Water Supply Pipeline Retrofit Optimization According to Spatial Clustering of Pipeline Groups

The pipeline group divided by physical attributes often includes many individual pipelines spatially scattered in a large geographical area, yielding the replacement of these pipelines requires frequent service interruption and additional costs on the scattered delivery of repair resources. To address this issue, this paper proposes a framework for pipeline replacement optimization on pipeline groups divided by spatial clustering, aiming to reduce the number of scattered individual pipelines in the replacement scheme. The proposed framework integrates spatial autocorrelation analysis for spatial clustering of pipeline groups as replacement candidates, the pipeline failure model to predict potential failures of pipeline groups, and the replacement optimization model of pipeline groups. The optimization model aims to minimize the number of potential failures with the constraint of the annual budget. The framework was implemented in a real WDN, and pipeline replacement schemes obtained by the proposed framework are compared with two other methods, namely, pipeline attribute clustering-based optimization and pipeline risk-based ranking. Results show that the spatial clustering-based grouping helps to reduce the number of spatially scattered individual pipelines by 36.7% and 64.6% compared to the other two methods, respectively. The proposed framework is expected to provide more cost-benefit schemes for pipeline replacement.