213. A hybrid solution approach for the Integrated Healthcare Timetabling Competition 2024
Invited abstract in session TD-12: Scheduling in Healthcare I, stream Health Care Management.
Thursday, 14:30-16:00Room: H10
Authors (first author is the speaker)
| 1. | Daniela Guericke
|
| Department of Industrial Engineering and Business Information Systems, University of Twente | |
| 2. | Matthias Walter
|
| Faculty EEMCS, University of Twente | |
| 3. | Rolf van der Hulst
|
| University of Twente, Department of Mathematics | |
| 4. | Asal Karimpour
|
| University of Twente | |
| 5. | Ieke Schrader
|
| University of Twente |
Abstract
The Integrated Healthcare Timetabling Competition (IHTC) 2024 posed a planning problem relevant to the hospital sector. The IHTC considers four coupled scheduling problems in an integrated problem setting, since patient flow through the hospital creates dependencies between them. The scheduling problems are: 1) deciding the patients’ admission day, 2) scheduling the patients’ surgeries in operating theatres, 3) assigning patients to rooms, and 4) assigning nurses to rooms to care for the patients. These decisions must satisfy several hard and soft constraints. The objective function of the IHTC consists of weighted penalties for the eight soft constraints.
Since mixed-integer programming models for the entire problem are prohibitively large, we propose a hybrid solution approach utilizing mixed-integer programming, constraint programming, and simulated annealing. Our approach decomposes the planning into different phases and explores feasible solutions from previous phases in parallel to find solutions for the overall planning problem. During the presentation, we will discuss the relevant models and design decisions, as well as results from the competition instances and the insights we gained into future research directions while solving them. Our approach, which proved robust across different instance settings, ultimately placed among the finalists of the competition.
Keywords
- Health Care
- Scheduling
Status: accepted
Back to the list of papers