Application of Bayesian Inference in Wood Pole Line Condition Assessment

This paper introduces the inference procedures based on Bayes theorem to perform point and interval estimations on a simulated wood pole line population. The objective is to define the scope of replacement according to wood pole condition assessment rules developed by asset analytics. The construction of prior distribution by numerical simulation provides an example to explain how it works with sampling evidence to calculate the posterior distribution. The results are also compared with estimations by classical statistics. Discussions of asset condition scoring algorithm, logistic regression, Bernoulli process and Bayes theorem are included in the paper.