Customer heterogeneity in revenue management for railway services

Choice models based on random utility theory are being used in revenue management problems because of their ability to deal with customer heterogeneity and preferences over a set of multiple products. In this article, both discrete and random mixture of logit are proposed to model ticket purchase ti...

Full description

Saved in:
Bibliographic Details
Published in:Journal of revenue and pricing management 2015-02, Vol.14 (1), p.28-49
Main Authors: Hetrakul, Pratt, Cirillo, Cinzia
Format: Article
Language:English
Subjects:
Air fares
Air travel
Airline industry
Airlines
Business and Management
Customers
Decision analysis
Linear programming
Optimization
Passengers
Railroad transportation
Research Article
Revenue management
Sociodemographics
Studies
Travel
Utility functions
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Choice models based on random utility theory are being used in revenue management problems because of their ability to deal with customer heterogeneity and preferences over a set of multiple products. In this article, both discrete and random mixture of logit are proposed to model ticket purchase timing over a finite sale horizon; in particular, parametric and non-parametric mixed logit are formulated and estimated. The analysis relative to intercity railway trips is performed on real data extracted from internet booking records, which contain very limited information that can be used for customers’ segmentation. The parameter estimates are then integrated into a nonlinear optimization framework, and fare strategies and seat allocations are derived. Results show that up to 20 per cent increase in revenue can be obtained when methodologies based on customer heterogeneity are considered. Eventually, this study demonstrates that advanced choice models can be efficiently estimated on real data with severe limitations, and operational optimization methods can be based on individual choice behavior.
ISSN:1476-6930
1477-657X
DOI:10.1057/rpm.2014.27