Subject Training

Context: Digital twins (DTs) have emerged as a cornerstone of Industry 4.0, enabling real-time monitoring, prediction, and optimization of complex systems. By creating a dynamic digital replica of a physical system, DTs provide decision-makers with enhanced visibility and the ability to test scenarios without disrupting operations. They are increasingly deployed in various domains, including smart manufacturing, energy systems, transportation, and healthcare [3, 1]. The power of a digital twin lies in its continuous synchronization with the physical asset, achieved through streams of sensor data, control commands, and feedback mechanisms. However, this tight integration also creates a wide attack surface. On one hand, the communication link between the physical and the digital systems is exposed to network-based threats. Data packets can be intercepted, delayed, modified, or injected by an attacker, leading to false states in the twin and potentially dangerous decisions in the physical world. For example, falsifying temperature readings in a smart factory could cause the DT to underestimate overheating risks, resulting in damage to equipments [2]. On another hand, the digital twin environment itself (typically deployed on cloud, edge, or virtualized platforms) faces classic cybersecurity risks. Intruders may attempt to access the DT platform to exfiltrate sensitive data, modify the simulation model, or execute malicious commands that propagate back to the physical system [2, 4].

Problem Statement: The reliability of a digital twin depends not only on its modeling accuracy but also on the trustworthiness of its data flows and the resilience of its hosting environment. If an attacker manipulates the communication channel, the DT may operate on falsified information, reducing its accuracy and possibly causing harmful decisions. Similarly, if the DT model or platform is compromised, attackers could gain unauthorized control, steal sensitive process data, or alter the behavior of the physical system itself. The challenge is therefore twofold : 1. How to protect the integrity, confidentiality, and availability of communication links between the physical system and the digital twin in the presence of cyber threats? 2. How to secure the digital twin model and platform against intrusions, tampering, and misuse, while maintaining real-time performance and scalability?

Références [1] Blessing Airehenbuwa et al. « Advancing Security with Digital Twins : A Comprehensive Survey ». In : arXiv preprint arXiv :2505.17310 (2025). [2] Mohammed El-Hajj. « Leveraging digital twins and intrusion detection systems for enhanced security in IoT-based smart city infrastructures ». In : Electronics (Switzerland) 13.19 (2024), p. 3941. [3] Abdul Rehman Qureshi et al. « A survey on security enhancing Digital Twins : Models, applications and tools ». In : Computer Communications (2025), p. 108158. [4] Ali Sayghe. « Digital Twin-Driven Intrusion Detection for Industrial SCADA : A Cyber-Physical Case Study ». In : Sensors 25.16 (2025), p. 4963.