An integrated quality engineering and evolutionary neural network procedure for product design

Ming Chyuan Lin, Meng-Dar Shieh, Shuo-Fang Liu, Yun Yun Wu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In product design, the accuracy of product information greatly affects design quality. Therefore, robust product design provides a critical role that sound product design plays in securing competitive advantages in product quality and production efficiency. In the area of robust product design, the Taguchi method of quality engineering simplifies the analysis method and provides an effective product design approach by confirming variable characteristics and determining the optimum combination of characteristics. The aim of this research is to introduce an evolutionary neural network into robust product design to help designers search for a more optimal combination of variable characteristic values for a given product design problem. In the product design procedure, the data resulting from the experimental design in the Taguchi method are forwarded to the back-propagation network training process and simulation to predict the most suitable combination of variable characteristic values. The recommended combination of variable characteristic values is represented in 3D form using a computer-assisted design system. A case study of design of a lat bar for pull-down fitness station is used to demonstrate the applicability of the design procedure. Note that the signal-to-noise ratios of the robust lat bar product design are derived from experiments that measure the back and bicipital muscle responses using an electromyography (EMG) apparatus. The results indicated that the proposed procedure could enhance the efficiency of product design efforts.

Original languageEnglish
Title of host publicationTransdisciplinary Engineering Methods for Social Innovation of Industry 4.0
Subtitle of host publicationProceedings of the 25th ISPE Inc. International Conference on Transdisciplinary Engineering
EditorsNel Wognum, Josip Stjepandic, Marcello Pellicciari, Cees Bil, Margherita Peruzzini
PublisherIOS Press BV
Pages441-450
Number of pages10
ISBN (Electronic)9781614994398
DOIs
Publication statusPublished - 2018 Jan 1
Event25th ISPE International Conference on Transdisciplinary Engineering 2018 - Modena, Italy
Duration: 2018 Jul 32018 Jul 6

Publication series

NameAdvances in Transdisciplinary Engineering
Volume7

Other

Other25th ISPE International Conference on Transdisciplinary Engineering 2018
CountryItaly
CityModena
Period18-07-0318-07-06

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Industrial and Manufacturing Engineering
  • Software
  • Algebra and Number Theory
  • Strategy and Management

Fingerprint Dive into the research topics of 'An integrated quality engineering and evolutionary neural network procedure for product design'. Together they form a unique fingerprint.

Cite this