Policy Implications of Hybrid Predictive Models for Pipeline Integrity Management: Enhancing Maintenance Decision Support in Nigeria’s Oil and Gas Sector

Abstract

Nigeria’s oil and gas sector depends on an extensive pipeline network that has sustained cumulative economic losses exceeding US $658.5 million from pipeline failures between 1990 and 2022 (Ekeu-wei and Ekeu-wei, 2024). Predominantly, reactive maintenance regimes have failed to arrest this trend. This paper explores the policy implications and national impact of a hybrid neural–statistical predictive maintenance framework applied to Nigeria’s pipeline sector. The framework integrates four deep-learning architectures—Feedforward Artificial Neural Networks (ANN), Long Short-Term Memory (LSTM) networks, Convolutional Neural Networks (CNN), and Graph Neural Networks (GNN)—with Generalised Additive Models (GAM) and Exponentiated–Exponentiated Weibull (EEW) survival distributions to predict Remaining Useful Life (RUL) with calibrated uncertainty bounds. Model evaluation against Nigeria-specific pipeline sensor data demonstrated that the hybrid framework outperformed all benchmarks: it achieved a Root Mean Square Error (RMSE) of 8.3 days, a Mean Absolute Error (MAE) of 6.1 days, and a coefficient of determination (R²) of 0.96, compared with R² values of 0.93 (LSTM), 0.91 (ANN), and 0.82 (traditional Weibull). For binary failure classification, the hybrid model attained 95.4% accuracy, 92.1% precision, 96.3% recall, and an F1-score of 94.1%. These results indicate that sector-wide adoption of predictive maintenance could reduce annual operational costs by approximately 35%, cutting average unscheduled downtime from 30 to 12 days per year and halving repair and replacement expenditure. The paper contextualises these findings within Nigeria’s regulatory landscape—particularly the Petroleum Industry Act (2021) and the Nigerian Upstream Petroleum Regulatory Commission (NUPRC) regulations—and provides evidence-based policy recommendations for policymakers and industry stakeholders to improve operational safety, reduce costs, and mitigate the environmental impact of pipeline failures.