Optimization of automated material handling systems regarding robustness and delivery time

Abstract

Automated Material Handling Systems (AMHS) are becoming more and more essential for economical working Megafabs. Managing large numbers of Work in Process (WIP) requires advanced logistical support strategies such that lots arrive Just In Time at their destination, to minimize cycle time. This makes robust AMHS design mandatory with short delivery times, at least from the operators perspective. In order to compare system and layout alternatives, appropriate metrics need to get defined, measured and compared. Traditionally, those alternatives are evaluated with discrete event simulations. But it remains cumbersome to determine certain parameters like the overall availability. A new approach will be presented to address this issue. The methods described above have been applied to AMD's Fab 30, and the results are briefly discussed.