Using AI to Predict Changes in Student Employee Staffing Needs

Managing staffing for student employees can be challenging, especially in high-demand service areas like interlibrary loan (ILL). At the Marriott Library, Acquisitions Supervisor Annelise Nicholes Xiao turned to AI to make this process smarter and more efficient. 

The Challenge 

Before this project, student schedules often didn’t align with peak request times. This led to inefficiencies—students were sometimes scheduled during slow periods or “clumped” together on shifts, while busy times lacked adequate coverage. 

The AI Approach 

Annelise used AI to analyze historical request data from library software and generate heatmaps showing the busiest times for ILL requests. These predictive models allowed her to: 

• Forecast demand down to the hour 

• Adjust student schedules to match peak workload periods 

• Create a more stable and productive job environment for student workers 

 

The Outcome 

The predictions proved highly accurate. Student schedules now align with real demand, improving service for library patrons and partner institutions while reducing idle time for student employees. 

Why It Matters 

This project demonstrates how AI can optimize operational efficiency in libraries, benefiting staff, students, and patrons alike. As Annelise notes, “I believe there are more areas in the library that could benefit from predictive AI models, which need to be explored.” She is already working on additional applications, including AI predictions for cataloging and call number creation. 

For more information, contact Annelise by email: Annelise.Xiao@utah.edu