A new and efficient computation method of IM (integrated importance measures) for components in binary coherent systems

The introduction of the full paper reviews a number of papers on importance measures in the open literature and points out what we believe to be their shortcomings; then, it proposes what we believe to be a new and efficient computation method mentioned in the title. Section 1 explains how we established our IM (Integrated importance Measures); its core consists of: (1) we brief the computation methods and meanings of classic importance measures; (2) we put forward IM for binary coherent systems to describe the mathematical expectation of the system reliability decrease based on current component unreliabilities; (3) we analyze the computation relationships between IM and classic importance measures. Section 2 discusses the properties of IM; its core consists of: (1) we prove mathematically the calculation method and physical meaning of IM for typical series system; (2) we prove mathematically the calculation method and physical meaning of IM for typical parallel system. Section 3 does the numerical study of a hybrid system; the results, given in Tables 1, and their analysis show preliminarily that, by considering the component reliabilities and failure rates together, our IM is indeed efficient for component importance analysis in binary coherent system.