Abstract
Injury risk models can play a key role in ergonomic worksite analysis directed at preventing low back disorders. Such models can be used to classify lifting tasks as having the same characteristics as those tasks which have had a high (or low) incidence rate of back injuries. Two evolutionary computation techniques (genetic algorithms GA, and genetic programming GP) were used to construct low back injury risk models. A GA model, GP model, logistic regression model, and an artificial neural network were constructed and tested using 235 documented lifting task cases. Results indicated that the evolutionary approaches were superior to the other models in terms of classification performance and parsimony.
Get full access to this article
View all access options for this article.
