Premium
The use of taguchi methods in performance demonstrations
Author(s) -
Bandurek G. R.,
Hughes H. L.,
Crouch D.
Publication year - 1990
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.4680060209
Subject(s) - taguchi methods , reliability engineering , reliability (semiconductor) , product (mathematics) , noise (video) , computer science , software , design of experiments , manufacturing engineering , engineering , industrial engineering , statistics , artificial intelligence , mathematics , machine learning , power (physics) , physics , geometry , quantum mechanics , image (mathematics) , programming language
The use of Taguchi methods for parameter design and tolerance design is well established. This paper describes a new application to qualification trials or type approval tests. The essential difference from the accepted use is that the purpose is to demonstrate performance when a product is subjected to internal or external noise. With a tolerance trial the aim is to find which sources of noise can be cost‐effectively controlled. Two case‐study examples are presented which show the value of the technique. The first shows how the measurements are used to provide one of the published performance specifications for a digital electronic product. The second reveals how an interaction between three environmental parameters took an analogue electronic product out of specification. There is a discussion on the suitability of the technique for software testing and for reliability demonstrations. This includes comments on which arrays are most appropriate for which situation.