论文信息 - Diagnostic Yield Characterization Expert (DYCE) - A Diagnostic Knowledge Based System Shell for Automated Data Analysis

Diagnostic Yield Characterization Expert (DYCE) - A Diagnostic Knowledge Based System Shell for Automated Data Analysis

Diagnostic Yield Characterization Expert (DYCE) is a knowledge based system shell used for automated data interpretation in semiconductor manufacturing. Using a combination of artificial intelligence techniques, DYCE assists manufacturing and development personnel to diagnose problems by automatically interpreting hundreds of pieces of data, just as a human expert would do. Diagnosis time has been reduced from several hours to minutes. DYCE's inference engine has generic meta-rules to represent the heuristic problem-solving knowledge of the experts; this permits the representation of hundreds of traditional expert system rules With a few dozen meta-rules. This generic shell architecture has important benefits in maintaining an expert system in a dynamic environment without the need for a knowledge engineer or AI training.

Donald D. Pierson | George J. Gallant

[1] Edward H. Shortliffe,et al. Rule Based Expert Systems: The Mycin Experiments of the Stanford Heuristic Programming Project (The Addison-Wesley series in artificial intelligence) , 1984 .

[2] G. Scher,et al. In-line statistical process control and feedback for VLSI integrated circuit manufacturing , 1989, Proceedings. Seventh IEEE/CHMT International Electronic Manufacturing Technology Symposium,.

[3] William F. Punch. Large interactions of compiled and causal reasoning in diagnosis , 1992, IEEE Expert.

[4] S. Saxena. Fault isolation during semiconductor manufacturing using automated discovery from wafer tracking databases , 1993, Proceedings of 9th IEEE Conference on Artificial Intelligence for Applications.

[5] Jay M. Tenenbaum,et al. PIES: An Engineer's Do-It-Yourself Knowledge System for Interpretation of Parametric Test Data , 1986, AI Mag..