Artificial intelligence for secure and sustainable industrial control systems - A Survey of challenges and solutions

In modern industrial environments, the security and sustainability of Industrial Control Systems (ICS) have become crucial. This comprehensive review examines the transformative potential of Artificial Intelligence (AI) in ICS, focusing on technologies like Machine Learning (ML), Deep Learning (DL), Large Language Models (LLMs), and cloud computing. Moreover, this research explores integrating existing and proposed sustainable practices within the ICS framework, with a particular emphasis on energy efficiency and carbon footprint reduction, to enhance the overall sustainability of ICS. This review employed a systematic approach to select relevant articles from multiple reputable databases, such as Scopus, IEEE Explore, Science Direct, ACM digital library, Web of Science, and IET digital library, including 250 articles that provide valuable insights into the intersection of AI, security, and sustainability in ICS. This review examines vulnerabilities in ICS, such as data breaches, insider threats, and malware, emphasizing the need for effective anomaly detection. It highlights how AI technologies like anomaly detection and predictive analytics can enhance threat detection and response in ICS by improving accuracy and efficiency. The review offers insights to researchers and professionals on the future of secure, sustainable ICS, supporting a resilient industrial landscape that meets cybersecurity, compliance, and sustainability goals.