Stochastic distribution of the required coefficient of friction for level walking - an in-depth study

Abstract

This study investigated the stochastic distribution of the required coefficient of friction (RCOF) which is a critical element for estimating slip probability. Fifty participants walked under four walking conditions. The results of the Kolmogorov–Smirnov two-sample test indicate that 76% of the RCOF data showed a difference in distribution between both feet for the same participant under each walking condition; the data from both feet were kept separate. The results of the Kolmogorov–Smirnov goodness-of-fit test indicate that most of the distribution of the RCOF appears to have a good match with the normal (85.5%), log-normal (84.5%) and Weibull distributions (81.5%). However, approximately 7.75% of the cases did not have a match with any of these distributions. It is reasonable to use the normal distribution for representation of the RCOF distribution due to its simplicity and familiarity, but each foot had a different distribution from the other foot in 76% of cases.

Practitioner Summary: The stochastic distribution of the required coefficient of friction (RCOF) was investigated for use in a statistical model to improve the estimate of slip probability in risk assessment. The results indicate that 85.5% of the distribution of the RCOF appears to have a good match with the normal distribution.

Keywords:

• friction requirement
• friction demand
• statistical distribution
• human locomotion

Acknowledgements

The authors wish to thank Ed Correa, Lauren Gwozdz, Richard Holihan, Jennie Jackson, Mary Lesch, Niall O'Brien, Margaret Rothwell and Peter Teare for their assistance during the course of this study.