| Abstrakt: |
The net financial return of an enterprise depends on the interaction among a great many factors. Some of these factors relate to the animal, some to its diet, some to its environment, some to the prevalence of disease and some to circumstances outside the production enterprise such as the premiums paid for products of different quality, the relative price structure of feeds and products, and the availability and cost of capital, labour, breeding stock and other resources. Although there has been a great deal of research into many of these factors, the complexity of the interactions between them makes it virtually impossible for the human mind to assess accurately the consequences of alternative management strategies on either the efficiency of production or the long-term profitability of a livestock enterprise. By transforming the concepts and knowledge into mathematical equations and integrating them in computer programs using simulation modelling techniques, this vast store of information can be applied directly to improving the management of commercial animal enterprises. Models are also valuable for defining research priorities. These simulation models should, as far as possible, be based on descriptions of the mechanisms perceived to determine animal function, not on empirical relationships of correlation and association. This need for mechanistic models has major implications for the direction and nature of future research into animal function. Mechanistic models of animal performance alone are unlikely to result in the widespread application of knowledge to the animal industries. Models must be integrated with other modules that cover the major areas of an enterprise determining its profitability, as well as with programming features that make the whole Decision Support Software System easy to use and interpret by industry personnel. The animal model is likely to represent less than 20% of a commercially useful package. A major factor limiting the application of animal growth models is lack of an adequate description of the conditions within commercial enterprises. Collection of such data is difficult and frequently regarded as unattractive by scientists and funding organisations, but it is essential for effective application of existing knowledge through simulation models. Furthermore, industry must make frequent measurements of factors determining animal performance and enterprise profitability if the significance of predictions from animal models is to be evaluated fully. An example is presented illustrating how simulation models can improve the biological efficiency and profitability of a commercial animal enterprise when this information is available. |
