This project is concerned with the advancing the fundamental knowledge of life-cycle reliability analysis (LCA) and design optimization with a particular emphasis on engineering microsystems. The goal is to create a rigorous computational framework for studying appropriate formulations of LCA. This framework is a synthesis of (a) reliability analysis and optimization algorithms, (b) life-cycle cost models accounting for the time-variant non-deterministic system response, (c) high-fidelity numerical simulation and sensitivity analysis, and (d) coupled reduced order models. The framework will be embedded into a design methodology that allows the maximization of the life-cycle benefits of microsystems while accounting for uncertainties and reliability constraints. Cost models will be developed that integrate manufacturing and system design requirements into the formulation of the device level optimization problem through cost-tolerance and cost-reliability relations.
The outcome of this work will advance the fundamental discovery and understanding of the life-cycle design of non-deterministic microsystems, in particular of systems whose complexity results from the interaction of multiple phenomena. This research will increase the awareness of the importance of accounting for uncertainties in engineering design and will provide a formal, mathematical approach for incorporating probabilistic life-cycle design concepts. The research activities will be integrated into undergraduate and graduate level courses. An Internet-based version of the computational framework will be developed, which will allow students to discover the importance of uncertainties and the concept of life-cycle reliability in the design process of microsystems.
