Real-Life OR Scheduling Data at UZGent (Mar-Aug 2021)
Description
This dataset addresses a specific issue in operating room scheduling, integrating the surgical case planning problem with the short-term surgeon scheduling problem within a one-to-two-week planning horizon. The problem is inspired by real-life situations at Ghent University Hospital (UZGent) in Belgium. The main objective is to efficiently sequence surgeons across different rooms and days, planning their surgical lists to maximize the daily priority attained. The dataset consists of two parts: Instances: This part represents real-life environments and includes a case study conducted at UZGent. To gather the necessary instance information, we conducted interviews with key personnel responsible for composing OR schedules and examined the OR logbook from March 2021. This resulted in the collection of data for 24 instances, with each instance representing the activities of a single week. The total number of patients ranges from 50 to 80 for medium-sized instances (e.g., G00) and 100 to 160 for large-sized instances (e.g., G01). Additionally, this part includes higher-level data in the MSS (Master Surgery Schedule) folder for each instance group. BCnP: This part contains the final solutions reported by the proposed Branch-Cut-and-Price (BCnP) algorithm. BCnP is divided into two subfolders, each containing 24 solutions in XML format, indicated by ["Ins_"]+[instance number][...].XML. The findings of this research are detailed in our recent study titled "A dedicated branch-price-and-cut algorithm for surgical case planning and surgeon scheduling," which is currently awaiting peer review. Further information, including the DOI, will be provided upon acceptance. The information regarding the MSS is already described in our study: Akbarzadeh, B., & Maenhout, B. (2024). A study on policy decisions to embed flexibility for reactive recovery in the planning and scheduling process in operating rooms. Omega, 103061. Notes: All files are in HTML format, organized in an object-oriented manner using related tag objects. Each file starts with a "description" tag that provides a preview of its contents.