Understanding the mechanisms behind pipeline failures is critical for identifying vulnerabilities in gas transmission pipelines and developing strategies to enhance energy supply chain reliability. Pipelines are recognised as the most cost-effective and reliable solution for energy transportation, playing a vital role in modern societies among increasing global demand for natural gas resources. Despite their reliability, pipelines are susceptible to various deterioration mechanisms, making them prone to catastrophic incidents. While fatalities due to pipeline leaks are relatively low in absolute numbers, the consequences pose significant threats to human safety, the environment, and economic stability. Pipeline failures result from cumulative aging processes influenced by physical, operational, and environmental factors. Time-dependent hazards such as corrosion evolve progressively, while time-independent hazards like natural disasters and third-party activities pose immediate risks. The interplay among these factors requires a comprehensive understanding to prioritize maintenance and implement effective risk control measures. This study proposes an innovative approach to analyse historical pipeline failure data, based on incident records from 1970 to 2023 provided by the Pipeline & Hazardous Materials Safety Administration (PHMSA) of the United States. With the United States housing 65% of the world’s pipeline length, a dataset of 12,182 incidents from 1970 to 2023 provides a unique opportunity for analysis. However, the lack of comprehensive historical failure data analysis underscores the need for predictive models capable of identifying multi-cause-and-effect relationships. To address this gap, by offering a data-driven and precise prediction of incident years, this study enhances understanding of underlying causes and circumstances, enabling interventions to mitigate future incidents. By adopting a “lessons learned” perspective, this study provides strategic insights for operators to proactively address potential vulnerabilities, promoting sustained operational integrity and minimizing unexpected events throughout pipeline service life. This study converts data from self-contained case reports into a user-friendly knowledge framework. The results are expected to assist pipeline operators in evaluating and predicting the condition of existing gas pipelines, enabling them to prioritize inspections and maintenance activities effectively. By using the heat maps created in this study, potential failure points can be identified proactively, assisting the implementation of timely maintenance procedures tailored to the estimated service life of each pipeline. By leveraging this knowledge for proactive maintenance, organizations can mitigate risks, enhance operational efficiency, and gain a competitive advantage in their industries.
