Overall fab efficiency [semiconductor manufacturing]

Abstract

The constant increase of the capital needed for a semiconductor facility has brought a huge interest in two methodologies: Theory of Constraints (TOC) and Total Productive Maintenance (TPM) which have been shown to be adequate in optimizing the return on capital equipment. This article shows that both methodologies are convergent. The appropriate use of both together can make it possible to maintain productivity improvement rates in the semiconductor business. The Overall Equipment Efficiency (OEE) measurements are the driver metrics for an effective TPM program. OEE measurements are easy to obtain, nevertheless, the analysis of these parameters requires a large amount of accurate data that are difficult to obtain in a production environment. In addition, it needs the dedication of a considerable amount of effort and human resources to improve in all the organizations (production, maintenance, engineering, etc.). It is widely accepted that bottlenecks should be entered in a TPM program, but for non-bottlenecks, it is not easy to determine which machines should enter the program, their target and their impact on the global fab efficiency. In this study we present a systematic method of approaching this problem.