Advanced Simulation Method for Robust Uncertainty Propagation in Complex Engineering Systems

Project: Research

View graph of relations


  • Siu Kui AU (Principal Investigator / Project Coordinator)


Modern engineering has seen a booming demand for analyzing complex engineering systems to unprecedented details paralleled with an increasing reliance on numerical models for performance predictions. Systems are designed with an increasing expectation of high performance reliability and robustness in functionality, which calls for a proper assessment of the effects of uncertainties and their mitigation in the design decision making process. Direct Monte Carlo simulation is a well-established versatile tool for uncertainty analysis, but it is computationally prohibitive for investigating rare failure scenarios, which are often the main interest in reliability and failure analysis. An advanced Monte Carlo method called Subset Simulation has been developed by the PI which is efficient for studying rare scenarios but still retains certain robustness feature of direct Monte Carlo. The method and applications so far have been limited to a single response variable. Recent experience with complex structural and aerospace systems calls for the efficient uncertainty propagation for multiple response variables. An initial study of the multiple variable setting also suggests the critical need to understand and master the trade-off between efficiency and robustness of the simulation algorithm in order for it to be sustainable. This project proposes to develop a generalization of Subset Simulation for multiple response variables and answer the associated questions on efficiency and robustness. The theory shall be applied to studying two classes of complex engineering problems, namely, reliability analysis of buildings structures (static and dynamic) and performance margin estimation of a satellite system.


Project number9041484
Grant typeGRF
Effective start/end date1/01/1027/02/13