Application of Taguchi's Design of Experiments to the Study of Biomechanical Systems
in Journal of Applied BiomechanicsClick name to view affiliation
Search for other papers by Nenzi Wang in
Current site
Google Scholar
PubMed
Search for other papers by Yen-Hsiu Liu in
Current site
Google Scholar
PubMed
An efficient global optimization procedure is presented by using Taguchi's design of experiments (TDE) as a means for undertaking biomechanical studies that rely on experimentations. The proposed TDE is a systematic method of fractional factorial designs for conducting experiments with many independent variables. The approach can provide a step-by-step means for predicting the results of a comparative full factorial design only with a small number of tests. In this study a three-level, four-variable heel-toe running model, and a two-level, seven-variable bicycle example were examined to show the capability and robustness of TDE. In the process of the analysis, the orthogonal array and signal-to-noise ratio analysis of TDE were used to set up the trial conditions and analyze the results. It is shown that in the heel-toe running analyses the TDE successfully predicted the optimum sets of variables with 89% fewer trials than the full factorial design. The reduction in number of trials in the bicycle example is 94%. As a result, the use of TDE analysis to replace a full factorial analysis can considerably reduce the number of trials and still provide a useful outcome in many multifactor biomechanical studies.
Dept. of Mechanical Engineering
Center for General Education, Chang Gung University, 259 Wen-Hwa 1st Rd., Tao-Yuan 333, Taiwan, R.O.C.
© 2023 Human Kinetics