A market-oriented wafer map optimization methodology using Differential Evolution to maximize wafer productivity

Abstract

In order to have a competitive edge in increasing global competition, memory industries need to improve productivity by a novel manufacturing strategy which is apposite to the rapid market changes. However, the conventional wafer productivity model only focuses on maximizing gross die which cannot address wafer productivity for profitability, i.e. return on investment (ROI), with respect to current market situations, since ROI is significantly influenced not only by the number of gross dies, but also by the number of shots and the market price. In this paper, we propose a novel productivity model based on ROI in order to compare wafer maps and to determine a chip size for productivity. To search the productivity-maximal wafer map in extremely large search space, we adopt Differential Evolution (DE) as the optimization technique. The computational results show that the proposed method can solve the problem of optimizing wafer map in minutes. Comparison results have demonstrated that the proposed method effectively improved wafer productivity by up to 1.82% in contrast with the old method. Ultimately, the proposed approach helps memory design engineers determine a chip size in an early design stage with consideration of the corresponding productivity.