Optimization of an automobile curtain airbag using the design of experiments

Abstract

A side collision of an automobile poses a higher risk of injury than a frontal collision does. Throughout the world, governments and insurance companies tend to establish and implement new safety standards for side collisions in order to ensure the safety of the occupants of vehicles. Most of the standards aim to reduce the head injury criterion. Currently, curtain-airbag systems are widely used and known to be the most effective means to reduce the head injury criterion in the event of a side collision. A curtain-airbag design procedure which uses modern optimization methods based on a newly defined design scenario to reduce the head injury criterion risks of the occupants of vehicles is introduced in this paper. The design scenario has two phases. In phase 1, various elements of the curtain airbag are defined as the design parameters. The variables are determined in a discrete space using an orthogonal array as well as in a continuous space using optimization where each function is approximated. In phase 2, important variables are extracted via a statistical method and those variables are determined in the same manner as in phase 1. As a result, the head injury criterion of the occupant is decreased significantly. The final design is discussed from a practical viewpoint.