Search

The present investigation was carried out on 216 records of Osmanabadi goat and 208 records of Deccani sheep. The relative efficiency of three different methods, viz. Shaffer's formula method, multiple linear regression and artificial neural network for prediction of live body weight were investigated. Individual animals were weighed on electronic weighing balance along with their body measurements like lengths, height and girths were measured. The explanatory variables were body lengths, body height and chest girths while dependant variable was body weight. A multilayer feed forward neural network with back propagation of error learning mechanism was developed using artificial neural network using bayesian regularization algorithms. It was observed that artificial neural network was best fitted with in goat and sheep, with the adjusted R2 of 0.93, explained by its linear relationship with the explanatory variables in goat. However, the prediction accuracy (R2 value) was observed as 94.21% with 2.35 kg error. While in sheep, the adjusted R2 was 0.82 and the prediction accuracy (R2 value) was observed as 85.29% with 3.48 kg error. The multiple linear regressions observed the adjusted R2 of 0.894, and the prediction accuracy (R2 value) as 90.03% with 4.73 kg error. The correlation coefficients for different body measurements using three different methods were ranged from 0.952 to almost one.

In Osmanabadi goats kids, stall feeding is the most economic and comfortable method of feeding management. The overall growth performance in Osmanabadi goat kids was enhanced by supplementation of concentrate mixture with regular roughage feeding. Further, in Osmanabadi goats, there was positive linear correlation of chest girth, height at withers, body length, height at hip with BCS, body weight and the age. The supplementation of concentrate mixture along with roughages in the feeding management of Osmanabadi kids resulted in heavier body weight. The trend of improvement in different growth and body parameters of Osmanabadi goat kids showed positive linear correlation with the amount of concentrate feed consumed by the kid. The supplementation of concentrate mixture with regular roughages in Osmanabadi goat kids improved the level of fodder intake. Stall feeding and the supplementation of concentrate mixture or nutritional enrichment of regular diet of Osmanabadi goat kids results in the improvement in voluntary fodder intake, however the rate of intake remained in positive linear correlation up to the 50% amount of concentrate feed consumed on dry matter requirement basis in Osmanabadi goat kids. Positive linear correlation of concentrate supplementation with voluntary fodder intake ultimately resulted in reduced fodder leftover and thereby the fodder wastage.

The present study was carried out on 538 monthly milk yield records of Marathwadi buffaloes maintained at Cattle Breeding Farm, College of Veterinary and Animal Sciences, Udgir Dist-Latur Maharashtra spread over a period of 9 years. The monthly milk yield was recorded at 6th, 36th, 66th and so on till standard lactation with 30 days interval. The goodness of fit of 3 lactation curve models, viz. parabolic exponential function, inverse polynomial function and mixed log function were used on monthly milk yields in Marathwadi buffaloes. The highest coefficient of determination for fitting of lactation curve models was with mixed log function (97.63%), followed by parabolic exponential function (95%) and inverse polynomial function (82.70%). Further, the root mean square error was found least for mixed log function (1.70), followed by parabolic exponential function (3.55) and inverse polynomial function (4.61). It is therefore inferred that mixed log function fitted best for prediction and modeling of lactation curves based upon monthly milk yields.

Artificial neural network models are machine-learning systems, a type of artificial intelligence. They have been inspired by and developed along the working principles of the human brain and its nerve cells. It is used in the modelling of non-linear systems. With the information learned through repeated experience, similar to human learning, artificial neural network can provide classification, pattern recognition, optimisation and the realisation of forward-looking forecasts. Artificial neural network has manifold applications in the field of livestock and allied sections for prediction of milk production, prediction of breeding values of bulls, estrous detection, mastitis prediction and lameness detection, detection of cows with artificial insemination difficulties, prediction of success rate of invitro fertilization, manure nutrient content, volatile fatty acids in the rumen of dairy animals. Artificial neural network models were determined to be more successful than cluster analysis. Most of the published works in data analysis use linear models for forecasting the production parameters; however, sufficient literature proved that by using artificial neural network better results obtained as compared to linear or classical methods. The present manuscript is an attempt to review the systematic information available in livestock and allied sector.

Gir cattle in Saurashtra provide milk, manure and draught power. Animal genetic resources of native breeds are largely underutilized in India and indiscriminate crossbreeding has degraded large numbers of local breeds. This is also true with Gir cattle in its native breeding tract. (i) About two-third population is either graded Gir or Gir like and hence not pure Gir, (ii) only one-third population i.e. 7.6 lakh could be 95% towards pure Gir. Hence, it is difficult to obtain large number of pure Gir animals. Since 2003–04 to 2012– 13, contribution of Saurashtra to state total indigenous cattle milk yield is 60%. A 25 years data analysis on Gir cows at CBF revealed average lactation yield as 2029 litre in 321 days lactation length and 1843.8 litre milk yield in 300 days. Production performance of Gir cows in three districts:(1) Junagadh (2) Rajkot and (3) Bhavnagar showed that average test day milk yield in Gir cows over a lactation were 7.99±0.05 litre/d for Junagadh, 6.43±0.03 litre/d for Rajkot and 6.71±0.04 litre /d for Bhavnagar district, projected values for 300-days milk yield being was 2397, 1929, and 2013 litre for the three districts respectively. Average AFC in Gir cattle of the station herd was 1527.7±14.1 days (51 mth) whereas survey study in Junagadh, Rajkot and Bhavnagar districts indicated AFC in field Gir cows as 45, 49, and 46 month, respectively. Parity had a significant effect on total lactation milk yield in Gir cows at CBF. Highest total lactation yield was observed in 2nd lactation (2132±123.4 litre) while lowest was observed in 10th lactation (1456.8±219.1 litre). Milk production traits LY and 300-d LY exhibited 10 to 13 % genetic variability in the herd. This suggests that to increase variability in the herd, genetic material from outside herd needs to be introduced.