Vehicle front crash design accounting for uncertainties

Uncertainty is naturally present in the early development phase of vehicle design: some parameters cannot yet be specified exactly, others may have to be changed over the course of development. When this results in the loss of required performance, expensive counter measures have to be taken. In vehicle architectures, this problem is even more severe, because several vehicles may be affected. A new approach for the design of robust vehicle structures is presented here and applied to the USNCAP frontal crash. It relies on a systems engineering technique, the V-model, where design goals for the vehicle are systematically broken down into subsystem and component design goals. Design goals for components are admissible ranges, i.e. solution spaces, for their functional properties within which the final design can be realized. They are uncoupled in the sense that the admissible range for the properties of one component does not depend on the choice of other components' properties. Every design within the solution space delivers the required performance. Susceptibility to uncertainty is reduced by seeking large solution spaces. Solution spaces are computed using a particular algorithm that was developed at BMW. As an example, the design of a front rail for its respective solution space is shown.