This content is not included in your SAE MOBILUS subscription, or you are not logged in.
Reliability-based multidisciplinary design optimization of vehicle structures
Published September 24, 2002 by Institution of Mechanical Engineers in United Kingdom
Crash simulation evaluates vehicle crashworthiness for occupant safety. It is a highly undeterministic process due to uncertainty in every aspect, e.g., material, gages, etc. As a result, conventional deterministic optimization may be insufficient for crash problems. In addition, a system approach, which considers multiple crash modes, must be applied for safety, as a design considering only a single crash event may lead to failures in the other modes. Reliability-based multidisciplinary design optimization is one of the best ways to address uncertainty problem efficiently. It provides an analytic and systematic tool for considering uncertainty in product development process. However, it is often infeasible due to the lack of computational resources, simulation capabilities, and efficient optimization methodologies, as reliability analysis itself is a design optimization problem that requires many function evaluations. This paper investigates two reliability- based optimization methodologies: reliability index approach and performance measure approach. They are applied to a full vehicle design of multiple impact modes including full-frontal impact, roof crash, side impact, 50% frontal offset impact, and other potential safety performance measurements, with uncertainties taken into consideration. The results of these two approaches are compared and discussed in detail. It is concluded that the performance measure approach is more efficient and robust than the conventional reliability index approach.