Greedy Heuristics and Simulated Annealing for Radiotherapy Scheduling
A new study on arXiv (2607.22539) tackles the Radiotherapy Scheduling Problem (RTSP), which optimizes patient treatment schedules to improve clinical outcomes. The daily batch approach using Integer Linear Programming is effective but computationally intensive. Researchers developed two novel greedy heuristics—RTSP First Fit and RTSP Best Fit—and combined them with Simulated Annealing (SA) to reduce time and memory demands. These methods were tested on a public dataset against an integer linear program, showing promising results for efficient scheduling.
Key facts
- arXiv paper 2607.22539 addresses Radiotherapy Scheduling Problem (RTSP)
- Daily batch approach with Integer Linear Programming is effective but resource-heavy
- Two greedy heuristics developed: RTSP First Fit and RTSP Best Fit
- Heuristics used as constructive heuristics for Simulated Annealing
- Methods evaluated on publicly available dataset
- Comparison made against integer linear program
Entities
Institutions
- arXiv