Modelling the repair warranty of an industrial asset using a non-homogeneous Poisson process and a general renewal process

This paper considers selecting the warranty period after the completion of a series of successive repairs on a product. Two stochastic failure models are used: the general renewal process (GRP) model and the non-homogeneous Poisson process (NHPP) model. Both use a Weibull distribution for the life time of the product, allowing the possibility to renewal (GRP) or not (NHPP) when successive repairs are performed. The NHPP is applied to estimate the warranty period after simple repairs (minimal), while GRP is used when repairs are complex (overhauls). The main novelty of this work is the process followed to estimate the warranty period, which can be defined according to the risk that a company is willing to assume in the market. Our procedure provides a useful tool for a maintenance company or for a manufacturer's after-sales division.