Expert System for Evaluation of Reproductive Performance and Management

Open Access
Publisher Website
Google Scholar
Abstract
A microcomputer expert system for dairy herd reproductive management was developed using an expert system shell and Turbo Pascal. The expert system initially examines the broad areas of days open, days to first breeding, detection of estrus, and conception rate to determine whether a problem exists. Interpretations ranging from "excellent" to "severe" were established for each trait. The system then selects an area for evaluation that has the largest negative influence on days open. Once an area has been selected for further evaluation, the expert system utilizes information from the user and DHI reports developed by the Dairy Records Processing Center in Raleigh, NC. These reports identify problems with conception categorized by production, parity, service number, days in milk, breed, and service sire. In addition, questions are presented by the expert system to isolate problems of accuracy of data, use of natural service, semen handling, AI technique, detection of estrus, signs of estrus, and other management areas. Recommendations and suggestions are given. Ten commercial herds having a conception rate less than 40% were evaluated by the expert system and by an extension reproduction specialist who supplied information for the system. Of 100 areas investigated, the expert system and extension specialist identified 47 as potential problem areas, agreeing on 85% of them. Most discrepancies resulted from the specialist applying a less restrictive standard when values were close to a preselected threshold.