A response surface approach to front-car optimisation for minimising pedestrian head injury levels

The paper presents a response surface approach to modelling car pedestrian impact events and a multi-objective optimisation technique aimed at finding a front-end car geometry that minimises the injury outcome. The results of the study, involving a parametric front car model and four anthropometric-group pedestrian models, demonstrate the ability of RBF-based response surface models to adequately describe car pedestrian impacts despite the high nonlinearity of such events. Moreover, the response surface models have been successfully used to find front-end car geometry that minimises injury levels across all four anthropometric groups.