Enhancing Manufacturing Security with AI-driven Anomaly Detection
Aims
This project aims to develop a secure AI model to detect and mitigate anomalies in manufacturing, improving efficiency and safeguarding against cyber threats.
Overview
“In today's manufacturing industry, where digital and AI technologies are becoming essential, the risk of cyber threats has significantly increased. Our project aims to tackle this issue by creating an AI-driven anomaly detection model tailored for the manufacturing sector, such as the food packaging industry. This model is designed not only to spot irregularities and anomalies in real-time but also to protect against specific AI-related threats like model poisoning and evasion attacks. By doing this, we hope to enhance operational efficiency, reduce downtime, and strengthen cybersecurity in manufacturing environments. Over the next six months, our team will design, test, and refine this AI model, ensuring it’s secure, reliable, and meets the unique needs of modern manufacturing.”
NW CyberCom is a £1.2 million project aiming to unlock the cyber security potential of the North West. Led by Lancaster University, the project sees six partner universities capture the latest cyber security innovations, working with entrepreneurs, investors, government and businesses to transform cutting-edge knowledge into new products, services and policy. The primary goal is to strengthen protection for consumers, businesses, and UK infrastructure.
Results and Outcomes
Tab Content: For Partners and Engagement
The AI-driven anomaly detection model is designed to make a tangible difference in manufacturing by enhancing both productivity and security. By collaborating closely with industry partners, the team have developed a prototype that demonstrates the model's ability to detect anomalies in real-time, which is critical for preventing defects, reducing downtime, and ensuring safety. They have also engaged with key stakeholders in the industry to validate the approach and gauge market interest. Feedback has been overwhelmingly positive, particularly from those concerned with the rising cyber threats in manufacturing. The success of this project is evident in the early adoption interest seen, along with the strong performance of the model in testing scenarios, where it has effectively identified anomalies and resisted potential cyber attacks.
Tab Content: For Academics
As part of the NWCybercom programme this project has been a valuable learning experience, particularly in the integration of AI and cybersecurity within the manufacturing industry. One key takeaway is the importance of developing AI models that are not only accurate but also resilient against adversarial attacks. The collaborative approach between cybersecurity experts and industry professionals was crucial in understanding the real-world challenges and refining our model to address them effectively. In hindsight, starting with a broader range of test scenarios could have provided even more robust initial data. For colleagues working on similar projects, one key recommendation is prioritising early and having continuous engagement with industry partners to ensure that the developed solutions meet practical needs and can be effectively deployed in real-world settings.
Back to listing