Applying AI to enhance Traceability Challenges and Food Authenticity in Food Supply Chains

Project Details


Food needs accurate records to demonstrate authenticity and quality, otherwise consumers can pay more for lower quality food, or worse, unsafe food. Scientific product testing has limitations and will not necessarily pick up fraudulent activities such as changing date codes or falsifying country of origin.

Our research-consultancy for Red Tractor in 2016/17 showed that traceability challenges - used by assurance and crime agencies in the sector - could be significantly enhanced by identifying opportunities to override systems. In 2018, scandals in the meat industry involve just this. Being able to interrogate all data rather than small samples would increase the chances of these anomalies being detected. Although Blockchain is being proposed as a solution by IBM, Walmart and others, it is still not proven to be fraud proof. Other AI solutions are likely to be more robust and more cost effective.

We will engage with organisations in food assurance services and standard setting to apply AI in the form of anomaly detection algorithms to look for patterns within the traceability data. This will have 2 key benefits:
1. It will identify management and reporting weaknesses in the quality assurance system.
2. It will identify potentially fraudulent behaviour.

Key findings

1. Clear road-map for future research direction
2. Funding bid to be submitted shortly
3. Review journal paper being prepared
Effective start/end date2/04/1831/07/18


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.