Reducing rescheduling events in operating rooms by applying pattern recognition techniques

The operating room department is one of the most important departments in hospitals. Managing it efficiently is of great concern. A part of this management consists of properly scheduling surgeries. In this paper we aim at examining a Dutch hospital’s current scheduling policy and investigating the amount of rework that is done during the scheduling process. We describe a method to gain insight into types of surgeries that are likely to be rescheduled. Using decision trees and association rule mining as pattern recognition techniques, we discover patterns that indicate a high probability of rescheduling. They are studied in more detail in order to explain why rework occurs in the scheduling process. A validation study is also performed to examine how strong the results are.