This paper addresses hierarchical modeling and control issues within a modern semiconductor fabrication facility. It is well known that fabs are appropriately modeled via discrete event systems. Such models, however, are difficult to analyze and even more difficult to use for design purposes. In this paper a new method for modeling the fab is presented. The method leads to models which approximate the fab over small time scales. These models provide synchronous discrete-time approximations which may be useful for analysis and control design. They also provide a natural tool for systematically addressing aggregation/de-aggregation issues. It is also shown how the fab can be approximated by a high-level flow model. Such a model is useful for making high-level long term decisions, determining realistic commands for low-level tracking policies, and for assessing achievable performance. A low-level tracking policy is presented and integrated with a high-level state variable feedback policy. The low-level tracking policy is shown to track low-frequency commands generated by the high-level controller.
[1]
P. R. Kumar,et al.
Re-entrant lines
,
1993,
Queueing Syst. Theory Appl..
[2]
R. David,et al.
A study of the continuous flow model of production lines with unreliable machines and finite buffers
,
1994
.
[3]
Armando A. Rodriguez.
Release policy governor (RPG) for re-entrant semiconductor fabrication lines
,
1997,
Proceedings of the 36th IEEE Conference on Decision and Control.
[4]
Stephen C. Graves,et al.
A Review of Production Scheduling
,
1981,
Oper. Res..
[5]
Stanley B. Gershwin,et al.
Manufacturing Systems Engineering
,
1993
.
[6]
S.C.H. Lu,et al.
Efficient scheduling policies to reduce mean and variance of cycle-time in semiconductor manufacturing plants
,
1994
.
[7]
Qing Zhang,et al.
Hierarchical Decision Making in Stochastic Manufacturing Systems
,
1994
.
[8]
Michael Pinedo,et al.
Scheduling: Theory, Algorithms, and Systems
,
1994
.