2633. Optimizing Discrete Dynamic Pricing in the Online Bus Market: A Choice-Based Network Revenue Management Approach
Invited abstract in session TA-29: Pricing and Contract Design in Logistics & Networks, stream Pricing and Revenue Management Innovations.
Tuesday, 8:30-10:00Room: Maurice Keyworth 1.04
Authors (first author is the speaker)
| 1. | Akshitha Suryapeta
|
| Operations Management, IIM Lucknow |
Abstract
The "One Nation, One Permit" policy is expected to benefit bus operators and passengers by reducing permit costs and fostering competition, potentially lowering fares. In India's online bus market, pricing plays a crucial role in operator success. This study examines the discrete dynamic pricing problem, where only a few predetermined prices are offered per origin-destination pair. A comparison is made between leg-based booking limits using expected demand and the Choice-based Deterministic Linear Program (CDLP). A choice-based network revenue management model is considered, addressing challenges such as estimating arrival rates and choice probabilities from past booking data. This requires a discrete choice demand model and optimization of large action spaces, similar to dynamic assortment optimization in e-commerce. Since choice-based revenue management models are intractable, CDLP is used to approximate value functions. By integrating customer choice and selecting prices from a finite set, the problem is formulated as a constrained maximization task. Numerical experiments will evaluate average revenue, seats sold per leg, seats sold per fare class, and computational efficiency.
Keywords
- Revenue Management and Pricing
Status: accepted
Back to the list of papers