Application Of Particle Swarm Optimization In Vehicle Routing Problem

Optimization is an important research target in supply chain management.With recent technological advances,optimization has become complex and more difficult to solve.Swarm intelligence optimization algorithms show good performance in solving optimization problems in terms of application area with high efficiency.Particle swarm optimization(PSO)algorithm is one of the swarm intelligence algorithms.It simulates the process of birds flying and foraging and finds the best solution for the problem.PSO has the advantages of fast computation speed and few parameters.However,PSO is prone to premature convergence,which will inevitably affect its practical application.Therefore,it is of great significance to improve PSO and expend its application.Vehicle Routing Problem(VRP)is a typical NP-hard problem,which can be solved efficiently with swarm intelligence algorithms.This thesis also applies PSO algorithm to VRP and studies the application of PSO.Firstly,the research status of PSO and VRP at home and abroad is introduced.Secondly,the principle and flow chart of PSO are described,and several common improvement directions are summarized.On the basis of these theories,this thesis develops a new PSO algorithm combining Levy flight and reverse learning mechanism.The new algorithm records the personal worst position of each particle.When a particle is stagnated,the reverse learning mechanism is adopted.The personal worst position is used to pull the particle out of local optimum,and the learning step is Levy fraction.The effectiveness and feasibility of the algorithm are proven by simulation and comparative experiments.Then,the improved PSO algorithm is applied to solve two different practical VRPs,which proves that it also has practical application value.Finally,according to the research content of this paper,a preliminary drug distribution system is designed and implemented,and the practical application of the PSO algorithm proposed in this thesis is realized.