Response policy for failure of parallel module in single-arm time constrained cluster tools

Abstract

Processing modules in cluster tools are prone to failure and thus it is important to have effective failure response policy. This problem is addressed in this paper for single-arm cluster tools with parallel processing module flow pattern and strict wafer residency time constraint. The system is modeled by a Petri net model that can model not only the steady state behavior of the system but also the transient behavior. Based on the model, schedulability is analyzed when one of the parallel modules for an operation fails. If it is schedulable, effective control policies are presented to keep the system in operation such that no wafer residency time constraint is violated. Examples are presented to show the application of the proposed control policies.