<![CDATA[Effective oil and gas pipeline management requires a data-driven approach to identify segments with varying risk characteristics. This study aims to classify pipeline segments based on protective infrastructure conditions using the Block-Based K-Medoids clustering method. The analysis considers six variables: Pipeline Burial, Pipe Along Road, Pipe Guards, Berm/Rail/Guard Condition, Public Road, and ROW HCA along a 59-kilometer pipeline corridor. Data were normalized, and the optimal number of clusters was determined using the Deviation Ratio Index based on Medoid (DRIM), which indicated three clusters as the most representative structure. The results demonstrate clear differentiation among segments in terms of exposure level, protective condition, and HCA involvement, enabling classification into low-, moderate-, and high-risk groups. Spatial visualization further confirms systematic risk distribution along the route. These findings provide a structured basis for prioritizing inspection, maintenance, and mitigation strategies in pipeline infrastructure management.]]>
Copyrights © 2026
