The EFRA project announces the release of the Poultry Safety Incidents Forecasting Dashboard, the second decision‑support tool developed under Use Case #1: Risks Prediction for Poultry Pathogens. This tool enhances the ability of food‑safety professionals to anticipate emerging poultry‑related hazards by analysing historical incident data and generating predictive insights.

AI‑Driven Forecasting for Proactive Risk Management

The dashboard analyses publicly reported poultry‑related food‑safety incidents and produces predictive trends that support early detection of potential risks. Through interactive time‑series visualisations, forecasted incident counts, and drill‑down access to incident‑level details, users can explore trends, identify hotspots, and investigate underlying causes.

A particular emphasis is placed on pathogens of concern, such as Salmonella, enabling regulators, food manufacturers, and risk‑management teams to strengthen preventive actions and improve operational planning.

Powered by the Poultry Incidents Time‑Series Model

The tool is built on the Poultry Incidents Time‑Series Model, a forecasting model developed and validated by Agroknow.

It integrates multiple data sources harvested and processed by Agroknow, including:

• Public food‑safety incident records
• Laboratory testing results
• Weather and meteorological data
• Unstructured text from incident and recall reports
• Structured time‑series incident data

This multi‑layered dataset enables more accurate forecasting and deeper insight into the drivers of poultry‑related food‑safety events.

Development and Validation

The tool was developed and validated by Agroknow, which led the AI model development, validation, deployment, and dashboard implementation.

Accessing the Dashboard

Access to this decision‑support tool is provided through the FOODAKAI application. As this is a subscription‑based dashboard, you may visit or request access to the: Poultry Safety Incidents Forecasting Dashboard