A stochastic model for the simulation of management decisions in dairy herds, with special reference to production, reproduction, culling and income

Abstract

A stochastic simulation model was developed to study management decisions in dairy herds. The primary purpose of this model is to quantify the economic effects of different culling policies with respect to productive and reproductive failure.

Each stimulated herd consists of a fixed number of up to 100 cows (with additional youngstock), individually generated according to various predetermined herd characteristics. Changes in the herd can be followed at 20-day intervals over 15 years. Calculations are repeated 20 times for each alternative, in order to obtain statistically reliable results. Milk production, fat and protein content of the milk, reproductive performance and involuntary disposal are stochastic variables in the model. The major deterministic variables are feed intake, live body weight (including normal carcass value) and number of youngstock.

Calculations to test the model's behaviour have been carried out. Results obtained in the basic situation were consistent with averages observed in practice. A sensitivity analysis was used to compare the economic importance of different management issues under identical conditions. Some parameters commonly used as economic indicators, namely 305-day milk production and Fertility Status, were shown to have limited value. Moreover, the results showed again that improvements in fertility traits alone (such as oestrus detection and pregnancy rate) do not affect income greatly. Therefore, a broadening of the content of herd health programs to include other aspects is desirable from an economic point of view. Breeding and culling policy are particularly important. Income decreased considerably when the maximum allowable calving interval was reduced, despite the positive effect on the average calving interval of the herd. Strategic breeding, taking into account the productive capacity of the cows, was the most profitable policy.

More precisely defined strategic breeding and culling policies will be evaluated and discussed in a subsequent paper. Meanwhile the present model is being adapted for interactive use in teaching of veterinary and agricultural students.