Context-Based Pricing for Revenue Optimization with Applications to the Airline Industry

Abstract

Most airlines use dynamic pricing to optimize the price of their base economy product by maximizing the expected revenue. However, when it comes to pricing of premium products, airlines often use static price increments that are applied to the best available economy fare based on simple business rules for adjusting the price based on supply. In this paper, we present a suite of machine learning algorithms that take advantage of the rich booking session context available at the time of the booking to make pricing policy recommendations. The challenge is to accurately predict bookings for new combinations of attributes by market and segment (departure time, length of stay, advance purchase, length of haul,...) using sophisticated machine learning models while keeping the resulting pricing policy interpretable. We present an approach based on a novel path-based Mixed-Integer Programming (MIP) reformulation that can efficiently recover simple yet near-optimal pricing policies. To generate practical pricing policies, the approach accommodates a variety of real-world business requirements into the decision optimization problem. We demonstrate the efficacy of our model with extensive experiments. Finally, we present an airline case study on deriving profitable prescriptive policies for premium cabin tickets based on easily interpretable pricing rules.