Securing home health care in times of natural disasters

Abstract

The demand for home health care services is rising tremendously. It is important to maintain these services especially in times of natural disasters. Therefore, powerful algorithms are required to assist decision making. This paper presents a model formulation and solution approach for the daily planning of home health care services. In cooperation with the Austrian Red Cross, one of the leading providers of these services in Austria, we defined seven aims for the objective function. It minimizes the sum of driving times and waiting times, and the dissatisfaction levels of clients and nurses. A feasible solution has to observe assignment constraints, working time restrictions, time windows, and mandatory break times. The model formulation is implemented with the solver software Xpress 7.0 and solved for small problem instances. Real life-sized problems are tackled with a variable neighborhood search (VNS)-based heuristic that is capable of solving even large instances covering 512 jobs and 75 nurses. Extensive numerical studies with real life data from three districts in Upper Austria are presented. A sensitivity analysis shows how different natural disasters may influence home health care services. In 2002 a flood disaster devastated the studied areas. Data from this time and standard flood scenarios, namely the 30, 100, and 200 year return period flood discharges, are taken to depict the consequences on these services. Furthermore, a comparison of the heuristic solution values with an actual route plan shows extensive improvements.