A Hybrid Model of Complex Automated Warehouse Systems-Part II: Analysis and Experimental Results

An automated warehouse system has two main components: an automated storage and retrieval subsystem consisting of a number of aisles, each one served by a crane, and a picking area which is formed by bays where stock units coming from the aisles are partially emptied by human operators. These two components are connected via an interface area consisting of carousels, conveyors and buffers. This area is usually modeled as a discrete event system, while the overall system performance depends also on continuous time phenomena. In Part I, a hybrid modeling approach based on a new Petri net formalism and a freeware simulation tool have been presented. The concepts of Hybrid Petri Nets and Colored Petri Nets are merged to obtain modular and compact models for automated warehouse systems. Part II now focuses on the application of this formalism to automated warehouse systems analysis and performance evaluation. Liveness analysis is performed by means of a hybrid automaton obtained from the net model. A deadlock prevention policy is synthesized working on an aggregated model. Finally, a real case study is considered to show the effectiveness of the approach. Note to Practitioners-The sizing and the control of modern automated warehouse systems are carried out by using formulas that do not take into account that they are dynamic systems. This leads to a big discrepancy between the expected performance and the real performance. A model-based approach is needed to obtain accurate results. Warehouse systems have been often considered discrete event systems, while their behavior depends also on continuous time phenomena. In the companion paper, a new formalism has been proposed to obtain a modular and compact hybrid model for these systems. A freeware simulation tool has also been developed. In this paper, it is shown how this model can be used to prevent deadlocks and to evaluate the system performance.