Posted in | News | Machine-Vision

Artificial Intelligence Helps Design Customized Pricing Model for Airline Customers

Artificial intelligence (AI) is being used by researchers to help the aviation sector to price their supplementary services like seat reservations and checked bags in a way that proves beneficial to the budget and privacy of customers and also to the bottom line of the airline industry.

Industrial and enterprise systems engineering professor Lavanya Marla and collaborators used artificial intelligence to design a customized pricing model for airline customers. (Image credit: L. Brian Stauffer)

In 2008, when airlines started to unbundle the price of their flights and supplementary services, several customers saw it as a ploy to quote a low base fee and then include extra to increase profits, stated the scientists.

In a recent study, the investigators are used unbundling to meet the requirements of customers and, at the same time, increased the revenue of airlines with individualized and smart pricing models provided in real time as the customers shop.

The study results will be presented at the 2019 Conference on Knowledge Discovery and Data Mining which will be held in Anchorage, Alaska on August 6th, 2019.

According to the researchers, airlines operate on extremely slim margins. While airlines earn a significant portion of their revenue on supplementary purchases, unbundling can offer cost-saving prospects to customers too.

If such measures are implemented, then customers would no longer have to pay for things they do not require. Moreover, discounts provided to customers who are likely to pass on the extras can help in changing a “no sale” into a purchase.

Most airlines offer every customer the same price for a checked bag. However, not every customer has the same travel and budget needs. With AI, we can use information gathered while they shop to predict a price point at which they will be comfortable.

Lavanya Marla, Study Co-Author and Professor, Department of Industrial and Enterprise Systems Engineering, University of Illinois

To accomplish this feat, the pricing models employ a combination of AI methods—that is, deep neural networks and machine learning—to observe and assign a level of demand on the flight preferences of individual customers, stated the researchers.

Numerous price factors, like the timing of travel, destination, flight origin, and the duration of a trip, are considered by the pricing models to assign a value on demand.

For example, a customer who is traveling for a few days may not be motivated to pay for a checked bag,” Marla stated. “But, if you discount it to them at the right price—where convenience outweighs cost—you can complete that sales conversion. That is good for the customer and good for the airline.”

In the analysis, the research team at the University of Illinois and Deepair Solutions teamed up with a European airline for a period of about six months to collect data and examine their models. While customers were shopping, they logged into a pricing page in which a predetermined percentage of customers are given discounts on supplement services.

We started by offering the AI-modeled discounts to 5% of the customers who logged in. The airline then allowed us to adjust this percentage, as well as to experiment with various AI techniques used in our models, to obtain a robust data set.

Kartik Yellepeddi, Study Co-Author and Co-Founder, Deepair Solutions

The airline started to observe a slight increase in the ancillary revenue as well as in ancillary sales conversions for each customer, and enabled the team to provide discounts to all of the customers who logged into a pricing page.

Because of the unique nature of personalized pricing, we built a high level of equity and privacy into our models,” Yellepeddi stated. “There is a maximum price not to be exceeded, and we do not track customer demographics information like income, race, gender, etc., nor do we track a single customer during multiple visits to a sale site. Each repeat visit is viewed as a separate customer.”

According to the study, with the uptick seen in ancillary revenue and ancillary sales conversions for each offer—up by 25% and 17%, respectively, the researchers believe that AI can help the aviation sector to migrate from the notion of the “average customer” and customize their offers to “individual travelers.”

In recent years, the airline industry has felt that it has been losing touch with its customer base,” Marla stated. “The industry is eager to find new ways to meet customer needs and to retain customer loyalty.”

Deepair Solutions, which is headquartered in London and has an office in Dallas, is an artificial intelligence company that serves the travel sector.


Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type