Replacement of PILC/PICAS joints using dynamic programming for optimization and Weibull model for reliability assessment

In response to a recent rise in the number of PILC/PICAS joint failures at a utility, the company has set targets to replace a significant number of ageing joints over the next 4 years. This paper aims to provide a methodology for replacement optimization using Dynamic Programming (DP). As part of the methodology, an objective function and constraints is proposed where the goal is to determine the optimal number of yearly replacements which minimizes total cost whilst maintaining an acceptable level of risk. DP is used to find the replacement combination which returns the minimum cost whilst satisfying the constraints. A case study is conducted to investigate the impact cost of replacement, cost of failure and total acceptable replacement limit has on the optimization results. The optimal replacement strategy was obtained. By comparing the optimal replacement strategy against an average unoptimized replacement strategy, it was found that utilities can expect a 14% cost saving and a 40% reduction in the total number of predicted failures by using the replacement optimization methodology detailed in this paper.