An exact two-phase approach to re-optimize tours in home care planning

Abstract

With the increase in demand, home care agencies must find efficient ways to schedule and route their staff. Unfortunately, the home care sector is never completely stable, the pool of patients constantly evolves, and the staff is subject to a high turnover. Therefore, home care agencies need to regularly update the schedules and to re-plan the visits of careworkers to patients. In this article, we present a method to re-assign the careworkers and re-design their tours whenever a schedule becomes obsolete due to the variations within the pool of patients or the staff. The originality of this work lies in the fact that we study the problem at the strategic level, with close-to-reality constraints. We analyze the impact of three different optimization criteria on the composition of the tours. We tested our algorithm on adapted instances from the literature and on instances extracted from real data provided by a home care agency in France, with up to 15 careworkers, 92 patients, and 337 visits over a week.