A decision model for shift scheduling of nurses

The day-to-day scheduling of nursing staff in an acute care hospital causes many problems for nursing administrators, one of which is the determination of very short-term changes in needs for staff. The difficulty lies not only in determining the number of patients for whom care must be provided, but the level of care that will be required by the patients. In this paper we refer to the level of care required as the acuity of care. We define acuity mathematically, for an individual patient, as the number of nurses in the unit needed by one patient during one shift. We represent the number of nurses needed in the unit during the shift as a multiplicative model of mean patient acuity, number of patients, and the mean rate of change in patient acuities. We show that changes in these factors can interact to cause wide swings in the number of nurses needed to staff the next shift. We derive conditions needed for level staffing under certainty, then introduce changes in those conditions in order to study the effect on the number of nurses required. Finally, a simulation model enables us to study the stochastic interaction between the alternative experimental factors for patient acuity, number of patients, and the rate of changes in acuity.