<p>Understanding subsurface faults is crucial for hydrocarbon exploration and production. To assess fault seal effectively, a combination of artificial neural network (ANN) and the “Move” was employed. The study aims to compare ANN, known for its non-linear regression capabilities, and the “Move,” which analyzes multiple fault seal factors for fault seal prediction. The objectives of this study are to delineate hydrocarbon-bearing reservoirs, map faults and horizons, and characterize faults in terms of their orientation and throw. Lithology differentiation was achieved using gamma ray logs, while reservoir identification and correlation across wells relied on resistivity and gamma ray logs, as well as log response similarities. A network of faults and three horizons were mapped, identifying a faulted anticline as the likely hydrocarbon-bearing structure. The study utilized well log and three-dimensional seismic data in the “Move.” The ANN’s performance was assessed using different evaluation metrics. The “Move” indicated that fault planes exhibited moderate to good sealing capacity, with average shale gouge ratio values of 35%, 36%, and 44% across wells. Lithology juxtaposition included shale on sand, sand on sand, and shale on silt. The ANN model accurately predicted fault seals with 93% <i>R</i>² (coefficient of determination) success rate. Validation of the ANN results, compared to Move predictions, showed superior performance through a scatter plot analysis. This study demonstrated that machine learning techniques, when applied to well logs and seismic data, offer substantial potential for enhancing fault seal prediction in the petroleum industry.</p>