Modeling And Optimization Of Tier-to-tier Shuttle-based Storage And Retrieval System

The traditional logistics warehousing and picking system is mainly based on manual operation.In recent years,with the rising of New Retail concepts and the booming of e-commerce business,the structure of customers' online orders has been gradually exhibiting the feature of "small-batch,multi-frequency".It is time-consuming and labor-intensive for manual picking.Therefore,the automated picking systems based on "parts-to-pickers" pattern emerged timely.The shuttle-based storage and retrieval system(SBS/RS)presented in this paper has attracted the attention of researchers due to its high flexibility and robustness.Depending on whether shuttles can perform cross-tier tasks with the help of a lift,two configurations can be defined:SBS/RS with tier-to-tier shuttles and that one with tier-captive shuttles.In This paper,the tier-to-tier SBS/RS has been investigated thoroughly.At the very beginning of the designation of automated picking system,most designers use commercial software to simulate the system to find the optimal equipment configuration.However,there are formidable challenges in such simulation method,including the excessiveness of parameters,the arduous abstraction process and the vast time investment.In addition,how to improve the efficiency of the system by reducing the waiting time between shuttles and lifts through optimizing the task scheduling is also an unsettled problem.Therefore,this paper studies operation process of the system and analyzes system performance through establishing a semi-open queuing network model,based on which the optimal number of shuttles for the system has been decided.Then an accurate task scheduling model has been established to find the optimal order sequence within a time window to ensure that the total time for completing orders within a time window is the shortest.Firstly,the movement of shuttles and lifts for a single order picking task are divided into six parts through analyzing the retrieval process of the tier-to-tier SBS/RS.An order cycle time model is built with consideration of the time that tasks waiting for shuttles and shuttles waiting for lifts.The factors which affect operation efficiency of the SBS/RS are determined and described in detail,through analyzing the operation time.Then,according to the internal working flow and operation time of the SBS/RS,a semi-open queuing network model of the system is established,in which shuttles are considered as the server of order tasks and lifts are considered as the server of shuttles.Then the Approximate Mean Value Analysis(AMVA)Algorithm and the Birth and Death process are used to analyze the order cycle time of the system and utilization of shuttles,as well as lifts.Then the accuracy of the model is verified by simulation experiments on MATLAB,and the optimal number of shuttles under various scenarios is further determined considering the optimal order cycle time of the system based on the model.Finally,in order to reduce the idle time of lifts and the waiting time of shuttles,some constraints are imposed on the order tasks and devices,and a task scheduling optimization model of the system is established with the shortest picking time of the total orders within a time window as the objective function.The GUROBI linear programming solver is used to solve the model to obtain the optimal order sequence and the shortest operation time,based on which the operation efficiency of the system is improved.