Application Of Improved Hybrid Particle Swarm Optimization Algorithm In Freight Train Maintenance Shop Scheduling

With the development of information technology,the scale of railway logistics industry is growing day by day,which has a new demand for the quantity and quality of freight trains.It is very important to finish the fast maintenance of the freight car which has reached the service life and continue to put it into use.So it is necessary to dispatch the maintenance workshop reasonably.Workshop scheduling problem is a typical NP-hard(Non-Deterministic Ploynomial-hard)problem.In recent years,many experts and scholars have made rich theoretical achievements in the research of workshop scheduling problem,but different workshops have different problem descriptions and objectives,so the classical model of workshop scheduling problem is difficult to apply to the actual workshop scheduling.After in-depth investigation,analysis and Research on a vehicle depot repair shop,this paper summarizes the repair scheduling model of trucks,and then uses the improved hybrid particle swarm algorithm to solve the model.In this algorithm,aiming at the problem that particles are easy to fall into local optimum in the process of optimization,resulting in low convergence accuracy,a variable named update factor is used as the starting point of improvement,which has a very small probability to update the positions of all particles at random.Considering that the convergence speed may be slower when the variable is increased,the velocity term of the particle is removed so that the particle swarm update formula changes from the original second-order differential equation to the first-order differential equation,which reduces the computational load of the algorithm.In addition,in order to speed up the particle search,a new inertial weight selection strategy is proposed.At the same time,this paper introduces the migratory bird optimization algorithm and the iterative greedy algorithm,combines the improved particle swarm optimization algorithm with the migratory bird optimization algorithm and the iterative greedy algorithm to complement each other in order to complete the optimization task.This paper compares the experimental results with those of other algorithms through simulation experiments,and verifies the effectiveness of the improved hybrid particle swarm algorithm proposed in this paper.The algorithm proposed in this paper is applied to a vehicle depot maintenance scheduling,and the generalized model is solved to obtain the optimal scheduling scheme.Finally,this paper designs a set of dispatching system for a vehicle depot repair shop,which effectively solves the problems existing in the repair scheduling of the vehicle depot,and improves the information management level of the vehicle depot.