Khushi Thakur, Damini Karnam, M. Rugved Ahmed, D. Saicharan, B. Siris Roya
Abstract
Military supply chain management plays a vital role in ensuring mission readiness and operational success in dynamic and hostile battlefield environments. Traditional military logistics systems largely depend on static planning, manual coordination, and limited situational awareness, making them vulnerable to disruptions caused by terrain uncertainty, infrastructure damage, weather variability, and adversarial threats. To address these limitations, this paper proposes an intelligent and resilient military supply chain optimization framework that integrates Explainable Artificial Intelligence (XAI), machine learning–based predictive analytics, Synthetic Aperture Radar (SAR) image analysis, and adaptive optimization techniques. The proposed framework leverages real-time multi-source battlefield data, including SAR imagery, convoy tracking data, inventory status, and environmental conditions, to enable proactive decision-making. Machine learning models are employed to predict supply demand, route feasibility, and disruption risks, while optimization algorithms dynamically adjust routing strategies and resource allocation. Explainable AI techniques are incorporated to provide transparent and interpretable decision justifications, enhancing trust and usability for commanders and logistics personnel. Simulation-based evaluations demonstrate that the proposed system significantly outperforms traditional military logistics approaches, achieving a 66.7% reduction in decision-making time, a 38.5% improvement in route efficiency, an 87.5% reduction in supply delays, and a 63.6% enhancement in operational resilience. These results highlight the effectiveness of integrating predictive intelligence, optimization, and explainability into military logistics systems. The proposed framework offers a scalable, resilient, and trustworthy solution for next-generation AI-driven military supply chain management.
Keywords
Military supply chain management, military logistics, explainable artificial intelligence (XAI), SAR images, predictive optimization, real-time data analytics, operational resilience.
Conclusion
This paper presented an intelligent, adaptive, and explainable military supply chain optimization framework designed to address the limitations of traditional logistics systems operating in dynamic and high-risk battlefield environments. By integrating real-time data analytics, SAR image–based situational awareness, machine learning–driven predictive modeling, optimization techniques, and Explainable Artificial Intelligence (XAI), the proposed framework enables faster, more accurate, and transparent logistics decision-making under uncertainty. The proposed system effectively overcomes the shortcomings of static and manual logistics planning by continuously analyzing heterogeneous battlefield data to predict demand fluctuations, identify potential disruptions, and dynamically optimize routing and resource allocation. Simulation-based evaluations demonstrate substantial improvements across key performance metrics, including a 66.7% reduction in decision-making time, a 38.5% increase in route efficiency, an 87.5% reduction in supply delays, and a 63.6% enhancement in operational resilience. These results confirm the effectiveness of combining predictive intelligence with adaptive optimization in military logistics contexts. A critical contribution of this research is the integration of Explainable AI, which ensures transparency and trust in automated decisions. By providing interpretable explanations for route selection, resource prioritization, and risk assessment, the system supports human–AI collaboration rather than replacing human judgment. This transparency is essential for mission-critical military operations, where accountability, situational awareness, and rapid validation of decisions are paramount. From a practical perspective, the proposed framework offers a scalable and resilient solution that can support both centralized command centers and decentralized field units, even in communication-constrained environments. The modular architecture allows gradual adoption and customization based on operational requirements and technological maturity, making the framework suitable for real-world military deployment.
References
1.[1] S. Liu, X. Chen, and Y. Wang, “Deep learning-based SAR image analysis for military target detection,” IEEE Transactions on Geoscience and Remote Sensing, vol. 59, no. 4, pp. 3214–3226, 2021, doi: 10.1109/TGRS.2020.3030567.
2.[2] M. T. Ribeiro, S. Singh, and C. Guestrin, “Why should I trust you? Explaining the predictions of any classifier,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, San Francisco, CA, USA, 2016, pp. 1135–1144, doi: 10.1145/2939672.2939778.
3.[3] H. Zhang and Q. Huang, “Explainable AI for supply chain optimization: A review and future directions,” Journal of Intelligent Manufacturing, vol. 31, no. 5, pp. 1179–1197, 2020.
4.[4] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA, USA: MIT Press, 2016.
5.[5] M. Gong, J. Liu, and Y. Chen, “SAR image denoising and feature extraction based on convolutional neural networks,” Remote Sensing, vol. 11, no. 20, pp. 1–17, 2019.
[6] S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” in Advances in Neural Information Processing Systems (NeurIPS), 2017, pp. 4765–4774.
How to Cite This Paper
Khushi Thakur, Damini Karnam, M. Rugved Ahmed, D. Saicharan, B. Siris Roya (2025). SAR Images based XAI Technique for Optimizing Military Supply Chain System. International Journal of Computer Techniques, 12(6). ISSN: 2394-2231.
