Growing demand for fuel-efficient or light weight vehicle has become a challenge for vehicle development. Upfront engineering process provides more opportunities for engineers to improve body weight efficiency. To accelerate the upfront body development process, the parametric concept modeling technology is commonly employed to generate parametric three-dimensional geometry, joints, modular components, concept welding, and finite element meshes. The topology optimization which determines the best structural layout without weight penalty has also been used during the conceptual design stage. The objective of this research is to explore the feasibility of integrating the advanced parametric concept modeling and both topology optimization and structural optimization technologies into upfront body architecture development process. The topology optimization identifies the critical body components and the parametric concept modeling methodology is used to parameterize body geometry and section properties. A process integration design optimization (PIDO) tool is then used to facilitate the whole process from the tasks of design of experiment sampling, job management control, response surface generation and optimization. A vehicle body example is used as an example to demonstrate the feasibility of the proposed design process.

This paper focuses on the methodology development and application of reliability-based design optimization to a vehicle exhaust system under noise, vibration and harshness constraints with uncertainties. Reliability-based design optimization provides a systematic way for considering uncertainties in product development process. As traditional reliability analysis itself is a design optimization problem that requires many function evaluations, it often requires tremendous computational resources and efficient optimization methodologies. Multiple functional response constraints and large number of design variables add further complexity to the problem. This paper investigates an integrated approach by taking advantages of variable screening, design of experiments, response surface model, and reliability-based design optimization for problems with functional responses. A typical vehicle exhaust system is used as an example to demonstrate the methodology.