Live animal predictions of carcass components and marble score in beef cattle: model development and evaluation

Predicting fat cover in beef cattle to make on-farm management decisions: a review of assessing fat and of modeling fat deposition

Demands of domestic and foreign market specifications of carcass weight and fat cover, of beef cattle, have led to the development of cattle growth models that predict fat cover to assist on-farm managers make management decisions. The objectives of this paper are 4-fold: 1) conduct a brief review of the biological basis of adipose tissue accretion, 2) briefly review live and carcass assessments of beef cattle, and carcass grading systems used to develop quantitative compositional and quality indices, 3) review fat deposition models: Davis growth model (DGM), French National Institute for Agricultural Research growth model (IGM), Cornell Value Discovery System (CVDS), and BeefSpecs drafting tool (BeefSpecsDT), and 4) appraise the process of translating science and practical skills into research/decision support tools that assist the Beef industry improve profitability. The r2 for live and carcass animal assessments, using several techniques across a range of species and traits, ranged from 0.61 to 0.99 and from 0.52 to 0.99, respectively. Model evaluations of DGM and IGM were conducted using Salers heifers (n = 24) and Angus-Hereford steers (n = 15) from an existing publication and model evaluations of CVDS and BeefSpecsDT were conducted using Angus steers (n = 33) from a research trial where steers were grain finished for 101 d in a commercial feedlot. Evaluating the observed and predicted fat mass (FM) is the focus of this review. The FM mean bias for Salers heifers were 7.5 and 1.3 kg and the root mean square error of prediction (RMSEP) were 31.2 and 27.8 kg and for Angus-Hereford steers the mean bias were -4.0 and -10.5 kg and the RMSEP were 9.14 and 21.5 kg for DGM and IGM, respectively. The FM mean bias for Angus steers were -5.61 and -2.93 kg and the RMSEP were 12.3 and 13.4 kg for CVDS and BeefSpecsDT, respectively. The decomposition for bias, slope, and deviance were 21%, 12%, and 68% and 5%, 4%, and 91% for CVDS and BeefSpecsDT, respectively. The modeling efficiencies were 0.38 and 0.27 and the models were within a 20 kg level of tolerance 91% and 88% for CVDS and BeefSpecsDT, respectively. Fat deposition models reported in this review have the potential to assist the beef industry make on-farm management decisions on live cattle before slaughter and improve profitability. Modelers need to continually assess and improve their models but with a caveat of 1) striving to minimize inputs, and 2) choosing on-farm inputs that are readily available.

Predicting Carcass Weight of Grass-Fed Beef Cattle before Slaughter Using Statistical Modelling

Gaining insights into the utilization of farm-level data for decision-making within the beef industry is vital for improving production and profitability. In this study, we present a statistical model to predict the carcass weight (CW) of grass-fed beef cattle at different stages before slaughter using historical cattle data. Models were developed using two approaches: boosted regression trees and multiple linear regression. A sample of 2995 grass-fed beef cattle from 3 major properties in Northern Australia was used in the modeling. Four timespans prior to the slaughter, i.e., 1 month, 3 months, 9-10 months, and at weaning, were considered in the predictive modelling. Seven predictors, i.e., weaning weight, weight gain since weaning to each stage before slaughter, time since weaning to each stage before slaughter, breed, sex, weaning season (wet and dry), and property, were used as the potential predictors of the CW. To assess the predictive performance in each scenario, a test set which was not used to train the models was utilized. The results showed that the CW of the cattle was strongly associated with the animal's body weight at each stage before slaughter. The results showed that the CW can be predicted with a mean absolute percentage error (MAPE) of 4% (~12-16 kg) at three months before slaughter. The predictive error increased gradually when moving away from the slaughter date, e.g., the prediction error at weaning was ~8% (~20-25 kg). The overall predictive performances of the two statistical approaches was approximately similar, and neither of the models substantially outperformed each other. Predicting the CW in advance of slaughter may allow farmers to adequately prepare for forthcoming needs at the farm level, such as changing husbandry practices, control inventory, and estimate price return, thus allowing them to maximize the profitability of the industry.

