2176. From Booking to Service: Adaptive and Learning-Driven Pricing for Hotel Revenue Growth
Invited abstract in session MA-29: Machine learning in pricing and revenue management, stream Pricing and Revenue Management Innovations.
Monday, 8:30-10:00Room: Maurice Keyworth 1.04
Authors (first author is the speaker)
| 1. | Aysajan Eziz
|
| Ivey Business School, Western University |
Abstract
Dynamic pricing for hotel rooms involving multi-night stays presents unique challenges due to complex capacity interactions across consecutive nights. Traditional revenue management methods that consider single-night stays fail to capture these intricacies, resulting in suboptimal pricing decisions. This research introduces a dynamic pricing approach specifically designed to address multi-night stay complexities. We propose two innovative and complementary solutions: a stochastic approximation algorithm (SAA), which uses simulation-driven learning to optimize prices without exhaustive enumeration, and a quadratic programming (QP) approximation that aggregates demand for computational efficiency. Our methodology effectively balances practical implementation with theoretical rigor, yielding near-optimal revenue performance. Extensive numerical experiments demonstrate that our SAA solution consistently achieves near-optimal revenue outcomes, significantly outperforms traditional benchmarks, and reduces computational complexity by approximately 90%. This research provides revenue managers practical tools for capturing previously untapped revenue potential in real-time hotel operations.
Keywords
- Revenue Management and Pricing
- Algorithms
Status: accepted
Back to the list of papers