An optimal method for bottleneck station short-term Scheduling in wafer fabrication line with nonzero initial state

Abstract

This paper presents initial results of a research in the short-term optimal scheduling on the photolithography station in wafer fabrication, which usually acts as a bottleneck station with multiple machines. It is desirable to minimize tardiness without loss the bottleneck utilization while today's wafer fabs are striving to maximize on delivery to their customers. Computational efficiency is one of the major challenges of applying short-term optimal scheduling of a wafer fab. This research adopts a scheduling policy that optimal scheduling for the heavily loaded bottleneck with an ant colony optimization (ACO) technique and heuristic rules for other lightly-workload stations, taking account of information of both online WIP and new release lots in scheduling horizon, which yields superior results with modest computational effort and enable the practical use of the optimal method.