The Security Concerns On Cyber-Physical Systems And Potential Risks Analysis Using Machine Learning
Muammer Eren Sahin, Lo’ai Tawalbeh, Fadi Muheidat
Abstract
The use of engineering to drive down costs and improve productivity has been an ongoing business exercise since the first Industrial Revolution. The term Cyber-Physical System is a wide range of different computing technologies embedded with the next-generation engineered systems into the physical world. Connected Cyber-Physical Systems (CPS) improve the lives of people and increase industry and manufacturing efficiency. It is affecting many branches of life such as transportation, healthcare and medicine, the environment, and energy. Industry 4.0 integrates humans, machines, and data to provide a holistic and interlinked approach to manufacturing, hence, increasing privacy concerns. For example, Autonomous Vehicles (AV) can be driven without a pilot and those systems can be hacked if there is a breach in the system. Nowadays, most of the systems are interconnected to the internet and nothing can be considered fully safe. Therefore, with this increase of security threats and privacy concerns, there is a need to assess and evaluate the trade-off between enhancements and improvements in manufacturing and the possible threats and security risks in the context of Cyber-Physical Systems. We need to bridge the gaps and overcome some of these limitations. In this work, we studied the security concerns emerging from interconnected Cyber-Physical systems, devices, and services in Industry 4.0. To identify security vulnerabilities, we have chosen the energy dataset because energy is the key point of every Cyber-Physical system so aimed to show the importance of energy, and the K-Means algorithm implemented which is an advanced Machine Learning and potential risks detected.