A fuzzy logic approach to the selection of the best silicon crystal slicing technology

  • Authors:
  • Doraid Dalalah;Omar Bataineh

  • Affiliations:
  • Industrial Engineering Department, Jordan University of Science and Technology, P.O. Box 620476, Al-hai Al-Janoubi, Irbid 21162, Jordan;Industrial Engineering Department, Jordan University of Science and Technology, P.O. Box 620476, Al-hai Al-Janoubi, Irbid 21162, Jordan

  • Venue:
  • Expert Systems with Applications: An International Journal
  • Year:
  • 2009

Quantified Score

Hi-index 12.06

Visualization

Abstract

Silicon wafer slicing is considered an intricate manufacturing process due to the high level of precision and stability required. The different machining and control factors involved in such technology may cause slicing precision to drift or result in defects in the produced wafers. The modern times impact of more accurate and efficient manufacturing has made it essential to understand and optimize this abrasive slurry cutting process. As a consequence to this understanding, we present a multi-criteria decision making model that assists in selecting the best slicing technology of silicon ingots in integrated circuit chips industry. Fuzzy reasoning is used to model the experts' comprehension and uncertainty in the factors used in the decision criteria. The factors were rated according to each alternative slicing technology using qualitative statements. The criteria factors were divided into two categories, namely, static and dynamic. Different fuzzy rules were summarized from the two categories and then used as an input to the model to calculate the competencies for the different alternatives. The alternative with the highest competency score represents the best choice. In this study, three alternative technologies were considered in the decision analysis, particularly, A-WD-300, DFD600, DFD6000 series. Although they cost more than the others, it was found that the 6000 series technology scores are the highest. The results of the analysis indicate that human-related mistakes and lack of expertise can be one of the main causes that negatively affect silicon wafer production. The proposed fuzzy model significantly contributes to the improvement of manufacturing quality in silicon wafers. Specifically, the model can assist semiconductor manufacturers in solving similar multi-criteria problems by offering a quick objective assessment and systematic model for selection of the optimal performing alternative.