Abstract
Contemporary simulation technology can produce accurate assessments of integrated circuit factory (fab) production performance, including the contribution by the automated material handling systems (AMHS). However, the corresponding simulation models are both expensive and time-consuming to construct, and require long execution times to produce statistically valid estimates. These attributes render simulation ineffective as a decision support tool in the early phase of system design, where requirements and configurations are likely to change often. In this paper, we describe an analytical approach to AMHS performance modelling for a simple closed loop AMHS, such as is typical in supporting a 300 mm wafer fab bay. In this system, due to the significant impact of vehicle blocking, a straightforward queueing network model which treats the material handling system as a central server can be very inaccurate. We propose an alternative model that estimates the MHS performance considering the possibility of vehicle-blocking. While the resulting large-scale model presents some computational challenges, it promises reasonably accurate estimates with computation times that are acceptable in a design environment.
Acknowledgments
We gratefully recognize the generous support of the Keck Virtual Factory Lab by the W. M. Keck Foundation. The suggestions of the anonymous reviewers and discussions with Professor Robert Foley have significantly improved the paper.
Notes
† The International Technology Roadmap for Semiconductors (ITRS) periodically lays out a technology plan to guide the semiconductor industry in the coming decades. The latest study is summarized in the ITRS 2003 annual report. It provides the current estimates for research and development that is required over the next decade to meet the historical numbers in performance growth, size reductions, cost, etc.