Paul K. Tower

Date of Award


Document Type


Degree Name

Master of Science


Department of Operational Sciences

First Advisor

Shane A. Knighton, PhD


Disruptions impacting workforce schedules can be costly. A 1999 study of the United Kingdom's National Health Service estimated that as much as 4% of the total resources spent on staffing were lost to schedule disruptions like absenteeism. Although disruptions can not be eliminated, workforce schedules can be improved to be more responsive to disruptions. One key area of study that has expanded over the past few years is the application of traditional scheduling techniques to re-rostering problems. These efforts have provided methods for responding to schedule disruptions, but typically require deviations to the disrupted schedule. This thesis examines five workforce scheduling models designed for a nurse rostering problem. Each model is designed to produce a robust workforce schedule that remains valid in the midst of disruptions and requires no schedule deviations. Each model is evaluated based on the number of disruptions it can receive before becoming invalid. Nonparametric statistical analysis is used to analyze the disruption data for each model and determine which workforce scheduling model produces the most robust schedule. The results of this research indicate that additional manpower must be applied to the correct skill sets in order to produce robust workforce schedules. Furthermore, workforce managers can consider leaving a portion of the workforce unscheduled (or in reserve) to accommodate schedule disruptions.

AFIT Designator


DTIC Accession Number