STUDY ON TEST DAY RECORDS TO PREDICT STANDARD LACTATION MILK YIELD USING ARTIFICIAL NEURAL NETWORK IN GIR COWS 3033
Loading...
Date
2020-02
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
JAU, JUNAGADH
Abstract
The present study was carried out using data of first lactation monthly test-day
milk yield (MTDY) and 300-day milk yield records of 513 Gir cows sired by 75 bulls
spread over a period of 33 years (1986 to 2018), maintained at Cattle Breeding Farm,
Junagadh Agricultural University, Junagadh. The overall least squares means for age
at first calving (AFC), first lactation 300 day or less milk yield (FL300DMY) and first
lactation length (FLL) were 1558.63 ± 23.30 days, 1506.36 ± 40.10 kg and 341.26 ±
6.62 days, respectively. Season of calving/birth had non-significant effect on AFC,
FL300DMY and FLL. Period of calving/birth had highly significant (P<0.01) effect
on AFC, FL300DMY and FLL. Age at first calving had non-significant influence on
FL300DMY. While calf birth weight had significant (P<0.05) effect on FL300DMY.
The overall least squares means for monthly test day milk yield (MTDMY)
varied from 3.39 kg (MTDMY10) to 6.40 kg (MTDMY3). Only
MTDMY3 (peak yield phase) was influenced significantly (P<0.05) due to variation
of season of calving. The period of calving had highly significant (P<0.01) effect on
all MTDMY. The AFC had significant (P<0.05) effect on MTMDY1, MTMDY2 and
MTMDY5. However, calf birth weight had highly significant (P<0.01) effect on
MTDMY1 and MTDMY2 while its effect was significant (P<0.05) on MTDMY4.
The heritability estimates for FL300DMY, MTDMY and FLL were found to
be 0.306 ± 0.139, 0.102 ± 0.117 (MTDMY10) to 0.657 ± 0.179 (MTDMY3) and
0.267 ± 0.135, respectively.The estimates of genetic and phenotypic correlation among all the monthly
test day milk yield ranged from 0.133 ± 0.440 to 0.996 ± 0.312 and 0.276 to 0.903
respectively, while between all MTDMY and FL300DMY were higher and positive
ranged from 0.523 ± 0.324 to 0.991 ± 0.049 and 0.482 to 0.884, respectively.
The optimum equation was developed for prediction of FL300DMY using
monthly test days records. The optimum equation had total four variables (test days)
viz. TD2 to TD5 for prediction of FL300DMY. This equation gave an accuracy of
prediction of 76.02% by MLR and 87.69% by ANN model. It was inferred that
FL300DMY could be predicted as early as 125th days of lactation with sufficiently
higher degree of accuracy. The comparison between MLR and ANN revealed that
ANN was better than MLR for prediction of FL300DMY in Gir cows.