An orthogonal-array-based design-of-experiments method for designing a vehicle hood and bumper structure

Abstract

Although the numbers of pedestrian fatalities and injuries are steadily declining worldwide, pedestrian protection is still an important issue. Extensive research has been carried out for pedestrian protection in order to establish appropriate regulations for pedestrian safety. The automobile hoods and bumpers are fairly important for pedestrian protection because pedestrians frequently collide with them during accidents, and they should be designed for the safety of the pedestrians. A new design method for hoods and bumpers is proposed to enhance pedestrian protection by using an orthogonal-array-based design-of-experiments method. Two analysis methods, namely a real experiment and a computer simulation, are utilized to design safe structures of the hood and the bumper. A real experiment is very expensive while computer simulation has modelling imperfections. A design method which uses an experiment and simulation simultaneously is developed. Orthogonal arrays are employed to link the two analyses. The minimum number of experiments is allocated to some rows of an orthogonal array and the simulations are allocated to the rest of the rows to save the cost. Real experiments and computer simulations are conducted for the rows of orthogonal arrays. Design problems are formulated and the orthogonal arrays are directly used to find the design variables by performing the analysis-of-means process in a discrete space. Mathematical error analysis is conducted. Based on the proposed method, a hood and a bumper are designed to protect pedestrians. The results show that the errors are distributed uniformly and a precise design is obtained. © IMechE 2008.