Iterative capacity allocation and production flow estimation forscheduling semiconductor fabrication

摘要

This paper presents an effective algorithm of determining dailyproduction targets and the corresponding machine capacity allocation forsemiconductor wafer fabrication. The algorithm adopts an iterativescheme and each iteration consists of two modules: the proportionalTarget Generation and Machine Allocation (TG&MA) and the Stage ofPenetration Estimation Algorithm (SOPEA). In TG&MA, machinecapacities are allocated to processing different types of products atvarious stages in proportion to their respective available workloads.With the capacity allocated to each product type, SOPEA then applies arecursive, deterministic queuing analysis to estimate the expectedflow-in workload of a stage within a day. The flow-ins are fed intoTG&MA for another iteration of capacity allocation. Fieldimplementation of this algorithm has demonstrated significant effects onproduction move increase, cycle time reduction and line balancing