• Dr. DHIRAVIDAMANI
• Mr. S.SANJEEV

In this study, we propose a cost-effective approach to predict inline inspection (ILI) results using machine learning (ML) models. Three prediction cases were considered: detecting defects, predicting defect dimensions, and estimating defect growth rates. Cat boost (CAT) emerged as the optimal ML method across all cases, offering high accuracy in predicting defect presence and characteristics. By leveraging pipeline attributes and environmental features, our approach significantly reduces unnecessary ILI costs and provides valuable insights for pipeline maintenance and management. Accurate predictions and insightful correlation analyses contribute to substantial cost savings and informed decision-making in pipeline integrity management.

In-line inspection, Machine learning, Pipeline defect, SHAP values, Remaining life

"Enhancing Pipeline Safety and Predicting Remaining Life: Leveraging Machine Learning Techniques and SHAP Interaction Values", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 6, page no.f351-f372, June-2024, Available :http://www.jetir.org/papers/JETIR2406539.pdf

"Enhancing Pipeline Safety and Predicting Remaining Life: Leveraging Machine Learning Techniques and SHAP Interaction Values", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 6, page no. ppf351-f372, June-2024, Available at : http://www.jetir.org/papers/JETIR2406539.pdf