Computer Experiments for Quality Control by Parameter Design
William J. Welch, Tat-Kwan Yu, Sung Mo Kang et al. · Journal of Quality Technology · 1990
Taguchi's off-line quality control methods for product and process improvement emphasize experiments to design quality “into” products and processes. In Very Large Scale Integrated (VLSI) circuit design, the application of interest here, computer modeling is invariably quicker and cheaper than physical experimentation. Our approach models quality characteristics generated by the computer simulation as functions of both the engineering and noise parameters. This single experimental design for both types of parameters typically requires far fewer runs. The model is used to predict the quality characteristics, from which loss statistics can also be predicted and optimized. In the VLSI applications described, we obtain effective prediction of product performance with comparatively few observations.
Cited by 367 works. doi:10.1080/00224065.1990.11979201
Authors
References
- Off-Line Quality Control, Parameter Design, and the Taguchi Method
- Introduction to Quality Engineering
- Designs for Computer Experiments
- Elimination of process-dependent clock skew in CMOS VLSI
- Response: Off-Line Quality Control, Parameter Design, and the Taguchi Method
- Performance Measures Independent of Adjustment: An Explanation and Extension of Taguchi's Signal-to-Noise Ratios
- Robust Design: A Cost-Effective Method for Improving Manufacturing Processes
- Design Optimization Case Studies
- A Mean Squared Error Criterion for the Design of Experiments
- ACED: Algorithms for the Construction of Experimental Designs