The Sustainability Framework Report 2019 raised opportunities to optimise performance, health, flock consistency and egg quality. Hen-house egg production can differ significantly in between flocks, but also in between housing systems.

The majority of challenges applies to the free-range egg industry, where a larger range of factors can compromise hen performance and subsequently farm economics. In addition, the free-range sector experiences the highest market value share, making this sector to be most risky to operate. Problems frequently seen in the Australian Egg free-range industry include peak-of lay timing (hens start laying too early or too late), drops in egg production related to environmental factors (range access, weather condition), the amount of off-grade eggs, and unfavourable egg characteristics at end of lay.

Machine learning algorithms are used in everyday life to help identify patterns and rules in datasets. Poultry production is a data rich environment and machine learning can be used as an early detection tool to spot disease. Similar models can be built using Australian production data.

This project will undertake two activities and associated extension to identify barriers to adoption in under performing farms to improve create targeted extension strategies based on behaviour change theory. The second part of the project will target farms with good egg production and preform in-depth diagnostics and data analysis to identify irregular patterns and create predicative models to help alleviate these with the help of machine learning to generate personalised management solutions on farm.

The objective is to help farmers achieve reliable production patterns and reduce the number of off grade eggs through increasing adoption of best practice and the use of machine learning to assist management.