Big data analytics helps with cow herding

Posted Aug 9, 2017 by Tim Sandle
Researchers have constructed a digital model that incorporates a financial function to explore the behavior of cow herds. This is an example of how big data analytics and computer modeling can be applied to farming.
File photo: Cows on a farm.
File photo: Cows on a farm.
William West, AFP/File
Agriculture is wreaping the benefits of new technology. One example of this process is with the herding of cows, which has been shown to be more complex than many farmers had realized. For instance, a research group has discovered that what appears to be a randomly dispersed herd, peacefully eating grass, is instead a complex system of individual animals, each facing differing tensions.
To understand this more fully, mathematicians and biologists from the Clarkson Center for Complex Systems Science have come together to construct a computer model that incorporates a cost function to behavior in herds. The aim is to better understand the dynamics of herds and to seek efficiencies in herd management.
The research approach falls into the realm of complex systems. Discussing the application of complex systems to farming, lead researcher Professor Erik Bollt explains: "Cows grazing in a herd is an interesting example of a complex system. An individual cow performs three major activities throughout an ordinary day. It eats, it stands while it carries out some digestive processes, and then it lies down to rest."
This may appear simple but the behaviors of individual cows can only be understood within the wider context of group dynamics. For example, cows eat at varying speeds resulting in herds seeking to move on before the slower cows have finished eating. Here slower cows either stop eating or follow (and remain hungry) or remain behind and are placed at risk from predators.
The model also revealed that ‘slow eating cows’ and ‘fast eating cows’ are not fixed positions; some cows oscillate between the two positions; thus, understanding the complexity of rhythms is also important. Farmers can learn from this and can consider splitting animals into subgroups based on differing nutritional needs. Whether this is worthwhile for the farmer also depends on a cost-benefit analysis, so for this reason a financial aspect was incorporated into the model.
The development of the model and the outcome has been presented to the journal Chaos: An Interdisciplinary Journal of Nonlinear Science. The research paper is titled “Modeling the lowest-cost splitting of a herd of cows by optimizing a cost function.”