Home health nurse routing and scheduling
The Supply Chain and Logistics Institute
Georgia Institute of Technology
Home health care is the business of delivering professional health care services in the home. The demand for home care services is growing as the elderly population and the number of people living with chronic illnesses increases. Unfortunately, the supply of skilled health care professionals is not expected to be sufficient to meet demand. This nation is experiencing a nursing shortage that is even more prevalent in the home care industry. One way to narrow the gap between the demand for and supply of home health care services is to use existing resources in the most effective, efficient manner. We study the problem of increasing resource utilization by improving home health nurse routing and scheduling. The resultant problems are variants of periodic routing problems.
In the home health nurse routing problem, there are a set of patients that need to
be visited during the planning horizon. Each patient has a frequency with which they
require visits, typically specified in visits per week. Each patient also has a set of allowable
visit day combinations. For example, if a patient has a frequency of two visits per week,
the allowable visit day combinations for that patient could be
{Monday, Wednesday} or
{Tuesday, Thursday}. Given this information, a set of routes for each day of the planning
horizon should be determined. Each route must start and end at the nurse’s home station
and visit each customer exactly once. Furthermore, the set of routes for the entire planning
horizon must visit each patient the number of times specified by their frequency. These
visits can only occur on a visit day combination which is allowable for that patient. The
most important ob jective is to minimize the total travel time or distance required by all
routes over the planning horizon. Secondary ob jectives include balancing workload and
maintaining regularity of visit schedules for patients from week to week.
Although special cases of the home health nurse routing problem can be reduced to the classical traveling salesman problem and vehicle routing problem, most problems encountered are much more complex instances of the periodic TSP or VRP. An additional complicating factor is the dynamic aspect to the problem. Not all patient demand is known a priori ; it becomes known at varying intervals during the planning process. Some patient demands are only revealed the day of the requested visit. We study first a static variant of the problem. In this phase, we determine how to assign customers to visit day combinations so that the resulting routes will be good. We also address route construction and improvement through local neighborhood search. Next, the dynamic version of the problem is considered. Here, we determine how to create flexible partial schedules that are robust with respect to adding new patients.
