Optimization of CNC end milling of brass using hybrid Taguchi method using PCÁ and grey relational analysis

Govindan Puthumana, Anish Nair

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

In order to improve the quality and productivity the present study highlights the optimization of CNC end milling process parameters to provide good surface finish. The surface finish has been identified as one of the main quality attribute and is directly related to the productivity of a machine. In this paper an attempt has been made to optimize theprocess such that the best surface roughness value can be obtained in a process. Hence a multi objective optimization problem has been obtained which can be solved by the hybrid Taguchi method comprising of principal components analysis as well as utility theory. To meet the basic assumption of taguchi method in the present work individual response correlation has been eliminated first by means of Principal components analysis (PCA). Correlated responses have been transformed into uncorrelated quality indices called as principal components. From the principal components the quality loss estimates have been calculated and the Grey relational values are found out for the quality loss estimates. Then the overall grey relational grade has been calculated. Finally taguchi method has been used to solve the optimization problem. Simultaneous optimization of huge number of responses can be done using the aforesaid method.
Original languageEnglish
JournalInternational Journal of Mechanical and Production Engineering Research and Development
Volume3
Issue number1
Pages (from-to)227-240
ISSN2249-6890
Publication statusPublished - 2013
Externally publishedYes

Keywords

  • Optimization of CNC
  • Hybrid Taguchi Method
  • Principal components analysis (PCA)

Fingerprint Dive into the research topics of 'Optimization of CNC end milling of brass using hybrid Taguchi method using PCÁ and grey relational analysis'. Together they form a unique fingerprint.

Cite this