Prediction of external corrosion using data from in-line inspections (ILI) can be instrumental in making proactive pipeline integrity management decisions, especially when integrated with geospatial data. This work investigates how a relationship may be established between the presence of bending-strain areas and the initiation and or progression of external corrosion. As the initiation of external corrosion requires both failure of the external coating and failure of the cathodic protection system, it may be possible to limit the damage caused in areas of bending strain by adopting inspection and repair strategies reflecting the pipeline integrity risks. The study examines how integrating geohazard characteristics into machine learning models improves the prediction of external corrosion. The paper provides insights into the complex interplay of geohazards, bending strain, and corrosion, informing better integrity management strategies. This study suggests that a shift towards a holistic, automated, machine learning-assisted approach for pipeline integrity can be attained in compliance with regulation and highlights the importance of data-driven approaches for safety and reliability.