Prediction models of intake and productive performance of non-castrated Nellore cattle finished in the feedlot system under tropical conditions

The objective of this study was to understand and predict the intake and performance of Nellore cattle finished in the feedlot. Individual data from 144 non-castrated male Nellore steers finished in the feedlot between the years 2016 and 2020 were used. Descriptive statistical analyses and Pearson's correlation were performed. The outliers were tested by evaluating the studentized residuals in relation to the values predicted by the equations. Residues that were outside the range of - 2.5 to 2.5 were removed. The goodness of fit of the developed equations was evaluated by the coefficients of determination (r²) and root mean square error (RMSE). The mean dry matter intake (DMI) was 10.2 kg/day, neutral detergent fiber intake (NDFI) was 3.4 kg/day, corresponding to 33.3% of DMI, crude protein intake (CPI) was 1.6 kg/day, and total digestible nutrient intake (TDNI) was 7.1 kg/day. The CPI to ADG ratio was 1.3 kg CPI/kg ADG and the TDNI to CPI ratio was 4.5 kg TDNI/kg CPI. The averages of productive performance were 1.3 kg/day for average daily gain (ADG), 152.6 kg for total weight gain (TWG), and 497.8 kg for final body weight (FBW) in average days in the confinement of 115.7 days. The intake measures correlated significantly with the performance measures, except for carcass yield and days in the feedlot. TWG had a high positive correlation with ADG (r = 0.84), while FBW had a positive correlation (r = 0.86) with hot carcass weight (HCW). Measures of intake, performance, and days in the feedlot can be used as predictors of DMI, FBW, HCW, TWG, and ADG. The prediction equations had satisfactory precision and accuracy for non-castrated Nellore cattle finished in feedlot systems under tropical conditions.

Advancements in sensor technology and decision support intelligent tools to assist smart livestock farming

Remote monitoring, modern data collection through sensors, rapid data transfer, and vast data storage through the Internet of Things (IoT) have advanced precision livestock farming (PLF) in the last 20 yr. PLF is relevant to many fields of livestock production, including aerial- and satellite-based measurement of pasture’s forage quantity and quality; body weight and composition and physiological assessments; on-animal devices to monitor location, activity, and behaviors in grazing and foraging environments; early detection of lameness and other diseases; milk yield and composition; reproductive measurements and calving diseases; and feed intake and greenhouse gas emissions, to name just a few. There are many possibilities to improve animal production through PLF, but the combination of PLF and computer modeling is necessary to facilitate on-farm applicability. Concept- or knowledge-driven (mechanistic) models are established on scientific knowledge, and they are based on the conceptualization of hypotheses about variable interrelationships. Artificial intelligence (AI), on the other hand, is a data-driven approach that can manipulate and represent the big data accumulated by sensors and IoT. Still, it cannot explicitly explain the underlying assumptions of the intrinsic relationships in the data core because it lacks the wisdom that confers understanding and principles. The lack of wisdom in AI is because everything revolves around numbers. The associations among the numbers are obtained through the “automatized” learning process of mathematical correlations and covariances, not through “human causation” and abstract conceptualization of physiological or production principles. AI starts with comparative analogies to establish concepts and provides memory for future comparisons. Then, the learning process evolves from seeking wisdom through the systematic use of reasoning. AI is a relatively novel concept in many science fields. It may well be “the missing link” to expedite the transition of the traditional maximizing output mentality to a more mindful purpose of optimizing production efficiency while alleviating resource allocation for production. The integration between concept- and data-driven modeling through parallel hybridization of mechanistic and AI models will yield a hybrid intelligent mechanistic model that, along with data collection through PLF, is paramount to transcend the current status of livestock production in achieving sustainability.