Multi-performance optimization of bone drilling using Taguchi method based on membership function

Research output: Contribution to journalArticle

36 Scopus citations

Abstract

In orthopaedic surgery, the thermal and mechanical invasion caused to the bone cells due to drilling can result in their damage or even death, leading to the loosening of the fixation and increase in healing time postoperatively. This article presents a hybrid algorithm (Taguchi combined with membership function) for the optimization of bone drilling process with multiple performance characteristics to minimize the drilling induced bone tissue damage. Taguchi method is a powerful optimization tool designed for the optimization of single performance characteristics, but most of the real engineering problems demands for multiple objective optimization. Therefore, a simple modification is incorporated in Taguchi method based on membership function to convert the optimization of the complex multiple performance characteristics into a single signal to noise as the performance index. Bone drilling investigations are carried out on bovine bone with spindle speed and feed rate as drilling parameters. The objective function is the simultaneous minimization of drilling temperature and thrust force. Analysis of variance (ANOVA) is employed to find out the effect of each parameter on the multiple performance signal to noise ratio. The effectiveness of the above approach is validated by the confirmation experiment.

Original languageEnglish (US)
Pages (from-to)9-13
Number of pages5
JournalMeasurement: Journal of the International Measurement Confederation
Volume59
DOIs
StatePublished - 2015

Keywords

  • Analysis of variance (ANOVA)
  • Bone drilling
  • Membership function
  • Taguchi method

ASJC Scopus subject areas

  • Condensed Matter Physics
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Multi-performance optimization of bone drilling using Taguchi method based on membership function'. Together they form a unique fingerprint.

  • Cite this