Flight crew rostering could be fun, unless, of course, you have to do it yourself


This story developed quite innocently during one slow weekend on the road. A late coffee at Starbucks ended up in a very interesting conversation and brought into focus a difficulty faced by flight and crew planning staff working for even the smallest airline operators.

I knew this area from my previous work, when I worked on IT consulting projects with major airlines. What got my attention on this occasion, was that my fellow coffee drinker was working for quite a small airline. An airline that size would not be on the radar of any major software vendor specializing in crew and fleet planning and scheduling. So, after some brief introductions, my coffee companion told me of his struggles with Excel and his huge white board with a lot of sticky notes on it. As he spoke, I pulled out my pen and wrote down a few numbers on the back of the napkin (I’m an economist by training, so it’s hard to resist sketching things out on the napkins).

A day in the life of an airline planner

An airline that follows a timetable is complex, but not overly so. Network optimization nicely takes care of creating traffic demand profiles, allocating fleet equipment, generating flight timetables, and creating workforce demand patterns. Output of optimal design is then used by roster planning and optimization software to match operational demands to available personnel. A single planning and scheduling optimization system can handle both passenger and freighter fleet needs. However, this particular operator, while small, had also a few all-cargo planes that could be used for unscheduled lifts to any destination. Therefore, part of his scheduling struggle was adding the dynamic component of short term scheduling in addition to the mid- (3 month) and long-term (6-12 month) rostering.

His “day-in-life” consisted of a lot of data entry into his massive Excel spreadsheets and then churning out countless “goal seek” formulas all over the worksheets. His data organization solution was to create “knowledge” spreadsheets where he would maintain more static source data describing individual employees and individual pieces of equipment (planes, but also a ground pool of trucks and assorted pieces of airside machinery).

His days were incredibly busy as he tried to juggle strategic planning while handling unpredictable events on the day of operations. In his job, the scheduler liaises closely with network planning to predict demand for crews and to ensure availability of crews to cover scheduled flights plus staff reserves to cover unplanned flights that could be added only a few hours before take-off. Not all employees are allowed to perform all tasks related to their role. Aviation is all about safety, so employees need to keep their certifications current. An employee can have between 5 and 100 certificates, each expiring at different times and at various intervals.

Keeping in mind that working periods vary for crews (this is dictated by the duration of a flight), my planner was in a constant fear of either creating a roster that would include people performing work with expired certificate(s) or not being able to find a complete crew with all (and valid) certificates necessary for the piece of equipment they had to operate. Oh yes, and also in fear of missing some labor union rules and workplace agreements applicable to each employee. Frequent audits by regulator and unions required my new friend to be ready with all information at his fingertips and demonstrate compliance in all aspects of labor practices.

Further headaches

As if operational planning was not stressful enough, there was also cost to consider. Apparently, my companion had some very demanding, cost-related KPIs. That is quite understandable in smaller airlines. Every OPEX dollar saved has impact on the bottom line and can mean the difference between operating for another year and folding up. In fact, the chief complaint about his KPIs was that they were impossible to meet consistently.

As is often the case, his salary depended on meeting these KPIs, and yet the HR department governed employee-related costs (salary, benefits, travel allowances, etc.), the administration department chose the crews’ hotel/accommodation and ground transport rates, and Procurement decided on rates for fuel and ad-hoc maintenance services costs. He had no influence over the frequent changes that wreaked havoc on his cost-effective roster calculations and plans. Thus the headache and frustration of my companion. And frequent revisions to his massive spreadsheets.

In the face of all this, I asked how he motivates himself and his planning team. As it happens, it is a struggle. It was difficult to keep the team members interested and challenged, constantly under pressure from other departments and employees, not to mention those impossible-to-meet KPIs which determined their salaries. Whenever one member of his team left, they’d quickly hire a junior analyst expected to acquire skills and expertise in a very short time. Then another person would burn out and the cycle would repeat itself.

This very interesting conversation and great lessons came to me at the cost of 4 café lattes. Never underestimate an opportunity to meet interesting people and striking friendly conversations. You will be rewarded in many ways.

The million-dollar napkin

And the numbers I jotted down on the napkin? Overtime that could be minimized: $450,000-750,000; cost of mistimed pilot training: $250,000; productivity efficiency: 76% due to mistimed training & certification of new hires but also unexpected airport closures due to weather; wrong crew assignments resulting in staff recalls and overtime: 7% at the cost of $250,000-$500,000. We speculated that those improvements would also allow his small team to meet the corporate KPIs and allow them to earn their performance bonuses more often.

Many of those costs could be eliminated or severely reduced, if only the planner had a better IT solution to optimize planning & scheduling. A better IT tool would free up his time to handle the really exceptional events requiring immediate human intervention. A planning system with built-in optimization and configured to solve his problem at the basic level, would easily pay for itself in less than 1 year. For a small airline, it could be a matter of extra time to survive and compete against the majors.

If you found this topic interesting, leave a comment below. I look forward to hearing what you think.

This post was previously published on LinkedIn. For more posts like this, follow Kris Kosmala on LinkedIn.