Efficient metaheuristics for the home (health)-care routing and scheduling problem with time windows and synchronized visits

Home healthcare and home care centers are facing increasing demands and costs all over the world. Researchers are attracted by several organizational issues related to home care centers’ activities, in particular by the daily routing and scheduling issue. The challenge is to deal with the assignment of visits to home caregivers and the design of sequences of visits execution, minimizing several objective functions separately such as traveling time, preferences and workload balance. This problem is presented in the literature as a vehicle routing problem with synchronization and time window constraints. A benchmark is available, and two efficient metaheuristics have been provided in the literature: a simulated annealing based algorithm (SA-ILS) and an adaptive large neighborhood search (ALNS). In this paper, new metaheuristics; a genetic algorithm, several variants of variable neighborhood descent: three nested and two mixed, and a hybrid genetic algorithm, are provided. Considering fairness as objective function, numerical results show the superiority of the two mixed variable neighborhood descent and the hybrid genetic algorithm, in comparison to SA-ILS and ALNS. Considering preference and traveling time as objective functions, numerical results show that the two mixed variable neighborhood descent outperform the SA-ILS and are competitive with the ALNS.