How Southwest Airlines used art and science to increase demand forecast accuracy

Southwest Airlines is benefitting from increased accuracy when predicting airline demand, with the help of machine learning and customer choice modeling.

The pandemic has required new scheduling techniques to be developed, with airlines no longer being able to rely primarily on historical schedules. In 2020, Southwest took on the challenge to identify and increase scheduling and revenue opportunities in a world that has changed enormously.

Amadeus SkyCAST is a key element in Southwest’s network planning decision-making process due to the solution’s high forecast accuracy and flexibility. The forecasting tool helps airlines find that perfect balance between schedule optimization and the highest revenue potential – no small feat for airlines with thousands of different flight combinations. Airline planning requires a careful balance between art and science to be successful.

Southwest Airlines can assess numerous 'what-if' scenarios in a matter of hours - allowing new growth opportunities to be identified - even during an uncertain and volatile period. Accurate forecasting has helped to limit the financial impact of the crisis while maintaining the high level of serice that Southwest customers expect.

The importance of customer choice modeling and machine learning for accurate forecasting
Special calibrations during COVID-19
Market size estimations

Available in other languages

EN DE

First Name This field is required

Last Name This field is required

Work e-mail This field is required

Country This field is required

Company name This field is required

What is your job level? This field is required

What job function do you work in? This field is required

Consent to receive further communication by Amadeus

By submitting this form, I confirm that I have read and understand Amadeus' Privacy Notice. This field is required

How Southwest Airlines used art and science to increase demand forecast accuracy

Southwest Airlines is benefitting from increased accuracy when predicting airline demand, with the help of machine learning and customer choice modeling.

This case study will explore:

There is a problem