17-10-2012, 04:38 PM
Modeling of an AS/RS Serviced by Rail-Guided Vehicles with Colored Petri Nets: a Control Perspective
RS Serviced by Rail-Guided Vehicles.pdf (Size: 363.48 KB / Downloads: 67)
Abstract
An Automated Storage and Retrieval System (AS/RS) automatically stores incoming material and retrieves stored parts with no direct human handling. This paper proposes a modular and unified modeling framework for heterogeneous automated storage and retrieval systems, comprising rail guided vehicles and narrow aisle cranes. We employ colored timed Petri nets, representing a concise and computationally efficient tool for modeling the system dynamic behavior, particularly suitable for real time control implementation. Indeed, the model can be utilized in a discrete event simulation to apply control policies in order to solve scheduling problems, as well as to avoid deadlock and collision occurrences.
INTRODUCTION
Automated storage and retrieval systems (AS/RSs) are widely used in warehousing and manufacturing systems for storing and retrieving finished products and parts. An AS/RS is a combination of automatic handling, storing/retrieval equipment and control systems, characterized by high accuracy and speed. Typically, an AS/RS consists of several aisles with storage racks on either side, each serviced by a crane (or, equivalently, S/R machine), operating storage and retrieval of the parts. The cranes move in three directions: along the aisle to perform transfers, sideways between the aisle and the racks at the sides, and vertically to reach the storage/retrieval location. Each aisle is equipped with a storage and a retrieval conveyor. Moreover, the AS/RS aisles are serviced by Rail Guided Vehicles (RGVs), unloading the parts to be stored or loading them after retrieval. Finally, a main input and an output buffer station, where the RGVs load or unload pallets, are present [6].
The benefits of AS/RSs include low labor cost, low inventory cost, enhanced space exploitation, improved material tracking and high system throughput. Nevertheless, advantageous operation of an AS/RS clearly depends on the control policies implemented. In the control architecture of an AS/RS three hierarchical levels may be identified [8]: the strategic level, containing control policies for long term expected system performance, the tactical level, collecting control rules for short term performance, and the operational level, dealing with real time behavior. This paper focuses on the operational control problems. In particular, the proposed management strategy is organized in two subsequent phases. The first and higher one is the scheduler level, taking decisions on selecting a suitable batching policy, a proper storage policy, a suitable position of the cranes when idle (dwell point selection problem) and a proper retrieval and RGVs routing policy. The second level is the real time controller, that is in charge of taking decisions on resource allocations in order to avoid conflicts and deadlocks. Numerous studies in the field appear in recent literature; in particular, investigations deal with typical operational problems, such as defining proper storage and retrieval sequencing policies, in order to maximize the system throughput [1, 6, 7]. However, authors do not examine the real time controller issues, including the RGVs and cranes management.