Bayesian stockpile reliability methodology for complex systems Academic Article uri icon

abstract

  • Bayesian reliability modeling of complex systems, such as a missile system, can allow considerable flexibility to incorporate component and subsystem level data, expert knowledge and full system tests. In this paper we present a unified method for developing a model that is able to consistently estimate reliability parameters for all elements of the model, from system level through to subsystem and component levels. The model can be adapted to model the various components of the system at different levels of granularity, depending on where data are available. In addition, this paper presents some direction about how Bayesian priors can be selected to incorporate expert knowledge of the system in a variety of ways. An example of a complex system based on a missile is used to illustrate the methods.

author list (cited authors)

  • Anderson-Cook, C. M., Graves, T., Hamada, M., Hengartner, N., Johnson, V. E., Reese, C. S., & Wilson, A. G

complete list of authors

  • Anderson-Cook, CM||Graves, T||Hamada, M||Hengartner, N||Johnson, VE||Reese, CS||Wilson, AG

publication date

  • January 2007