2651. Estimating customer choice model and discrete dynamic pricing in the Online Bus Market
Invited abstract in session MC-59: Pricing and learning 2, stream Pricing and Revenue Management.
Monday, 12:30-14:00Room: S08 (building: 101)
Authors (first author is the speaker)
| 1. | Akshitha Suryapeta
|
| Operations Management, IIM Lucknow | |
| 2. | Suresh Jakhar
|
| Operations Management, Indian Institute of Management Lucknow |
Abstract
The Indian outstation tourism sector, valued at $66 billion, sees substantial bus travel (50% of outstation journeys). The online bus market, currently at 5% ($1.2 billion), is projected to reach 9.5% by FY25. One of the driving factors for this growth is the shift in customer behavior. Hence, our research focuses on studying customer behavior. The research objective is to evaluate the applicability of the choice model in predicting bus booking patterns and further solve discrete dynamic pricing problem. The primary hurdle is constructing an appropriate choice model, which involves various options that a customer contemplates while making their decision and identifying relevant attributes influencing their choices. To model choice behavior, the widely used Multinomial Logit model is adopted. The second challenge lies in the incomplete available data, with access only to booking records and no insight into browsing behaviors. To address this, we utilize an Expectation-Maximization (EM) method. In the first stage, we apply this estimation approach to simulated data, and then to the real bus booking data. The accuracy of our estimation is verified through calculations of asymptotic standard errors and goodness-of-fit tests, comparing the observed and expected booking counts predicted by our model. The estimates are then used to solve the discrete dynamic pricing problem of the choice-based network RM model.
Keywords
- Revenue Management and Pricing
Status: accepted
Back to the list of papers