Predicting feed intake using modelling based on feeding behaviour in finishing beef steers.
Predicting feed intake using modelling based on feeding behaviour in finishing beef steers.
复制标题
DOI：
10.1016/j.animal.2021.100231
复制
发表时间：
2021-07
期刊：
影响因子：
--
通讯作者：
Duthie CA
中科院分区：
文献类型：
--
作者：
Davison C;Bowen JM;Michie C;Rooke JA;Jonsson N;Andonovic I;Tachtatzis C;Gilroy M;Duthie CA
目前用于测量圈养牛的饲料摄入量的技术既昂贵又耗时，使得它们不适合在商业农场使用。为了评估生产效率，需要估计每头动物的摄入量。本研究的目的是预测个体动物摄入量的参数，可以很容易地获得商业农场，包括饲养行为，体重和年龄。总共使用了80头阉牛，每头阉牛被分配到两种饮食中的一种（每种饮食40头），其中包括（g/kg; DM）饲料与精料的比例为494：506（混合）或80：920（CONC）。在56天期间，使用32个电子喂食器记录每只动物的个体每日鲜重摄入量（FWI; kg/天），随后计算个体DM摄入量（FWI; kg/天）。从电子喂食器计算测量期间每天的个体喂食行为变量，包括：喂食器访问总次数、在喂食器上花费的总时间（TOTFEEDTIME）、消耗饲料的总时间（TIMEFEED）和每次喂食器访问期间的平均时间长度。由于易于从加速度计获得，因此选择这些进食行为变量。基于（i）个体动物TOTFEEDTIME相对值（表示为饮食组（GRP）和总GRP摄入量的比例）、（ii）多元线性回归（REG）、（iii）随机森林（RF）和（iv）支持向量回归（SVR），检查了预测个体动物摄入量的四种建模技术。每个模型分别用于预测C 0 NC和MIXED饮食，得到八个预测模型，（i）GRP_C 0 NC、（ii）GRP_MIXED、（iii）REG_C 0 NC、（iv）REG_MIXED、（v）RF_C 0 NC、（vi）RF_MIXED、（vii）SVR_C 0 NC和（viii）SVR_MIXED。每种模型都在FWI和ESTA上进行了测试。使用重复测量相关性（R2_RM）评估模型性能，以捕获与标准R2、RMSE和平均绝对误差（MAE）相比每日摄入量的重复性。REG、RF和SVR模型预测FWI的R2_RM = 0.1-0.36，RMSE = 1.51-2.96 kg，MAE = 1.19-2.49 kg;预测FWI的R2_RM = 0.13-0.19，RMSE = 1.15-1.61 kg，MAE = 0.9-1.28 kg。GRP模型预测FWI的R2_RM = 0.42-0.49，RMSE = 2.76-3.88 kg，MAE = 2.46-3.47 kg;预测FWI的R2_RM = 0.32 - 0.44，RMSE = 0.32 - 0.44 kg，MAE = 1.55-2.22 kg。虽然更简单的GRP模型显示出比回归和机器学习技术更高的R2_RM，但这些模型具有更大的误差，这可能是由于没有捕获个体喂养模式。虽然回归和机器学习技术产生了与个人摄入量相关的较低误差，但预测的总体精度太低，无法实际使用。
Current techniques for measuring feed intake in housed cattle are both expensive and time-consuming making them unsuitable for use on commercial farms. Estimates of individual animal intake are required for assessing production efficiency. The aim of this study was to predict individual animal intake using parameters that can be easily obtained on commercial farms including feeding behaviour, liveweight and age. In total, 80 steers were used, and each steer was allocated to one of two diets (40 per diet) which consisted of (g/kg; DM) forage to concentrate ratios of either 494:506 (MIXED) or 80:920 (CONC). Individual daily fresh weight intakes (FWI; kg/day) were recorded for each animal using 32 electronic feeders over a 56-day period, and individual DM intakes (DMI; kg/day) subsequently calculated. Individual feeding behaviour variables were calculated for each day of the measurement period from the electronic feeders and included: total number of visits to the feeder, total time spent at the feeder (TOTFEEDTIME), total time where feed was consumed (TIMEWITHFEED) and average length of time during each visit to the feeder. These feeding behaviour variables were chosen due to ease of obtaining from accelerometers. Four modelling techniques to predict individual animal intake were examined, based on (i) individual animal TOTFEEDTIME relative expressed as a proportion of the dietary group (GRP) and total GRP intake, (ii) multiple linear regression (REG) (iii) random forests (RF) and (iv) support vector regressor (SVR). Each model was used to predict CONC and MIXED diets separately, giving eight prediction models, (i) GRP_CONC, (ii) GRP_MIXED, (iii) REG_CONC, (iv) REG_MIXED, (v) RF_CONC, (vi) RF_MIXED, (vii) SVR_CONC and (viii) SVR_MIXED. Each model was tested on FWI and DMI. Model performance was assessed using repeated measures correlations (R2_RM) to capture the repeated nature of daily intakes compared with standard R2, RMSE and mean absolute error (MAE). REG, RF and SVR models predicted FWI with R2_RM = 0.1–0.36, RMSE = 1.51–2.96 kg and MAE = 1.19–2.49 kg, and DMI with R2_RM = 0.13–0.19, RMSE = 1.15–1.61 kg and MAE = 0.9–1.28 kg. The GRP models predicted FWI with R2_RM = 0.42–0.49, RMSE = 2.76–3.88 kg and MAE = 2.46–3.47 kg, and DMI with R2_RM = 0.32–0.44, RMSE = 0.32–0.44 kg, MAE = 1.55–2.22 kg. Whilst more simplistic GRP models showed higher R2_RM than regression and machine learning techniques, these models had larger errors, likely due to individual feeding patterns not being captured. Although regression and machine learning techniques produced lower errors associated with individual intakes, overall precision of prediction was too low for practical use.
登录
查看更多内容
影响因子：
3.5
作者：
Chizzotti, M. L.;Machado, F. S.;Ribas, M. N.
通讯作者：
Ribas, M. N.
影响因子：
2.3
作者：
Martiskainen, Paula;Jarvinen, Mikko;Mononen, Jaakko
通讯作者：
Mononen, Jaakko
影响因子：
3.2
作者：
Koknaroglu, H.;Otles, Z.;Hoffman, M. P.
通讯作者：
Hoffman, M. P.
影响因子：
3.8
作者：
Bakdash JZ;Marusich LR
通讯作者：
Marusich LR
影响因子：
3.6
作者：
Montanholi, Y. R.;Swanson, K. C.;Miller, S. P.
通讯作者：
Miller, S. P.