System reliabilty based on diffusion models for fatigue crack growth

Diffusion processes are commonly used to describe the dynamics of complex systems arising in a wide range of application fields. In this paper we propose, on the basis of diffusion processes, two models concerned with the stochastic behavior of fatigue cracks in a system. They are then used to get the distribution of the failure time, the first time the crack size of at least one of the cracks exceeds a given value. Several properties of our proposed models are presented, and the unknown parameters are estimated by the method of maximum likelihood. From these an estimate of failure time distribution is obtained. In this part, contrary to common practice, we do not assume availability of failure data.