The Research On Intelligent Scheduling And Inventory Control Of Automated Storage And Retrieval System

AS/RS,Automated Storage and Retrieval System,is a discrete,stochastic,dynamic,multiple factors,multiple targets complex system.And the intelligent management of AS/RS is a complex optimization problem.The solving process of the traditional method is not only a long time,high cost process,but also difficult to get the optimal solution.In order to further improve the management level and the operation efficiency of the AS/RS,this paper has a research on the optimization of the AS/RS by applying the modern intelligent optimization theory.This paper researches a number of optimization problems of the AS/RS with Genetic Algorithm,and proposes the corresponding specific experimental programs to verify the feasibility of the algorithm.On the intelligent scheduling of the AS/RS,The research contents are as follows,1.According to the principle of the analytic hierarchy process(AHP),this calculates the weight ratios of the combinations of scheduling rules,which relatives to the overall optimization goal.2.According to the different of the task types,the allocation process of tasks is divided into two parts,the outbound task allocation and the store task allocation.To balance the storing amount of materials in each storage unit and balance the utilization rate of storage equipments of each unit,the intelligent optimization system distributes the outbound tasks On the basis of the dynamic storing status of materials in each storage unit,and allocates storage tasks according to the task scale of each storage unit.3.According to the storage allocation rules,the optimization system establishes the mathematical models of the multiple objective optimization.Applying the different multi objective optimization approaches of the genetic algorithm,this paper proposes two experimental programs,the parallel selection program and the weighting coefficient transformation program.By comparing the optimization results of the two programs,it is proved that the weight coefficient method is better than the parallel selection.4.the goal of the task queue optimization mathematical model is gaining the minimum operating time of the stacking machine.This paper proposes the instruction model of mixture integer coding GA optimization method.The experimental results show that,through optimizing the job queue time savings of about 14.05～14.9%,feasibility of the method is tested.On the inventory control of the AS/RS,The research contents are as follows1.This paper researches the characteristics and operation processes of both serial inventory and parallel inventory,and establishes two appropriate mathematical models.Based on the Genetic Algorithm,a new optimization method of the inventory task queue is proposed and verified its reliability.The results of experiments prove that the time-consuming of the optimized inventory queues can be effectively reduced when using the method,inventory costs are declined,efficiency of inventory operations is improved.2.By analyzing the processes of the stacker which carries out the different types of the pallet shift operation,this paper proposes a classification method of the pallet shift operations,which bases on the set operation.At the same time,the process of the pallet shift operation is divided into two parts,the operated library screening process and the task queue optimization.On the operated library screening process,a intelligent optimization mathematical model is established that the optimization target is improving the concentration degree of the material distribution and the constraint is getting the least number of shifts.Comparing with the optimization curve under the constrained and unconstrained conditions,the optimization result of the constrained condition is better than that of the unconstrained condition.About the task queue optimization,the mathematical model gains the minimum task time as the optimization target.Comparing with the optimization curves of the pallet shift queue optimization and the storage task queue optimization,it is proved that the size of the solution space has a direct impact on the convergence rate of the optimization curve.3.According to the four different determined inventory control system model,this paper describes the calculation method of some inventory control parameters,such as the maximum inventory,the safety inventory,the replenishment cycle,and so on.