Interactions between optimal replacement policies and feeding strategies in dairy herds

a dynamic performance model was integrated with a model that optimised culling and insemination policies in dairy, herds using dynamic programming. The performance model estimated daily feed intake, milk yield and body weight change of dairy cows on the basis of availability and quality of feed and potential milk yield. A set of cow-states was defined by lactation number (1 to 12), calving interval (11 to 16 months), potential milk yield (15 levels) and stage of lactation (months 1 to 16). Actual performance was obtained taking into account potential performance, feed properties, and feed intake constraints. Biological and economical parameters used in the model represented actual production circumstances in Costa Rican herds. Eight feeding strategies combining two forages and four concentrate allocation systems were simulated. Different feeding strategies resulted in maximal changes of 6.8 mo. in optimal average herd-life, US$26.1 in monthly income per cow and 1.9% in replacement rates, while average carving interval was not affected. The main difference was found between feeding strategies based on flat ratios of concentrate compared with feeding strategies based on daily milk yield. Feeding flat ratios of concentrate altered the course of profitability due to restriction of the variation in feeding costs between cows and its effect on animal performance, Average herd-life and monthly income under the optimal feeding strategy were highly sensitive to changes in milk price, but less sensitive to changes in price of concentrates or price of forage. Calving interval was not sensitive to any of the factors. Comparison of optimal policies with actual parameters obtained from field data indicated that cows are being culled close to the optimal herd-life with calving intervals longer than optimal. The model is an efficient tool to study interactions between nutrition, reproduction and breeding at the animal and herd level. (C) 2001 Elsevier Science B.V All rights reserved.