Long-term workforce planning for home healthcare11This project was not funded. All the data used in the study is contained in the paper

This paper presents a new mixed-integer linear programming model for managing the size and composition of a workforce that provides home healthcare services. Decisions center around hiring, training, and downgrading in the face of high resignation rates and a fluctuating imbalance between supply and demand. Novel features of the model include a workforce that is characterized by hierarchical skills and various levels of experience, both affecting individual productivity and operational costs. The optimization problem is to determine a weekly hiring, training, and downgrading plan over the long-term to minimize the weighted sum of costs. Constraints include meeting demand, assuring that patients can be assigned the most appropriate caregivers, and maintaining a target level of skills and experience among the staff. Complications concern an annual turnover rate that exceeds 60% as well as uncertain demand. To validate the model, extensive tests were conducted using data provided by a U.S. home health agency. The results show that optimal solutions can be obtained in a few minutes or less for most instances, depending on the number of patients and caregivers. A major insight gained from the study is that it is possible to derive hiring rules that are simple to implement and closely match optimal plans.