Explainable AI For Military Supply Chain Optimization Using Sar Images

Abstract

Efficient supply chain management is a fundamental requirement in modern military operations, where rapid and reliable logistical decisions directly influence mission success. Planning transportation routes, allocating resources, and evaluating terrain conditions must often be performed in uncertain and dynamic environments. Synthetic Aperture Radar (SAR) remote sensing has emerged as a valuable technology for such tasks because it provides high-resolution terrain information independent of weather conditions, daylight availability, or atmospheric disturbances.Conventional approaches for interpreting SAR imagery and supporting military logistics frequently depend on manual analysis or rule-based decision systems. These approaches are often labor-intensive, difficult to scale, and susceptible to human error. Although deep learning models have demonstrated strong performance in image analysis tasks, their limited interpretability often prevents their adoption in critical defense applications where transparency and accountability are essential.To address this challenge, this study proposes an Explainable Artificial Intelligence (XAI) framework for military supply chain optimization based on SAR imagery. The proposed system employs a Convolutional Neural Network (CNN) to classify terrain categories that influence military logistics and route planning. To enhance interpretability, explainability techniques such as Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) are integrated with the CNN model. LIME highlights the image regions that influence individual predictions, while SHAP provides both local and global insights by quantifying feature contributions to the model’s output.The proposed framework enables accurate terrain classification while simultaneously providing transparent explanations for model decisions. Such interpretability supports informed decision-making and increases confidence in AI-based systems used in defense logistics. The results demonstrate that integrating explainable methods with deep learning can significantly improve trust, usability, and operational efficiency in military supply chain planning.