Research On Location Allocation Of Iron And Steel Logistics Park Based On Improved Multi-objective Optimization Algorithm

With the steel industry entering the era of low profit,more steel enterprises have turned to the upsurge of building logistics park to extend supply chain services.Due to the late start and immature management technology,there are many problems in steel logistics park,such as low modernization level,high storage cost and low operation efficiency.Therefore,it is very important to reduce the storage cost and improve the operation efficiency of park.In addition,the application of multi-objective optimization algorithm can get the solution of the above problems,so the research of efficient algorithm for park storage allocation problem is also worthy of attention.This paper focuses on the storage allocation of steel logistics park in the purpose of improving the operation efficiency of the steel logistics park,mainly on the three parts of the multi-objective optimization model of storage allocation,the solving method of the model and the experimental case verification.First,conduct research on the steel logistics park,and build a multi-objective optimization model for the park's storage allocation.Analyze the characteristics of different steel logistics parks and their needs in the steel market,and summarize the key development directions of modern steel logistics parks.And in order to promote the steel logistics park to comply with the development requirements of the steel industry,according to the operation of the yard in the park,with the purpose of improving operational efficiency,taking the shortest travel time of the vehicles in the park and the minimum difference of the work volume of each yard as the optimization objectives,the park's storage allocation is constructed,which can not only meet the basic principles of steel storage in the park,but also avoid overloading of the loading and unloading tools in each yard Multi objective optimization model.Secondly,in order to efficiently solve the multi-objective problem,the improvement strategies of the Fast Non-dominated Sorting Genetic Algorithm with Elitist Strategy(NSGA-II)is studied.By analyzing the shortcomings of the NSGA-II algorithm from the perspective of the population as a whole,two strategies of adaptive hierarchical retention and interval population expansion are constructed,and an improved algorithm(Hierarchical Retention and Interval Population Expansion NSGA-II,HRIPE-NSGA-II)is obtained.The standard test function is used to compare HRIPE-NSGA-II with standard NSGA-II and MOEA/D in terms of population distribution,convergence and comprehensive performance.The results show that HRIPE-NSGA-II is superior to the comparison algorithm in terms of comprehensive index and distribution index,and has good convergence.Finally,based on a case of a steel logistics park,the multi-objective optimization model of the park storage allocation in this paper is verified.By understanding the general situation of the park and collecting and processing the steel data of the park,the effective parameters of the model are determined.And,design an appropriate coding method,use HRIPE-NSGA-II and standard NSGA-II,MOEA/D to solve the above multi-objective model,analyze the experimental results and use k-means clustering method to cluster the optimal solution set to obtain a representative solution.The results of different algorithms in the case show that HRIPE-NSGA-II can obtain a better solution than the comparison algorithm.