A Location-Inventory-Routing Problem Integration Optimization And Research In Uncertain Environment

With the development of economy and society,people pay more and more attention to time,the diversity and randomness of demand are becoming more and more obvious.In order to provide customers with better service,companies must provide differentiated products and services,develop strategies to deal with various uncertainties,and gradually incorporate time-related factors into the enterprise’s reform.For most companies,it will inevitably involve the location,inventory,routing and other factors,which they most concerned about is that how to create new opportunities in the new competitive environment,and to obtain more economic value.In this context,this paper carries on the research to the location inventory routing problem in uncertain environment,and provides the manager with a certain theoretical basis.This paper first analyzes the realistic background and research status of location-inventory-routing integration optimization problem.Since there are not many researches on this kind of problem at home and abroad,this paper reviewed the status quo separately from four aspects: location-inventory,location-routing,inventory-routing,location-inventory-routing.Because many of the current research on this issue are only considered the cost factors,and customer needs are determined,so for the study of this problem will provide managers with better decision-making,and thus create more economic benefits.After introduced the theoretical knowledge needed in the topic research,taking into account the factors that most managers care about in reality,established a mathematical programming model of cost as the general objective,which mainly includes the cost of location,the cost of inventory and the cost of distribution.Where the customer’s demand is random and takes into account the customer’s soft-time window factors.Then,in order to be closer to the reality,then take the time as a objective,established a two-objective mathematical programming model with cost and time,by using the Pareto theory,we obtained the Pareto solution set.Aiming at the mathematical model and the nature of the problem in this paper,the improved genetic algorithm and the decentralized search algorithm are designed to solve the single objective model considering the cost and the multi-objective model considering the cost and time respectively,It is found that the proposed genetic algorithm and the decentralized search algorithm can obtain a satisfactory solution,and the decentralized search algorithm has a better relation to the genetic algorithm than the genetic algorithm.Finally,the paper summarized the research contents,and points out the factors that have not yet been considered in the study and the shortcomings that still exist,and puts forward the problems that need to be improved and further studied so as to be improved in subsequent studies.