Data-based node penalties in a path-finding algorithm in an automated material handling system

Abstract

Increasing factory throughput is a critical issue in the semiconductor industry, and a quick transition of material to the next location in the automation system plays a significant role in increasing throughput. A dynamic pathfinding algorithm for a vehicle-based automated material handling system (AMHS) is discussed in this paper. The dynamic path-finding algorithm uses distance between nodes, node penalties, and the number of vehicles queued to calculate the total cost of a path. This paper introduces the use of historical data from the AMHS and discusses how to effectively utilize such data in critical situations to improve overall AMHS performance.

Congestion-aware dynamic routing in automated material handling systems

“…Han et al [19] compare single and double level tracks using a 22 bay simulation and conclude that the double level track provides a significant improvement in the number of moves per hour. Fukunari et al and Bahri and Gaskins [15,5] use AutoMod to compare node-based methods for managing congestion on several layouts. Fukunari et al conclude that incorporating historical information in setting node penalties improves performance.…”

“…The need for dynamic routing that considers congestion in unified AMHSs has been expressed, including the existence of patents [17,22], but limited literature addresses specific methods or demonstrates improvement via high-fidelity simulation. Fukunari et al and Bahri and Gaskins demonstrate that considering and/or balancing congestion in routing decisions reduces delivery times [15,5]. With respect to layout or flow path design, we are not aware of any literature that considers layout effectiveness under dynamic routing in AMHS or AGV systems, although [65] stresses the importance of considering the interaction of operational and design decisions.…”

Levels of Capacity and Material Handling System Modeling for Factory Integration Decision Making in Semiconductor Wafer Fabs

“…More specifically, these models are used in the design of AMHS layout (for example, see [8], [45], [46], and [47]), the analysis of the hallway transportation system (for example, see [7], [44], [48], and [49]), determination of vehicle and storage capacities (for example, see [50] and [51]), analysis of transportation technology (for example, see [52]), evaluation of vehicle control algorithms (for example, see [53]- [55]), and the analysis of AMHS delivery paradigms (for example, see [56]- [58]), among others.…”

On-line reconfiguration to enhance the routing flexibility of complex automated material handling operations

“…There have also been other approaches where the routings are generated dynamically in a single phase. For example, Montemanni [11] used an ''ant-colony'' algorithm to solve a dynamic vehicle routing problem, Fukunari [12] used a data-based node penalty approach for a dynamic path-finding algorithm for an AMHS in a wafer fab, and Haghani [13] used a genetic algorithm to solve a dynamic vehicle routing problem with time-dependent travel times. However, the two-phase approach is usually faster and preferred for time-critical applications.…”