ALA Unit/Subunit: LITA
Meeting Type: Program
Cost: Included with full conference registration.
Rush Rhees library is an old and complex building with there are multiple entry and exit points, and many walkways and connections to other buildings on campus. The architecture makes it very difficult to determine how many people are inside the building at any given point in time. As such, it can be difficult for us to make evidence-based operational decisions regarding staffing and hours, and it can be be hard for us to fully understand how the building is being used by patrons as we work to design programming and other activities. To overcome this gap, we embarked on an ambitious project to use data science methods and predictive analytics to model and better understand traffic flows through the library. In this presentation we'll discuss how we used our library's access control gates to create a stream traffic data, how we pushed those data into Tensor Flow to develop a predictive model, and how we developed data visualizations in Tableau to help decision-makers across the library gain a better understanding of the building's use. We'll also discuss our experiments using computer vision to enrich our dataset, and discuss our plans to move from the pilot phase of the project into sustainment.