Research On Maintenance Support Resource Prediction And Scheduling Optimization Method

Maintenance support resources are the material guarantee to complete equipment maintenance support.Efficient maintenance support activities are the key to ensure the smooth progress of operations and even affect the success or failure of military operations.Troops need to determine the amount of maintenance resources required by the demand point according to the specific maintenance task,and dispatch maintenance resources to all demand points quickly,timely,reasonably and effectively.Combined with the practical problems of maintenance support,the forecasting and scheduling optimization scheme of equipment maintenance support resources are studied.The research contents of this paper is as follows:1)First of all,for the forecast of spare parts demand based on time series,considering that the historical data accumulation of spare parts demand is relatively small,it is difficult for traditional forecasting model to achieve accurate prediction.Because the grey prediction model has good prediction performance for small sample data,the Markov model is combined with the grey prediction model,and the residual error is used to correct,so as to improve the prediction accuracy,and the required spare parts can be predicted more accurately.Secondly,considering the characteristics of maintenance tasks,the planned maintenance tasks,random failure maintenance tasks and battle damage maintenance tasks of vehicles are estimated.According to the basic characteristics of existing maintenance personnel,this paper studies the planning of vehicle equipment maintenance personnel in wartime,and establishes the number prediction model and personnel allocation model.Using the improved Drosophila algorithm to solve the model,the maintenance personnel assignment scheme is obtained.2)Aiming at the problem of fixed maintenance resource scheduling,in order to solve the key problems such as inaccurate prediction and unreasonable allocation in the process of maintenance resource scheduling,a dynamic maintenance resource scheduling optimization model of multiple supply centers and multiple demand points in different operational stages is established,which enables multiple supply centers to optimize the maintenance resource scheduling timely,effectively and quickly,and to meet the corresponding resource requirements of each demand point as much as possible.The co-evolution strategy of three evolution operators(normal distribution crossover operator,global search enhanced differential evolution operator and adaptive mutation operator)is adopted to improve the traditional evolution algorithm.The improved MOEA/D algorithm is used to solve the model.The simulation results show that the combat commander can obtain the optimal maintenance resource scheduling scheme according to the specific conditions of different stages.3)For long-distance maintenance,fixed support points can not meet the characteristics of long-distance maintenance and will cause a waste of maintenance resources,inaccurate resource prediction,high timeliness and high requirements for resource support.According to the above factors,a multi-objective and multi-constraint optimization model with the shortest maintenance resource scheduling optimization time,the least unsatisfied resources and the least mobile support points is established,so that each mobile support points can coordinate and optimize the resource scheduling of demand points.The multi-objective evolutionary algorithm is improved from three aspects: weight vector,population update strategy and evolutionary operator,which effectively improve the efficiency of the algorithm,and eliminates redundant non-inferior solutions by using sparsity method.The simulation results show the scheduling optimization scheme of maintenance resources at each mobility support point.