population aggregation,the weighted combination of two mutation strategies,DE/best/1 and DE/rand/1,is proposed to give consideration to the global optimization ability and local development ability of the algorithm.At last,the individuals judged as precocity are disturbed and accelerate to jump out of the local optima,so as to ensure the population diversity and improve the exploration performance of the algorithm.In the experiment,the algorithm is compared with the standard DE algorithm,SADE algorithm,JDE algorithm and JADE algorithm on 8 Benchmark functions.It is proved that the improved algorithm has better performance,stronger global optimization ability,faster convergence speed and stronger stability.It is concluded that a series of improvements for Differential Evolution algorithm can balance its global optimization ability and local development ability effectively,and improve the robustness and adaptability of the algorithm.Finally,the emergency materials scheduling problem is divided into distribution center location problem and multi-objective materials scheduling problem in this paper.The location operation was carried out with the lowest scheduling cost as the goal,and then the shortest scheduling time was added as another goal for scheduling planning.At last,considering the dynamic change of demand in the rescue process,a dynamic scheduling model with the influence of disturbance quantity is constructed.The improved algorithm proposed in this paper was used to solve the constructed model,and the data recorded in the Wenchuan Earthquake was taken as the simulation object of the emergency materials scheduling.The location planning of the emergency materials distribution center was carried out first,and then the scheduling scheme to achieve the optimal transportation time and transportation cost was solved.Finally,the new demand effect is added to extend the previous scheduling scheme.The experimental simulation results show that,compared with the Genetic Algorithm and the standard Differential Evolution algorithm,the improved Differential Evolution algorithm achieves better performance in solving multiple models under the synergistic effect of multiple objectives.Thus,the feasibility and efficiency of the Dynamic Adaptive Weighted Differential Evolution algorithm proposed in this paper are obtained.