Big data analytics for empowering milk yield prediction in dairy supply chains

Accurate prediction of daily milk production is a crucial aspect of the dairy industry. During the past decades, although many models using various data analytic techniques have been proposed in literature to address the milk prediction problem, these models have yet to be widely applied in daily operations. Dairy producers need to predict milk yield at individual cow and group level. Given the increasing amount of milk production information collected every year, difficulty also arises from analyzing big data. To address challenges in dairy supply chains and help dairy producers, especially small-scale producers, make use of data analytics in milk supply decision-making, a targeted effort to develop a feasible and cost-effective tool, Milk Yield Prediction and Analysis Tool (PAT), is launched. This tool allows dairy producers to use various prediction models to discover insight into milk production and forecast future milk yield at both the individual cow and the group level. This paper provides a detailed discussion on the design of this tool and demonstrates how big data analytics can be applied in a cost-effective manner.