PEO Willowdale-Thornhill Chapter Virtual Technical Seminar

Taylor Martin's Bio:

Taylor is a Professional Engineer and PhD candidate in Healthcare Operations Research at the University of Toronto. His research applies machine learning to simulate and optimize the structures and schedules of the nursing workforce. Taylor is also a graduate of the Masters of Health Administration program at the Institute of Health Policy, Management and Evaluation at the University of Toronto. Taylor is presently the Director of Workforce Optimization at University Health Network. He has led the hospital scheduling transformations initiatives and managed staffing operations across several GTA organizations.

Description of Technical Seminar:

Hospital staffing is critical to the provision of quality care, financial outcomes, and staff wellbeing. Despite even the best of workforce planning and scheduling practices, the reality of daily operations is that staffing is highly vulnerable to staff absenteeism. This talk will describe long standing operational and staff scheduling challenges in healthcare and share recent research applying machine learning and operations research to forecast and optimize staffing levels to achieve material operational and financial benefits. In this study, a nurse-aware absence forecast was developed using XG Boost classifiers. The ML forecast was compared to other forecasting techniques to positive result. Forecast outputs were used as demand in an Integer Program to set optimal replacement staffing levels under operationally-relevant constraints. Results showed that replacement staffing could be reduced by 5% using the nurse-aware forecast model.