Cyclic Scheduling Of Multi-Cluster Tools With Temporal Constraints

Semiconductor enterprises should continue to provide more equipment capacity and reduce production cycle to meet the demands of the highly competitive semiconductor market. Cluster tools are widely used as semiconductor manufacturing equipment. In recent years, throughput analysis and robotic scheduling of single-cluster tools have been well studied, however research work on multi-cluster tools is still at the early stage. This paper mainly studies the periodic problems of multi-cluster tolls for both cases of excluding temporal constraints and considering temporal constraints.In the case of not considering the temporal constraints, a decomposition method is proposed to reduce a multi-cluster tools problem to multiple single-cluster tools problems. We then apply the extended results of throughput analysis and robotic scheduling for each single-cluster tools. For a M-cluster tools, we present a fundamental period (FP) calculation and robotic scheduling algorithm.For the scheduling problem of complicated multi-cluster tools with temporal constraints, an algorithm is proposed based on the time window constraints. First, a MIP model is built for the cluster tool containing the temporal constraints. Then the solution of the MIP model as the inner scheduling algorithm, combined with the decomposition method as the outer layer optimization algorithm schedules the overall multi-cluster tools. An example of multi-cluster tools is used to illustrate the proposed method and indicate that the proposed algorithm is available and effective. Finally, simulation experiments were designed to evaluate the equipment performance. These conclusions have certain guiding significance on the actual manufacturing process of multi-cluster tools.