293. Stochastic surgery scheduling with downstream capacity constraints in the presence of side data
Contributed abstract in session HB-5: Operating Room Scheduling /2, stream Regular talks.
Thursday, 11:00-12:30Room: Room S6
Authors (first author is the speaker)
| 1. | Ricardo Otero
|
| KU Leuven | |
| 2. | Erik Demeulemeester
|
| KBI, KU Leuven |
Abstract
This study addresses the stochastic Advance Scheduling Problem with Downstream Capacity Constraints (ASPDCC), particularly focusing on scenarios where decision-makers have auxiliary data at their disposal. Our approach initiates modeling the stochastic ASPDCC as a two-stage stochastic program, employing a sample average approximation (SAA). We propose a data-driven framework that integrates a machine learning algorithm (decision trees) to estimate the conditional probability distribution for the surgery durations and the length of stay (LoS), in the presence of side data. Then, these estimated probability distributions are integrated into the SAA model. To improve the efficiency of our methodology, we implement a benders-branch-and-cut algorithm to solve the SAA model. Computational experiments conducted using data from a reference hospital in Colombia demonstrate that our algorithm generates schedules that surpass those generated by the conventional method of fitting a probability distribution for each surgery specialty.
Keywords
- Operating room planning and scheduling
- Analytics
- Optimization algorithms
Status: accepted
Back to the list of papers