Sustainability & efficiency

Genetic parameters of feed intake, eating behaviour, efficiency and body weight recorded using 3D cameras in three dairy cattle breeds

---

Abstract

Despite feed efficiency in dairy cattle is a complex and costly trait to measure, in the last decade it became an important economic parameter incorporated into several breeding programs. However, the amount of data still makes it difficult to make multitrait genetic analysis of related traits to estimate the consequences of selecting for improved efficiency. The CFIT system is a 3D based cow recognition technology that automatically records body weight when a cow leaves the milking unit and measures individual feed intake from a total mixed ration and feeding behavior (including visit duration and frequency) at the feeding station. Currently the CFIT system is implemented on 30 farms in Denmark providing continuous data collection from Holstein, Jersey and Red Dairy Cattle. The aim of this study was to estimate genetic and phenotypic correlation between feed efficiency, production and eating behavioral traits using data from CFIT system. A total of 2.3 mio feed intake records and corresponding milk and daily body weight records collected from October 2024 to October 2025 for 8286 Holstein, 5043 Red Dairy Cattle and 5035 Jersey cows in the period on a total of 30 herds. Five traits were defined and estimated as daily averages from weekly records: feed intake, number of meals, eating time, eating speed, feed efficiency, body weight and energy corrected milk (ECM) from the national milk recording system. Mean and standard deviation (std) across breeds was 58.8 (11.1) kg for feed intake, 35.1 (8.6) kg for ECM, 628 (86.5) kg for body weight, 0.67 (0.11) kg ECM/kg feed for efficiency, 4.43 (1.84) hours/day for eating time, 13.09 (3.3) kg feed/hour for eating speed and 12.7 (3.2) number of mealsData was analyzed using breed-specific multitrait animal model in the DMU software. The model included fixed class effect of herd, lactation and month and a fixed regression of age at calving and squared age at calving as well as a fixed regression of weeks in milk and a squared regression of weeks in milk. Heritability estimates across breeds were moderate to high for all traits ranging from 0.16 for eating speed to 0.65 for body weight. The genetic correlation structure (rg in parenthesis) indicated that feed efficient cows have fewer meals (-0.15), eat for a shorter time (-0.13) but no relationship to eating speed (-0.01) was detected. Heavy cows have far fewer daily meals (-0.43) but no relationship with eating speed (0.09), number of meals (-0.02) or efficiency (-0.09). Cows with high feed intake were less efficient (-0.44), high yielding (0.31), heavy (0.49) and eat for a long time (0.47) but rather slowly (-0.17).The results indicate that feed efficiency has some relationship to number of meals and eating time but not eating speed.

Keywords: 2026

How to Cite:

Bengtson, C., Kupisiewicz, K., Lassen, J. & Stephansen, R., (2026) “Genetic parameters of feed intake, eating behaviour, efficiency and body weight recorded using 3D cameras in three dairy cattle breeds”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2283836. doi: https://doi.org/10.31274/wcgalp.23520

Rights: 1