Uncertainty modeling in multidisciplinary design optimization

Uncertainty is pervasive throughout the design process of complex engineering system. The sources of uncertainty in multidisciplinary design optimization (MDO) were summarized and classified into three categories; requirement error, input data error and model error. A mathematical model was built up for the estimation of propagated uncertainty in MDO system that contained two coupled disciplines (subsystems), which represented the relationship between input data error, model error and uncertainty in results of MDO system analysis to evaluate the influence of uncertainties for system performance. The agreement of the final computation results with the experimental data indicates this model is effective and reliable. Having extended, this model could be used to estimate the uncertainty in results of MDO system analysis, which contains more coupled disciplines (subsystems). The research is useful for the development of robust MDO methods.