A novel approach for multi-variable analysis in ergonomic design

Little research exists on predicting subjective comfort as it relates to the interior design of a vehicle. Moreover, current design approaches have not provided a complete solution to this issue. The shortcomings of current design approach have motivated this research. In this research we used a new comfort measurement method and propose a new approach for designing the driving compartment configuration of automobiles. We develop methods which integrate experimental data relating human and design parameters to subjective comfort. We propose integrated statistical and neural network methods to develop predictive models of "driving" comfort. We then propose to use these analytic methods to construct tools for computer aided ergonomic design (CAED). Our research and modeling efforts lead to the conclusion that our new modeling approach can help to develop ergonomic design models with better predictive power. Most of all, the new CAED tool can help designers find the optimal comfort design for vehicle occupants and avoid costly testing and modification process.