Preventive maintenance (PM) is a vital activity in semiconductor manufacturing. A good PM schedule can increase the availability of tools by trading off between the planned unproductive down time versus the risk of much costlier unscheduled down time due to tool failures. Cluster tools are highly integrated systems made up of several processing modules (chambers) mechanically linked together, which can perform a sequence of semiconductor manufacturing processes. We present a two-layer hierarchical modeling framework for addressing the PM optimization problem for cluster tools, i.e., a Markov decision process (MDP) model at the higher level, and a mixed linear programming (LP) model at the lower level. Production planning data such as WIP levels are incorporated in these models. The LP model is solved using optimization packages EasyModeler and OSL. A case study comparing the results with a reference schedule is conducted in the simulation environment AutoSched AP, and numerical results are reported.
[1]
M.C. Fu,et al.
A Markov decision process model for capacity expansion and allocation
,
1999,
Proceedings of the 38th IEEE Conference on Decision and Control (Cat. No.99CH36304).
[2]
Richard M. Feldman,et al.
A survey of preventive maintenance models for stochastically deteriorating single-unit systems
,
1989
.
[3]
J. McCall.
Maintenance Policies for Stochastically Failing Equipment: A Survey
,
1965
.
[4]
William P. Pierskalla,et al.
A survey of maintenance models: The control and surveillance of deteriorating systems
,
1976
.
[5]
Dimitri P. Bertsekas,et al.
Dynamic Programming and Optimal Control, Two Volume Set
,
1995
.