Speaker
Description
Artificial insemination (AI) is widely used to improve the genetic quality and productivity of water buffalo in the Philippines through the nationwide program of the Department of Agriculture–Philippine Carabao Center (DA-PCC). Despite ongoing efforts, calf production remains low and inconsistent across regions. This study analyzes five years of AI service data comprising 278,978 records from DA-PCC regional centers to identify factors associated with successful calf production. Several machine learning models were developed and evaluated to predict calf production outcomes, with Extreme Gradient Boosting achieving the best performance across standard evaluation metrics. The results show that the assigned service center and the distance between technicians and buffalo farms are among the most important predictors, with longer travel distances associated with lower calf production rates. Motivated by these findings, a mathematical optimization framework is proposed to improve the delivery of AI services. The resulting AI Technician Routing Problem integrates farmer–technician assignment, route planning, and semen inventory considerations into a unified model formulated as a mixed-integer linear program. The framework provides a quantitative approach for improving the efficiency and reliability of AI service delivery in water buffalo production systems.
