Journal article 406 views
A coupled penalty matrix approach and principal component based co-linearity index technique to discover product specific foundry process knowledge from in-process data in order to reduce defects / Rajesh, Ransing; Cinzia, Giannetti
Computers in Industry, Volume: 64, Issue: 5, Pages: 514 - 523
Full text not available from this repository: check for access using links below.
A coupled penalty matrix approach and principal component based co-linearity index technique to discover product specific foundry process knowledge from in-process data in order to reduce defects
|Published in:||Computers in Industry|
Check full text
No Tags, Be the first to tag this record!
Traditional techniques are unable to discover correlations among factors in the ‘noisy’ in-process data. The proposed technique of discovering correlations in the reduced space defined by the principal components is shown to be a novel and robust method. It allows process engineers to view limited number of important penalty matrices from the thousands of possible combinations. The approach has been embedded training courses offered by the American Foundrymen and Institute of Cast Metal Engineers in UK. Elsevier publishers have chosen to make this paper open source, for a period of three months, for the benefit foundry engineers around the world.
College of Engineering