The EFRA project is pleased to announce the release of the Flock Performance Outlier Detection Dashboard, a new decision‑support tool developed under Use Case #1: Risks Prediction for Poultry Pathogens. This tool enhances poultry‑farm management by enabling early identification of abnormal flock behaviour, supporting faster and more informed decision‑making.

A New AI‑Powered Tool for Early Detection

The dashboard analyses flock‑level mortality and weight‑gain patterns to detect abnormal behaviour early in the production cycle. By combining interactive visualisations with AI‑generated alerts, it helps farm managers identify emerging issues, prioritise interventions, and prevent performance loss.

The tool integrates 2 dedicated functionalities:

• Mortality‑Rate Outlier Detection Dashboard
• Flock Weight‑Gain Outlier Detection Dashboard

These capabilities allow users to monitor flock performance from multiple angles and respond proactively to deviations that may indicate health, welfare, or management challenges.

AI Models and Data Behind the Tool

The dashboard is powered by the Clustering Risk Assessment in Poultry Farming model, a clustering‑based AI approach designed to identify outlier patterns in production data.

It uses:

• Structured flock‑level production data from Moy Park (daily mortality counts, weight‑gain measurements, flock characteristics)
• Complementary laboratory test data

This combination ensures robust detection of anomalies and supports evidence‑based decision‑making.

Co‑Development and Validation

The tool was jointly developed and validated by:

• Agroknow – AI model development, validation, and dashboard deployment
• Moy Park – provision of historical flock data and domain expertise

Accessing the Dashboard

The Flock Performance Outlier Detection Dashboard is deployed on the FOODAKAI Platform. If you already have access to FOODAKAI, you can visit or request access to the dashboard here: Flock Performance Outlier Detection