Project Details
Description
SafeSCMS introduces an advanced AI framework to interpret and strengthen decision-making processes in SCMS. The project accentuates transparency and fairness in AI decisions, addressing the interpretability gap and instilling trust in AI's role in critical operational decisions. This is essential for systems where AI significantly influences complex choices.
Layman's description
SafeSCMS is a pioneering project aimed at enhancing the cybersecurity of Supply Chain Management Systems(SCMS) in SMEs through applying Trustworthy AI. Recognizing the increasing reliance on Reinforcement Machine Learning (RML) in SCMS and the accompanying cyber vulnerabilities, SafeSCMS seeks to safeguard these systems against sophisticated cyber threats. This project is an industrial project that we aim to develop with our industrial partner, Digital Readiness and Intelligence Ltd (D-Ready) and 20 other SMEs.The project's innovation lies in its unique focus on monitoring, detecting, and responding to cyberattacks, specifically targeting RML-driven decision support systems. By employing state-of-the-art AI techniques, SafeSCMS will ensure the integrity, fairness, and transparency of AI-driven decisions in SCMS, even in the face of data manipulations by cyber threats.
Key findings
*Real-time anomaly detection.*Bias correction mechanisms.*A human-in-the-loop framework for enhanced decision-making processes.This improves the resilience and reliability of SCMS and fosters stakeholder trust in AI technologies.SafeSCMS aligns with the National Cyber Strategy 2022, supporting its pillars of enhancing cyber resilience, fostering technological innovation, and ensuring national security in the digital domain. This project is a step towards a more secure and resilient digital future for SMEs, essential for the UK's economic prosperity and technological advancement.
Short title | SafeSCMS |
---|---|
Status | Finished |
Effective start/end date | 1/04/24 → 31/07/24 |
Funding
- Innovate UK
Fingerprint
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.