Survival analysis of water distribution pipe failure data using the proportional hazards model

In this paper the hazard rates of the cast iron 6 inch (about 150 mm) pipes of a study area are modeled by implementing the proportional hazards modeling approach for consecutive pipe failures. A methodology of identifying individual pipes according to the internal and external characteristics of pipe is applied to a case study water distribution pipe break database. The individual cast iron 6 inch pipes are categorized into seven ordered survival time groups (STGs) according to the minimum total number of breaks recorded in a pipe. The modeling results show that the failure times of all of the STGs have the Weibull distribution. In addition, the estimated baseline survival functions show that the survival probabilities of the STGs generally decrease as the number of break increases for a given time. It is found that STG I has an increasing baseline hazard rate whereas the other STGs have decreasing baseline hazard rates.