Productive life span and resilience rank can be predicted from on-farm first-parity sensor time series but not using a common equation across farms

Abstract

A dairy cow's lifetime resilience and her ability to recalve gain importance on dairy farms, as they affect all aspects of the sustainability of the dairy industry. Many modern farms today have milk meters and activity sensors that accurately measure yield and activity at a high frequency for monitoring purposes. We hypothesized that these same sensors can be used for precision phenotyping of complex traits such as lifetime resilience or productive life span. The objective of this study was to investigate whether lifetime resilience and productive life span of dairy cows can be predicted using sensor-derived proxies of first-parity sensor data. We used a data set from 27 Belgian and British dairy farms with an automated milking system containing at least 5 yr of successive measurements. All of these farms had milk meter data available, and 13 of these farms were also equipped with activity sensors. This subset was used to investigate the added value of activity meters to improve the model's prediction accuracy. To rank cows for lifetime resilience, a score was attributed to each cow based on her number of calvings, her 305-d milk yield, her age at first calving, her calving intervals, and the DIM at the moment of culling, taking her entire lifetime into account. Next, this lifetime resilience score was used to rank the cows within their herd, resulting in a lifetime resilience ranking. Based on this ranking, cows were classified in a low (last third), moderate (middle third), or high (first third) resilience category within farm. In total, 45 biologically sound sensor features were defined from the time series data, including measures of variability, lactation curve shape, milk yield perturbations, activity spikes indicating es-trous events, and activity dynamics representing health events (e.g., drops in daily activity). These features, calculated on first-lactation data, were used to predict the lifetime resilience rank and, thus, to predict the classification within the herd (low, moderate, or high). Using a specific linear regression model progressively including features stepwise selected at farm level (cutoff P-value of 0.2), classification performances were between 35.9 and 70.0% (46.7 ± 8.0, mean ± SD) for milk yield features only, and between 46.7 and 84.0% (55.5 ± 12.1, mean ± SD) for lactation and activity features together. This is, respectively, 13.7 and 22.2% higher than what random classification would give. Moreover, using these individual farm models, only 3.5 and 2.3% of cows were classified high when they were actually low, or vice versa, whereas respectively 91.8 and 94.1% of wrongly classified animals were predicted in an adjacent category. The sensor features retained in the prediction equation of the individual farms differed across farms, which demonstrates the variability in culling and management strategies across farms and within farms over time. This lack of a common model structure across farms suggests the need to consider local (and evidence-based) culling management rules when deve...

An Estimate of the Effects from Precision Livestock Farming on a Productivity Index at Farm Level. Some Evidences from a Dairy Farms’ Sample of Lombardy

“…Simple examples follow. The data from each milking is one of the most useful to monitor the risk of mastitis [ 20 ]. This result, jointly with other sensors for early mastitis detection, allows a positive feedback in maintaining high milk yield in the short time as well as in the whole lactation performance (305-d milk yield).…”

Milk losses linked to udder health treatments at dairy farms with automatic milking systems

“…24 hours), while the Lely AMS software, before storage, corrects for the varying number of milkings per day per cow when milking with AMS. To obtain a similar variance as for the Lely milk production data, the daily milk production time series of DeLaval farms were smoothed with a moving average using a span of 3 days before the analysis (Adriaens et al, 2020). The second data set for Lely AMS systems contained all data since the installation of the AMS and thus matches with the daily data, whereas for DeLaval systems the second data set typically comprised a shorter period compared to the daily data.…”

“…Previous work showed that mastitis influences the lactation curves, and therefore classical lactation models including the entire time series cannot be used as such (Andersen et al, 2011). Several authors have recently developed methodologies that predict the unperturbed milk yield trajectory of a cow (Adriaens et al, 2020(Adriaens et al, , 2021Poppe et al, 2020;Ben Abdelkrim et al, 2021), from which the residuals can be used to quantify deviations. The methodology that we developed previously has the advantage of being applicable for QMY trajectories too, which allowed us to use an similar method at both the cow and the quarter level in this study.…”

“…In brief, at cow level, a perturbation was defined as an episode for which the CMY was below the expected yield (ULCF) for at least consecutive days, with at least one day below 80% of the expected production. These criteria are the result of discussions with veterinary experts and are motivated in Adriaens et al (2020Adriaens et al ( , 2021. For the QMY, the higher variability compared to CMY data needed to be considered.…”

“…To this end, a good estimation of the expected milk production in absence of perturbations is needed. For example, we (Adriaens et al, 2018(Adriaens et al, , 2020(Adriaens et al, , 2021 and others (Poppe et al, 2020;Ben Abdelkrim et al, 2021) developed methodologies to predict milk yield in an unperturbed state. These predicted trajectories and their corresponding deviations can subsequently be used to study resilience, for improved monitoring and precision phenotyping as well as for the characterization of milk yield dynamics and losses during challenges.…”