3077. A Learning-Enhanced Matheuristic Approach for Optimizing Recreational Activity Planning in Medical Tourism
Invited abstract in session TD-15: Methodological developments in public transportation and medical tourism, stream Combinatorial Optimization.
Tuesday, 14:30-16:00Room: Esther Simpson 1.08
Authors (first author is the speaker)
| 1. | Burak Pac
|
| Industrial Engineering, Gebze Technical University |
Abstract
In medical tourism, the integration of medical treatments with recreational activities requires careful planning to ensure tourist satisfaction and operational efficiency. The Recreational Activity Planning Problem involves assigning recreational activities to medical tourists while respecting their medical schedules, budget constraints, and activity preferences. Activity durations and capacities must be adhered to, with the aim of maximizing a weighted sum of profit from touristic activities and tourist satisfaction. The problem is formulated as an Integer Programming model, and its NP-hardness is established through a reduction to the Multiple Knapsack Problem with Assignment Restrictions. A matheuristic approach based on Variable Neighborhood Search is proposed, leveraging diverse neighborhood structures to efficiently explore alternative solutions with a broad scope. A key aspect in the methodology devised is the use of neural networks to predict at incumbent solutions the improvement potential of different neighborhood structures. These predictions enhance the prioritization of neighborhood structures to be searched, significantly improving the efficiency of the algorithm. Experimental results highlight the added value of neural network predictions in guiding the search process, enabling the solution of large-scale instances that are unsolvable within reasonable time by commercial solvers.
Keywords
- Metaheuristics
- Machine Learning
- Health Care
Status: accepted
Back to the list of papers