AI-powered early warning and surveillance system to identify the risks within the food supply chain

Projektinformationen

iComplai

ID Finanzhilfevereinbarung: 101114035

DOI

10.3030/101114035

Projekt abgeschlossen

EK-Unterschriftsdatum 1 Juni 2023

Startdatum 1 Juli 2023

Enddatum 31 März 2024

Finanziert unter

• European innovation ecosystems

Gesamtkosten

Keine Daten

EU-Beitrag

€ 75 000,00

Koordiniert durch

ICOMPLAI UG
Germany

Periodic Reporting for period 1 - iComplai (AI-powered early warning and surveillance system to identify the risks within the food supply chain)

Berichtszeitraum: 2023-07-01 bis 2024-03-31

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

iComplai's AI system is a groundbreaking platform designed to enhance food safety risk management using advanced AI technologies. It processes vast amounts of data from various sources in the food supply chain, providing real-time risk predictions and continuous monitoring without human intervention. The system's foundation is its proprietary taxonomy and ontology, which standardizes and organizes data for AI interpretation. The platform also boasts an automated data preprocessing pipeline that refines incoming data, ensuring its quality for AI analysis. A notable feature is the AI models' continuous training capability, which refines predictions over time.

Within the Women TechEU project we aimed:

- To test iComplai in a real operational environment
- To evaluate the test results and optimise iComplai based on these results
- To promote our product and increaste recognition
- To perform an FTO analysis
- To get a complete list of requirements that iComplai needs to comply with
- To carry out a market research study to assess user needs, demand, and market size, to formulate an exploitation and dissemination plan, and a Go2Market Strategies
- To conduct assessment of product development costs, operational costs, marketing and sales costs based on market analysis, BP framework, and design specifications
- To formulate the elaborated BP and Work Plan

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

We have performed a test of iComplai and elaborated an in-depth analysis of raw material risks in food production, with a focus on chilli products.key principles of food safety and risks management included hazard identification, risk assessment, and the implementation of control measures to mitigate risks. The data set revealed significant instances of pesticide residues (69 incidents) and microbial hazards like Salmonella (25 incidents).
We have detected for the first time the following substances in 5 years for chilli peppers: Acesulfame Potassium, Benomyl, Capsaicin, Fenpropathrin, Imidacloprid, Isoprothiolane, Phenthoate, Pyriproxyfen, and Pyrrolizidine alkaloids. The analysis of the substances that have led to recalls in chilli products reveals the following:
- Allergen: 3 recalls;
- Bacteria, Bacillus Cereus, Bacillus, Undeclared Sesame, Salmonella, Undeclared Milk, Undeclared Soy, Foreign, Misbranding: 1 recall each.

The highest number of recalls are associated with allergens. This indicates a significant concern in the food safety domain, particularly regarding the accurate labeling of potential allergens in food products. Recalls due to bacterial contamination, including Bacillus and Salmonella, and mislabeling issues such as undeclared sesame, milk, and soy, also emerge as critical issues.

These findings highlight the importance of robust allergen management and accurate labeling practices in the food industry. Ensuring accurate and comprehensive labeling, especially for allergens, is not only a regulatory requirement but also a crucial step in protecting consumer health and preventing costly recalls.

The results of our work within the Women TechEU project indicates that companies should prioritize effective allergen control programs, rigorous testing for microbial contaminants, and meticulous review of product labeling to mitigate these risks. For importers in Europe, where regulations on allergens and labeling are stringent, these practices are particularly vital to maintain compliance and safeguard public health.

Through our pesticide risk prediction, which focuses on unknown-unknowns, we were able to predict the pesticides that are not yet publicly known but are likely to come up on the pepper. The prediction is based on a product / origin combination (e.g. chilli pepper from India).

Through identification of the risks through news streams we were able to identify further risks in the area of food fraud and pests.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

We have developed our fully automated AI-powered analytics platform and extended our ML algorithm to cover further hazard areas like food fraud, additionally, our platform also collects data on risks in the food supply chain related to extreme weather, modern slavery (forced labour and child labour), human environmental damage (deforestation, pollution, and waste.), economical risks (price changes and sanctions), geopolitical conflicts, pest & diseases, geophysical events, animal welfare, and supplier risks. Using 80 sources from Regulatory Authorities including the EC RASFF, FDA, CFIA, Australia New Zealand notifications, Japan border controls, UK recalls, Korea notifications, Taiwan border controls, Ireland recalls; And local, national and international news (ca. 2200 sources) including the WHO, FAO, EU Law, EFSA, CBC, NGOs, etc. The platform includes translations from 8 different languages.

Our current system offers real-time monitoring, enabling businesses to address potential safety risks promptly. Personalization is central to iComplai's approach, tailoring risk assessments to individual business needs. The platform is scalable, catering to businesses of all sizes. Emphasizing transparency, the system uses explainable AI techniques and adheres to ethical AI principles, prioritizing data privacy and unbiased predictions. In essence, iComplai is automating food safety management with its AI-driven, ethical, and personalized approach.

transparent-logo.png