Retrieval Scheduling And Performance Analysis On Miniload Multi-shuttle Automated Storage And Retrieval System

With the advance of warehousing technics and increasing industrial demand,automated storage and retrieval systems(AS/RSs)are applied widely in recent years.Miniload multi-shuttle AS/RSs are used for small parts storage and item picking.The system consists of multi-shuttle storage and retrieval(S/R)machine,racks and picking stations.Bins and totes are the commonly used storage units which are stored and retrieved by the S/R machine.Item picking and replenishment tasks are fulfilled at the picking station.As a part-to-picker system,miniload multi-shuttle AS/RS has the merit of labor-saving,increased picking accuracy and high throughput capability.In this paper,mathematical model and heuristics are designed for retrieval scheduling of the S/R machine.Based on the heuristics,analytical and simulation study on the system performance is conducted.Taking into consideration the characteristic of the retrieving and restoring process,an integer programming model is constructed,optimizing the total travel time of the S/R machine to complete a given set of retrieval requests.Two heuristics,the modified nearest neighbor(NN)heuristic and the Tabu Search(TS)are developed to solve the problem.The performance of the heuristics is demonstrated through numerical experiments.With the increase of the batch size,the computational time to acquire optimal solution grows exponentially.Both NN and TS can improve the quality of the solution remarkably compared with the First Come First Serve rule.Meanwhile,TS can obtain near optimal schedule within short computational time.An analytical model is proposed to predict the travel time of the S/R under NN heuristic.Considering constant and exponential unit picking time,the expected cycle time,throughput as well as the utility of the S/R machine and the picker are analyzed.The analytical model is validated through numerical experiments.The variation on unit picking time,the batch size and the number of shuttles of the S/R are critical factors that influence the system throughput.Numerical experiments show that retrieval sequencing can improve the performance of system significantly when the system bottleneck is the traveling time of the S/R machine.High system utilization can be achieved when the time of the picking cycle and the S/R traveling cycle are of equivalence.With the S/R travel schedule obtained by minimizing the total travel time,relatively large variance on the travel time of each single cycle is observed,which may influence the operational efficiency of the system.The retrieval scheduling model is then further developed with a new objective aims at optimizing the system throughput.Compared with the original travel plan,the schedule obtained by the new model could improve system throughput at the expense of a slight increase on the traveling time of the S/R machine.Considering the dynamic and stochastic characteristic of the demand arrival process,a simulation model is established.Factors including queue length,request waiting time and operation cycles per hour etc.are selected to evaluate the long-run performance and stability of the system.The influence of the sequencing rules,demand arrival rate and the number of shuttles are analyzed through simulation experiment.The retrieval scheduling approaches and the models for performance analysis developed in this paper can be of reference to the design and operational management of multi-shuttle miniload AS/RSs.