Investigation into Lean Meat Yield (LMY) algorithms in Livestock Data Link (LDL)

Summary

Livestock Data Link (LDL) was established by MLA as a tool for producers and processors to source, review and receive feedback on carcase data, compliance to market specifications and the cost of non-compliant carcases for beef and sheep.

This project assessed the performance of the Lean Meat Yield percentage (LMY%) predication algorithms in LDL on beef and sheep carcases, to determine any potential improvements in accuracy for commercial processing facilities and their supply chains.

The project report details several areas in which improvements could be made, or where additional activities could be included in the LDL program, such as the use of an LMY% prediction based on carcase weight and P8 fat for cattle.

Objectives

The overall objectives of the project were to:

• prepare a register of existing LMY% prediction algorithms, determine what attributes they use, the expected range of carcase LMY% and the accuracy of such algorithms
• test the accuracy and range of LMY% predications from a range of cattle and sheep processors in different markets
• assess the potential to capture missing attributes that prevent LMY% from being predicted in LDL.

Key findings

Cattle:

• There was a moderate correlation between LMY% derived from rib-fat versus P8.
• Although using an equation based on P8 fat (rather than rib-fat) would significantly increase the number of animals with an LMY% prediction, implementation would be limited due the current level of reduced accuracy, which means further research is required in this area.
• It is currently unlikely that single point measures of carcase traits (e.g. rib fat and P8 fat) are going to provide robust estimates of LMY% when compared to other objective measurement technology such as Dual Energy X-Ray Absorptiometry (DEXA).

Sheep:

• The conversion of “fat depth measured in mm” to “fat score” resulted in a loss of discriminatory information from LMY%, which only caused a small change in the actual LMY% calculated.

Benefits to industry

LMY% is a key indicator of carcase efficiency. Enhancing LDL to include data for LMY% will increase the prediction, understanding and extrapolation of data that can improve the production and processing efficiency of beef and sheepmeat in the Australian red meat industry.

MLA action

MLA is investing in the research and development of alternative methods for estimating LMY% in sheep and cattle using technologies such as DEXA, microwave and video image analysis.

As processors implement these technologies, they will capture LMY estimates for a far wider range of carcase and animal attributes than are currently possible.

Future research

The project report details specific recommendations to improve the prediction of LMY% in the LDL program, including:

• improving the algorithm that utilises P8 fat measurement
• adjusting the prediction equation to exclude, or allow for, independent values outside acceptable limits
• beef processors that want to produce LMY% using LDL data should invest in alternative technologies that potentially measure LMY% directly
• LDL should seek to capture breed data, to allow for a more tailored prediction of LMY% that removes some of the inherit breed biases.