Short-term workforce management: Cross-training, scheduling and allocation of heterogeneous nurses in a hospital
Additional Document Info
In this paper we present an integrative approach that addresses cross-training, scheduling, and allocation of nurses in a hospital and evaluate various cross-training policies. Using data collected at a hospital, we study several elements of the workforce planning process. The workforce planning process is considered in its entirety, including: (1) planning, (2) scheduling, and (3) allocation and adjustment. Although our main focus is on the latter two phases, the results from an experimental study can be used to help make planning decisions. In the scheduling phase, we assign heterogeneous workers to a set of shifts over a planning horizon. We use a bi-criteria formulation that minimizes nurse labor costs and the number of undesirable shifts for each employee. We include an objective for minimizing the number of undesirable shifts because of the increasing rate of employee turnover in healthcare settings. By explicitly accommodating schedule desirability, we aim to generate tours for nurses that are as close to their choice schedule as possible. In the allocation and adjustment phase, we allocate cross-trained float pool nurses and reassign cross-trained unit nurses at the beginning of a shift to one of several units in the hospital. This phase is where healthcare managers respond to real-time bed census information and patient acuity. This is an area that has received limited attention in the literature. In addition, we investigate the impact of cross-training policies through a full factorial experiment. Several experimental factors such as level of cross-training and proportion of cross-trained nurses are used to help make planning decisions (i.e., what cross-training policy to implement). Thus, although we do not explicitly model the planning phase, we show how the results from our analysis can assist in the decision processes found in that phase.
name of conference
Proceedings - Annual Meeting of the Decision Sciences Institute