Evaluating failure time probabilities for a Markovian wear process (Q2485167)

Starting from some results of the first author regarding to the closed-form lifetime distribution for a continuous Markovian wear process as a two-dimensional Laplace transform, the authors propose to reduce the lifetime distribution to a one-dimensional Laplace transform in order to facilitate simpler numerical implementation. To this end they present simplified analytical results for the numerical evaluation of failure time probabilities for a single-unit system whose cumulative wear over time depends of its external environment. The main result is a closed-form, analytical expression for the system lifetime in the form of a one-dimensional Laplace transform as opposed to two dimensions. Their result is proved in two ways: a) by directly solving an ordinary (matrix) differential equation via standard methods; b) analytically investing a two-dimensional result with respect to the spatial dimension \(x\), obtained previously. Finally, two numerical examples are provided to compare the one- and two-dimensional results with simulated cumulative probability values. The last section contains some conclusions.