Seminar: “A Predictive and Prescriptive Analytics Approach to Next-Day Operating Room Scheduling Problem,” Enis Kayış, Özyeğin University, EA-409, 1:40PM April 15 (EN)

Title: A Predictive and Prescriptive Analytics Approach to Next-Day Operating Room Scheduling Problem
by Enis Kayış,Özyeğin University, Industrial Engineering Department

April 15, Friday 13:40

Operating room (OR) is one of the most critical and expensive resources in hospitals. OR scheduling has an impact on many key performance criteria for healthcare delivery systems such as resource utilization, patient satisfaction, and staff overtime. However, due to many sources of uncertainties, multiple stakeholders, and complexity of operations, OR planning is a challenging problem and still an open research area. Data resulting from increasingly widespread deployment of electronic health record (EHR) systems are starting to provide an important foundation to improve operational efficiency using data driven models at hospitals. I will describe an OR planning and scheduling system that predicts procedure duration and prescribes next-day OR plans to meet multiple criteria.
First, I will present an adjustment method based on a combination of operational, temporal, and staff related factors that improve traditional surgery duration estimates. Using two years of detailed operational data from an EHR system, we conclude that while improving estimates of surgery durations is possible, the inherent variability in such estimates remains high, necessitating caution in their use when creating OR schedules.
In the second part of the talk, I will present the daily scheduling problem of a single OR with uncertain surgery durations. Our aim is to find the optimum sequence and scheduled starting times of the surgeries to minimize weighted sum of expected patient waiting times and OR idle times. In addition to analytical results providing useful insights on the characteristics of the optimum solutions, we develop heuristics for sequencing and duration assignments and compare their performance. Our heuristics provide insights for the practitioners that do not have the resources to implement more advanced models.

Bio: Enis Kayış is an Assistant Professor of Industrial Engineering at Özyeğin University. He received his PhD from Management Science and Engineering from Stanford University in 2009 and his MS in Statistics from the same university in 2007. During 2009-2012, he had worked at Hewlett-Packard Labs as a research scientist in several projects including demand estimation, product portfolio management and pricing, healthcare operations, and forecasting. His current research interests include data-driven decision making and business analytics with applications in outsourcing, supply chain contracting, and healthcare operations. Prior to his graduate studies, he received BS in Industrial Engineering and Mathematics, both from Bogazici University.