By Richard Ratliff, Executive Scientist at Sabre Labs, and Sergey Shebalov, VP and Head of Research at Sabre Labs
In Part 1 of this three-part series looking at advanced Artificial Intelligence (AI) and Machine Learning (ML) applications for airlines, we explored how airlines are using AI/ML today, along with the first of five new AI/ML applications: dynamic pricing and availability of airfares.
In Part 2, we considered three more new AI/ML applications: air ancillary dynamic pricing, experimentation engines and market size forecasting.
In this final installment, we describe new tools for handling disrupted passenger re-accommodation and conclude with a brief overview of how cloud-based computing and improved machine learning operations (MLOps) are making AI/ML more useful and powerful than ever.
Disrupted passenger re-accommodation
Some of the most common disruptions airline passengers face are flight cancellations and delays, but other kinds of disruptions exist. Sabre’s AI/ML models for reaccommodation during irregular operations such as unplanned flight schedule changes, cancellations, etc. (called IROPS) can help in such scenarios by providing new itineraries accounting for global availability, passenger prioritization and airline practices.
These AI/ML models can also be used as a ‘what-if’ tool where airline analysts evaluate the effect of a particular disruption (say, a change of equipment to a smaller capacity plane). Based on the solution, they can decide if this is an acceptable action to perform. If the tool is used, and results are applied, passengers are automatically guaranteed a ticket in the new itinerary as the rebooking process takes care of this.
At the core of this IROPS application, a sophisticated model is used to generate alternative paths and decide which paths should be assigned to each booking based on availability and prioritization (using AI/ML to classify passenger acceptance). Whenever the IROPS application is run, and passengers are informed of their new itineraries, those passengers need to decide if the new trip is something they are still interested in taking or they prefer rescheduling or completely canceling.
A businesswoman with an important meeting Monday morning would cancel rather than take an alternative itinerary arriving Monday lunch time. Someone on a week-long vacation in Italy would likely accept to be rebooked to arrive half a day later. Experienced analysts can recognize some of these patterns and offer only those reaccommodation options that are likely to satisfy travelers. To help airline analysts further, we developed a machine learning model that helps evaluate if the replacement itinerary would be accepted or not.
What does the future hold?
As airlines move to more granular control over pricing, offer a growing variety of ancillaries (which can differ by market, volume and weight), and apply more real-time experimentation – the parameters that optimization algorithms have to consider will also grow exponentially. Thankfully, with costs of computing, storage and networking going down and with increased power of modern AI/ML algorithms, industry participants have been able to tackle bigger problems over larger domains. This has been possible through investments in sophisticated data and analytics platforms and the applications that can leverage them.
For example, to help support our customers, Sabre is building new AI/ML applications that are natively cloud-based with a microservices-based architecture (see Sabre Travel AI™). Each module, called a microservice, is independent from the other components and is dedicated to a particular task. The advantage of that approach is a very agile development process and ability to leverage modules across different solutions. We build a microservice once and then can use it again and again in different applications. Since it’s a modular component, it is easily scalable, has better fault isolation, and is (somewhat) programming language and operating system agnostic – all of which leads to faster time to market.
The use of microservices is of particular benefit for AI/ML models due to the emergence of another software practice, machine learning operations (MLOps). Without going into technical details, the business benefit of MLOps is that it allows for AI/ML models to be automatically retrained, recalibrated and retested as new data becomes available, and these updated models are automatically deployed into the production environment without the need for manual intervention. MLOps capabilities allow Sabre’s new AI/ML models to be automatically updated as new data is collected, resulting in reduced cycle time and improved AI/ML model accuracy.
We are also working to provide the option for our customers to build their own (or heavily customized) applications that make web-based calls to Sabre Travel AI™ microservices via well-defined application programming interfaces (APIs). Such API-based calls are not new for Sabre; in fact, that’s how most airlines and travel agencies access Sabre GDS and PSS content today. Our objective is to extend those API-based capabilities to include AI/ML model components, allowing Sabre to serve as a value-added technology partner for airlines using our products as well as accelerating and simplifying airlines’ in-house applications development.
About the Authors
Richard Ratliff, Executive Scientist at Sabre Labs, has held roles of increasing responsibility in IT at Sabre for over 30 years with a focus on travel retailing and revenue management.
Sergey Shebalov, VP and Head of Research at Sabre, leads a team responsible for the development and implementation of decision support systems in the travel industry. Sergey holds a PhD in math from the University of Illinois and has two decades of experience in airline and agency IT.
Keep the lines of communication open
Be the first to know when we release new information on emerging technology, upgrading of existing systems, or updates to our point of view on the future of travel retail.