Yield model for fault clusters within integrated circuits

  • Authors:
  • Charles H. Stapper

  • Affiliations:
  • IBM General Technology Division, Burlington facility, Essex Junction, Vermont

  • Venue:
  • IBM Journal of Research and Development
  • Year:
  • 1984

Quantified Score

Hi-index 0.01

Visualization

Abstract

Generalized negative binomial statistics turns out to be a model of the fault distribution in very large chips or wafers with internal defect clusters. This is expected to influence large chip and full wafer redundancy requirements. Furthermore, the yield appears to be affected by an experimental dependence of the average number of faults on chip area.