Methodology for building a discrete-event model of a temporary storage warehouse in Anylogic

Abstract

The paper describes an approach to constructing a simulation model of a temporary storage warehouse (TSW) within a transport-logistics complex. The model is implemented in the AnyLogic environment using a discrete-event approach and is intended to assess the throughput and operational efficiency of the TSW under various parameters (goods flow, level of automation, number of resources). The architecture of the model and its main blocks are presented, together with details of their interactions. Special attention is paid to the integration of several AnyLogic libraries (process simulation, road and rail transport, AS/RS storage), the use of embedded Java code to control process logic (goods generation, waiting for full train loading, etc.), and to selected simplifications (for example, omission of minor AS/ RS operations).
Practical techniques for configuring the model (use of probability distributions, conditional transitions, conveyor logic) are demonstrated to ensure that the methodology is reproducible by other researchers. The simulation confirmed that even partial automation (AS/RS) substantially improves warehouse performance. The proposed solutions and recommendations can be applied when developing simulation models of similar logistics facilities.