Deep Learning-Driven Cybersecurity for Digital Supply Chains: A Hybrid Ensemble Approach to Real-Time Threat Detection

Ajonuma Michael E.; Mgbeafulike Ike J.

doi:10.56201/ijcsmt.vol.12.no4.2026.pg126.133

📄 Download Paper

Deep Learning-Driven Cybersecurity for Digital Supply Chains: A Hybrid Ensemble Approach to Real-Time Threat Detection

Ajonuma Michael E., Mgbeafulike Ike J.

Abstract

Rapid digitalization has increased the attack surface in Supply Chain Management , making organizations highly vulnerable to emergent sophisticated cyberattacks. Traditional signature-based security remains for the most part reactive and unable to cope with fast- evolving malware, phishing, ransomware, and insider intrusion. This research proposed a hybrid ensemble deep learning cybersecurity system for real-time anomaly detection within Digital Supply Chains . The existing system faced some challenges which include deficiency in both theoretical frameworks and practical solutions to proactive cyberattack detection mechanisms and inability to adapt to evolving threats in digital supply chain. The above challenges was tackled by developing a deep learning models that optimized deep learning algorithms capable of analyzing diverse data sources within the digital supply chain, including transaction records, network logs, and sensor data, to detect and mitigate cyber threats in real-time and also created an adaptive cyber security system for digital supply chain systems using deep learning algorithms capable of continuously learning from new data. Following an Agile iterative methodology for system development in six sprints involving activities on data preprocessing, classical machine learning baselines, neural network development, hybrid model integration, model evaluation, and operational deployment, using the MalwareData dataset, the best performance of 98.38% accuracy with an F1-score of 0.97 from Random Forest (RF) was obtained in comparison with 95.62% from Logistic Regression and 94.49% from the standalone Neural Network. Even more robustness with fewer false positives was exhibited with the hybrid ensemble model by amalgamating machine learning (ML) and deep learning (DL) techniques in order to reinforce threat detection. Streamlit powered by SQLite has been used for deployment, rendering a responsive and user-friendly interface for real-time monitoring within simulated DSC environments. This research proposed a scalable and adaptive cybersecurity solution for modern supply chains.

Keywords

Supply Chain ManagementDeep LearningHybrid EnsembleDigital Supply ChainDigitalizationmalware IJCSMT

References

[1] Abdusalomov, A. B., Nasimov, R., Nasimova, N., Muminov, B., and Whangbo, T. K. “Evaluating synthetic medical images using artificial intelligence with the GAN algorithm”, Sensors, 2023; 23(7), 3440. https://doi.org/10.3390/s23073440. [2] Abid, R., Rizwan, M., Veselý, P., Basharat, A., Tariq, U., and Javed, A. R. “Social networking security during COVID‐19: A systematic literature review”, Wireless Communications and Mobile Computing, 2022; (1), 2975033. https://doi.org/10.1155/2022/2975033. [3] Abosuliman, S. S. “Deep learning techniques for securing cyber-physical systems in supply chain 4.0”, Computers and Electrical Engineering, 2023; 107, 108637. https://doi.org/10.1016/j.compeleceng.2023.108637. [4] Ageron, B., Bentahar, O., and Gunasekaran, A. “Digital supply chain: Challenges and future directions”, Supply Chain Forum: An International Journal, 2020; 21(3), 133– 138. https://doi.org/10.1080/16258312.2020.1816361. [5] Akam, M. J., Sunday, E. G., Etuk, I. U., Ejikeme, O. B., and Arikpo, N. N. “The role of integrated coordination in supply chain performance of firms in the manufacturing industry”, International Journal of Integrated Supply Management, 2023; 16(1), 26– 51. https://doi.org/10.1504/IJISM.2023.130853. [6] Al Mashalah, H., Hassini, E., Gunasekaran, A., and Bhatt, D. “The impact of digital transformation on supply chains through e-commerce: Literature review and a conceptual framework”, Transportation Research Part E, 2022; 165, 102837. https://doi.org/10.1016/j.tre.2022.102837. [7] Aldasoro, I., Gambacorta, L., Giudici, P., and Leach, T. “The drivers of cyber risk”, Journal of Financial Stability, 2022; 60, 100989. https://doi.org/10.1016/j.jfs.2022.100989. [8] Alkhalil, Z., Hewage, C., Nawaf, L., and Khan, I. “Phishing attacks: A recent comprehensive study and a new anatomy”, Frontiers in Computer Science, 2021; 3, 563060. https://doi.org/10.3389/fcomp.2021.563060. [9] Basha, S. S., Dubey, S. R., Pulabaigari, V., and Mukherjee, S. “Impact of fully connected layers on performance of convolutional neural networks for image classification”, Neurocomputing, 2020; 378, 112–119. https://doi.org/10.1016/j.neucom.2019.10.008. [10] Cha, S. Y. “The art of cybersecurity in the age of the digital supply chain: Detecting and defending against vulnerabilities in your supply chain”, The digital supply chain (pp.215–233), 2022. [11] Chauhdary, S. H., Alkatheiri, M. S., Alqarni, M. A., and Saleem, S. “An efficient evolutionary deep learning-based attack prediction in supply chain management systems”, Computers and Electrical Engineering, 2023; 109, 108768. https://doi.org/10.1016/j.compeleceng.2023.108768. [12] Khan, S. A., Naim, I., Kusi-Sarpong, S., Gupta, H., and Idrisi, A. R. “A knowledge- based experts system for evaluation of digital supply chain readiness”, Knowledge- Based Systems, 228, 107262, 2021.

Deep Learning-Driven Cybersecurity for Digital Supply Chains: A Hybrid Ensemble Approach to Real-Time Threat Detection

Abstract

Keywords

References

More Articles from INTERNATIONAL JOURNAL OF COMPUTER SCIENCE AND MATHEMATICAL THEORY