Asymptotic Properties of the Hazard Function and the Mean Residual Life Function

Dejan Škanata
DOI: https://doi.org/10.37394/23206.2020.19.27
2020-05-12
WSEAS TRANSACTIONS ON MATHEMATICS
Abstract:The memoryless or non-aging property of systems is of special relevance in reliability theory, which implies that the hazard function is constant in time, and the corresponding mean residual life function takes a reciprocal value. The only known continuous distribution with that property is the exponential distribution. However, many other distributions exist whose asymptotic behavior of underlying hazard functions approaches a constant, while the mean residual life function approaches a reciprocally constant value. Here we provide an analysis which enables us to study a class of distributions that asymptotically approach the memoryless property, and which include gamma, Erlangian, exponential resilience, exponential geometric, hyper exponential, logistic exponential and the inverse Gaussian distribution.
English Else
What problem does this paper attempt to address?