Using Adaptive Replacement to Minimize Risk in the Oil and Gas Industry

Abstract

presented a decision theoretic approach for determining optimal replacement strategies under replacement and repair scenarios. The Bayesian approach, adaptive in nature, takes into account failure and survival information at each planned replacement stage to update the optimal time until the next planned replacement. Under the assumption in which an item is replaced by a new one upon failure, the underlying process between two planned replacement times is a renewal process. The replacements upon failure that may occur between the planned replacement stages constitute renewals of the underlying renewal process. The times between renewals are the lifetimes of the items. Mazzuchi and Soyer (1996) made the Weibull assumption for the lifetime distribution of an item and used an approximation due to Smeitink and Dekker (1990) to compute the renewal function. Robinson and Aboura 2015) enhanced the adaptive approach by presenting a method for the exact calculation of the renewal function and its derivative due to Constantine and Robinson (1997). The method of finding zeros of a function, by Muller (1956) and Frank (1958), is adapted to the maintenance optimization problem, making use of the availability of the derivative of the renewal function. Robinson and Aboura (2015) made further improvements by providing a methodology for the assessment of the joint prior distribution of the parameters of the Weibull lifetime model. The prior distribution is determined through the specification of initial reliability estimates for different mission times. To provide a simple approach to