Differential Evolution Algorithm In The Application Of Joint Replenishment

Joint replenishment problem,is refers to the variety of goods from a supplier's purchase plan of the joint procurement strategy.The purpose of joint replenishment is through the development of a reasonable procurement plan,making the number of purchases decreased,the number of discounts,etc.,into the reduction of the total subscription fee.But at the same time because the demand,available funds,transportation conditions and other factors will be affected by external factors such as market environment.And some commodities have perishable nature and so on.Leading to a variety of unpredictable changes.A number of factors are uncertain.It Can not be based on historical statistical data to infer the exact value of the parameters.These uncertain environmental factors should be taken into account when we develop a procurement plan.Thus formed a new problem,that is,the joint replenishment problem under the uncertainty.At the same time,the solution to the joint replenishment model is also a very complex problem.The main content of this paper,the first design of an efficient solution algorithm,Differential algorithm as a new optimization algorithm,superior performance but also easy to "premature" and other shortcomings.In this paper,according to the differential algorithm which the parameters of the performance of the algorithm.The optimization of the control parameters is selected in the algorithm improvement mode.A detailed analysis of how to more effectively set the control parameters,This paper proposes a new adaptive control algorithm.Reduce the sensitivity of the algorithm to the parameters,and avoid the premature convergence of the algorithm.And the performance of the improved algorithm is proved by a function example.On the other hand,the establishment of a close to the reality of the fuzzy environment under the joint replenishment model.And the improved difference algorithm is used to solve the problem of fuzzy replenishment model.Finally,we design an inventory management prototype system.In this paper,the content of the study applied to the system,Through the specific content of practice,to solve practical problems.The validity of the improved algorithm is proved to be valid for the fuzzy replenishment model.