Arrow DownArrow ForwardChevron DownDownload facebookGroup 2 Copy 4Created with Sketch. linkedinCombined-Shape mailGroup 4Created with Sketch. ShapeCreated with Sketch. twitteryoutube

Frontier decision-making is here

Find out more

Case study

Saved £10 million a year for a leading airline by forecasting standby crew

We built a machine learning model for a short-haul airline that would accurately predict the amount of spare staffing capacity it would need to roster a full cabin crew for flights, saving £10 million a year. 


A leading short-haul airline operating in more than 30 countries.


Each day, approximately 4,000 crew report for duty and operate between two and five flights. Each flight requires a certain number of crew. Inevitably, disruption happens that affects how many crew are available, including crew sickness, resignations, or day-of-operation disruption (e.g. weather, airport delays). The airline must predict the level of disruption and allocate enough standby crew to ensure that its planes can fly.

The airline’s previous solution was to allocate a fixed proportion of crew on standby (typically a fifth more than were needed for each flight). Because of the high operational cost and financial impact of not having enough staff to fly a plane, the airline was conservative in its estimate of how many staff needed to be on standby. In reality, few flights were cancelled because of lack of crew, meaning that this was an inefficient use of resources.

The airline wanted to have a more accurate gauge of how many staff it needed to have on standby. Faculty was charged with identifying the significant factors that affect standby demand and creating an improved method for crew rostering.


We worked with the airline’s analytics team and human resources department, using staffing data to build a machine learning model that more accurately predicted the amount of spare staffing capacity it would need to roster a full cabin crew for flights.

Once the model was developed, tested and discussed with staff, a dashboard tool was built using the model for staff schedulers to help the airline make more accurate estimates of the standby staff it would need on any day and in any location. Our data analysis also revealed which months and which locations being served by the airline led to higher standby requirements.



The dynamic model reduced standby staffing levels from a constant 21% to an average of 14%. As a result, the headcount was dynamically adjusted to suit the time of year, location, and other affecting factors. Needing fewer staff on standby has resulted in the airline saving more than £10 million a year.

The insight into the months and locations with particularly high standby requirements has provided the airline with the evidence to conduct further investigations into the causes of staff shortages. It is aiming to develop interventions that will further reduce the crew standby requirements.


million savings
in a year

To find out more about what Faculty can do
for you and your organisation, get in touch.