How dynamic scheduling can provide more efficient delivery of home care services

Senior couple with man in wheelchair

New research from the Schulich School of Business shows that the use of “dynamic scheduling models” can be applied to more effectively assign patients requiring home care services to medical providers such as registered nurses, physical therapists, or personal support workers.

The findings are contained in the paper, “Dynamic scheduling of home care patients to medical providers,” published in Productions and Operations Management. The article was co-written by Adam Diamant, associate professor of operations management and information systems at Schulich, together with Andre A. Cire, associate professor of operations management at the Rotman School of Management and the Department of Management at the University of Toronto Scarborough.

Adam Diamant
Adam Diamant

“Our research proposes a scheduling framework that assists in the assignment of new patients to home care providers employed at a home care agency and determines the order in which those providers should visit patients that have been assigned to them,” says Diamant. “Our proposed model balances the revenue associated with the provision of health services with the cost of making inefficient assignments, which can include everything from excessive amounts of travel to providers working overly long shifts.”  

The model, adds Diamant, captures key real-world factors such as the uncertainty associated with the number of daily referrals, a patient’s varying health requirements and their expected duration of care, continuity-of-care constraints, shift-length regulations, and the spatial distribution of both patients and home care providers.  

“From a managerial perspective, our analysis reveals three key insights,” says Diamant. “First, we demonstrate that when assigning patients to home care providers, incorporating accurate predictive models about their future workload is essential to minimizing excessive costs. This is a non-trivial task as these predictions are a function of the scheduling policy, and we provide guidance as to the structure of these predictive models in the paper. Second, we find that well-performing policies typically assign home care providers to patients living within a small set of geographic regions. However, effective scheduling policies must also consider a patient’s expected duration of care to appropriately balance practitioner workload in the long-run. Finally, our quantitative framework is especially effective in environments where travel time is a large component of a home care provider’s shift. This includes large geographic regions such as rural communities or dense urban centres with high average commute times.”