Integrated Framework for Dairy Feeding

Abstract

Herds are typically composed of hundreds to thousands of cows, and are managed in groups of roughly 25 to 100 cows. Since each group of cows contains so many cows, and each herd has so many groups, individual management of each cow within these groups is impossible, as it would take too much time. Also, feed costs represent roughly 60-70% of cost of milk production, so feed efficiency is extremely important. The team approached this problem by attempting to create a system that would allow farmers to manage individual cows in a timely manner. This is done by creating a web-based framework where farmers could simply upload data, and view reports generated by data analysis. The Integrated Framework for the Dairy Feeding project aims to provide a central online platform for dairy production monitoring and optimization. The web platform created by our team provides the farmer with an easy-to-use interface where the farmer can construct new diets for the individual cattle or pens of cattle based on statistical feedback on previous milk composition data and health reports. Throughout the project term, the Dairy Feeding Framework team found that Django is not very compatible with CentOS 7. Rather a containerized virtual machine on the Virginia Tech cloud container cluster was more compatible and flexible than using RLOGIN on CentOS 7.