Scalability and robustness of feed yard mortality prediction modeling to improve profitability

Abstract Cattle feed yards routinely track and collect data for individual calves throughout the feeding period. Using such operational from nine U.S. years 2016–2019, we evaluated scalability economic viability of using machine learning classifier predicted mortality as a culling decision aid. The expected change in net return per head when predictions aid compared to status quo protocol having been pulled at least once bovine respiratory disease was simulated. This simulated ranged −$1.61 $19.46/head. Average standard deviation this study $6.31/head $7.75/head, respectively.