Search for
Login | Username Password Forgot? | Email: | Create Account

Science | Entries: 4656 | Views: 1333 | Modified: 2 years ago | | Add to My Feeds

The primary purpose of this thesis was to measure arm strengths, in combinations of exertion directions, and to evaluate the importance of knowing the precise posture of the arm and specific joint locations in 3D space when predicting female arm strength. A stepwise multiple regression approach was utilized in the prediction of female arm strengths, using kinematic measures of hand location, arm posture and 26-force directions from 17 subjects and 8 hand locations as inputs. When including measures of arm posture, the regression model was indeed improved, explaining 75.4% of the variance, with an RMS error of 9.1 N, compared to an explained variance of 67.3% and an RMS error of 10.5 N without those postural variables. A comparison was also made between the empirical strength data from this thesis and the outputs from the University of Michigan’s Center for Ergonomics 3-Dimensional Static Strength Prediction Program (3DSSPP) software. A poor correlation (R-square = 0.305) and high RMS error (39 N) was found, indicating a definite need for further evaluation of the 3DSSPP package, as it is one of the most commonly used ergonomic tools in industry.

More from DigitalCommons@McMaster

^ Back To Top