Abstract While it has long been realized that surface texture can be used for functional correlation and process diagnostics, it is rarely done so in practice. Surface texture measurements are currently only used as a go/no-go gage for tolerance compliance. Therefore, there is a need to assess current industrial practices and identify directions that will advance the practice of surface metrology to diagnostics and correlation applications. This paper presents a case study based approach to this issue. Several industrial applications are compiled into case studies. Each case study requires the development of novel analytical techniques in order to advance the practice of surface metrology beyond simple parameter computation. After compiling several industrial applications into case studies, a tools requirement matrix is developed. Subsequently, a requirements specification matrix is drawn for an advanced analysis system to enable diagnostics and correlation application to be performed at the floor level.
[1]
Kayvan Najarian,et al.
Surface wavelength content based clustering using neural networks for manufacturing process mapping
,
2003
.
[2]
Simon Haykin,et al.
Neural Networks: A Comprehensive Foundation
,
1998
.
[3]
D. Hosmer,et al.
Applied Logistic Regression
,
1991
.
[4]
Laurene V. Fausett,et al.
Fundamentals Of Neural Networks
,
1994
.
[5]
Jay Raja,et al.
Inference engine for process diagnostics and functional correlation in surface metrology
,
2004
.
[6]
Laurene V. Fausett,et al.
Fundamentals Of Neural Networks
,
1993
.
[7]
B. Muralikrishnan,et al.
Recent advances in separation of roughness, waviness and form
,
2002
.
[8]
David J. Whitehouse,et al.
Handbook of Surface Metrology
,
2023
.