Since Volantio launched its Yana™ platform in 2017, we have been leveraging machine learning to better understand the primary drivers of passengers’ willingness to change their travel plans, and use these insights to help our partners better guide how they approach passenger reaccommodation for incremental revenue, IROPs, and denied boarding purposes.
Machine learning feature influence (normalized to 1)
Key insights learned from our initial analysis:
Monetary compensation was NOT the primary driver of whether or not a customer accepts an offer to move to an alternative flight
The leading predictor in this data set was the number of passengers on the PNR, and the optimal number was 2 passengers
Other factors which had a stronger influence on predicting acceptance by passengers of offers to switch flights included the time difference between the original and alternative flights, and the amount of advance notice provided to passengers about the offer to switch
Key implications for airlines:
The ability to quickly move passengers between flights in order to react to revenue opportunities or operational challenges is critical to airlines today
Traditionally, airlines have focused on monetary compensation as the primary lever to incentivize passengers to change their travel plans
A revised approach, leveraging machine learning to contact the passengers with the highest likelihood of acceptance, could reduce unnecessary passenger contact, improve acceptance rates, and lead to improved operational and profitability outcomes for carriers
Finally, it’s important to note that the above results will differ from airline to airline, and will change over time. However, they provide a useful starting point to think about improved passenger interactions.