Published Paper Details:

The rapid adoption of Industrial Internet of Things (IIoT) in the pharmaceutical sector has significantly enhanced automation, real-time monitoring, and intelligent decision-making. However, increased connectivity exposes Pharma IIoT systems to sophisticated cyberattacks, including false data injection, ransomware, and denial-of-service attacks, posing risks to drug quality and patient safety. This paper proposes a self-learning digital twin-enabled security framework that integrates real-time telemetry, digital twin modeling, and deep reinforcement learning (DRL) for dynamic attack detection and mitigation. The framework leverages a digital twin environment to simulate operational and attack scenarios, enabling the DRL agent to learn optimal defense strategies. Experimental results demonstrate that the proposed approach outperforms baseline models in terms of accuracy, F1-score, false positive rate, and detection latency, achieving a detection accuracy of 95.8% and a latency of 12 ms. The framework ensures proactive, adaptive, and resilient cybersecurity for next-generation Pharma IIoT systems, providing a viable solution for real-time threat mitigation and operational safety.