Identifying causality and contributory factors of pipeline incidents by employing natural language processing and text mining techniques

Using thousands of PHMSA pipeline incident narratives, the authors apply NLP and text mining to identify contributory factors and infer latent causal dependencies. They compare K‑means clustering with co‑occurrence network analysis and show the network approach better captures dependencies among factors. For PSM incident investigation, this provides a scalable way to extract underlying factors from free text, reveal recurring causal linkages across incidents, and feed those insights into risk models, barrier management, and prevention programs.