Equation Chapter 1 Section 1A Method for Component Parameter Importance of Phased Mission Systems Based on Kriging Model

Importance measures for parameters of system components can provide useful information for system design and management. For phased mission systems (PMSs), such measures need to consider the dynamic dependence of shared components and the change of reliability logics across phases. This paper presents a Kriging model-based method for estimating the relative importance of component parameters, particularly, mean time between failures (MTBF). Based on simulation results of mission reliability with random sampling of component parameters within their value ranges, Kriging model is built and then its coefficients of deterministic trend terms corresponding to component parameters are used as measures of parameter importance. The proposed approach for importance measure is computational efficient and can reasonably reflect the relative importance of a parameter to mission reliability of PMSs. Numerical results are obtained under different configurations and compared with Sobol's indices for an example phased mission system.