How will flight schools schedule in the future? According to Michael Rasmussen, Product Manager at FlightLogger, suggested scheduling will build the schedule for them: the scheduler picks the students and days, and FlightLogger creates a schedule that follows every rule the school has set, prioritizing the students who are furthest behind. It is rule-based rather than AI guesswork, and a human always reviews and approves the result. In the second episode of The Climb Briefing, he talks with FlightLogger CEO Kasper Valgren about where scheduling is heading.
Michael Rasmussen is Product Manager at FlightLogger and owns the product roadmap. Before moving into product, he spent years on FlightLogger's client success team onboarding flight schools of every size. This is the second part of a short series; in part one, he covered combined syllabus, prerequisites and aircraft model requirements.
Michael describes a large flight school scheduling hundreds of students a day across many aircraft, but only able to publish the schedule about 22–24 hours ahead. Students didn't know when they were flying tomorrow, instructors didn't know whether they were working, and administration had no view of how the operation would run over the next few days. A lot of resources went into the last hours before each day, and Friday afternoons got very long.
It isn't only a large-school problem. Medium-sized schools see it too, and the difficulty grows quickly with size: the larger you get, the more of a headache scheduling becomes.
Every booking has to respect many rules at once: which lessons a student has completed, which instructors are certified on which aircraft models, availability, maintenance and more. Checking all of that by hand takes a lot of time, which is why FlightLogger has been building those rules into the system with prerequisites and aircraft model requirements.
Lost production time is the obvious cost, and in aviation production time is expensive. But the bigger cost may be the lack of insight into the coming week. Instructors don't know how many hours they will get, and flight instructors above all want to fly and teach. Scheduling further ahead gives the whole academy alignment: everyone knows when they are working and can plan their private lives too.
Today, most FlightLogger clients schedule flight training about four days ahead. Michael's ambition is to raise that to at least a week and a half.
They act as a filter, removing every combination that shouldn't be in a schedule. Suggested scheduling enforces the same rules: which aircraft models can be used for a specific program or session, and which instructors are certified to fly them. Optional preferences are added on top, such as a student's preferred instructors (tier one first, then tier two) or a preferred aircraft model, for example because of the package the student bought.
A system needs to know the rules before it can suggest anything. Some rules apply to every school; others are optional and can be left out.
Suggested scheduling creates a schedule based on all the variables a school has put into FlightLogger, the same things schedulers check manually today. The scheduler picks the students, the days or number of days, and the program if a student is enrolled in several, then clicks to create the schedule.
The first and most important variable is student progression. Students who are furthest behind their planned progression are prioritized first, so that when instructors or aircraft are scarce, the students who most need flying get it.
Version one covers flight training and looks at prerequisites, aircraft model requirements and availability, with some fixed rules to start with, such as keeping an aircraft that already has bookings flying continuously unless it is unavailable, for example for maintenance.
Version two is planned to add predictive warnings: avoiding aircraft that are predicted to need maintenance, and taking flight and duty time limits into account so no one is scheduled into a violation. Ground training is planned later, which is where combined syllabus helps, because prerequisites can link ground lessons to flight lessons in one program. After that, development will follow feedback from schools using the feature, with more settings that schools can switch on or off.
Yes. Suggested scheduling is deterministic: it enforces the rules the academy has set, with no AI judging or guessing. Because scheduling is safety-critical, it will not run automatically without people. A human always reviews and approves the schedule.
In the future, the system will also flag risks and suggest alternatives, for example when an aircraft needs unplanned maintenance or an instructor with three bookings calls in sick. It flags and suggests; people decide.
Michael sees AI in FlightLogger as an assistant that finds patterns in operational data and makes recommendations, for example bookings that are repeatedly created and cancelled, instructors grading the same exercise differently from the norm, or students who use more time than reserved and perform below their class, which could flag a student at risk. In every case, the system flags and a human judges what to do.
Schedulers will review the suggestions, apply professional judgment, check that the setup is correct, and handle edge cases. One example is operating hours for night flights, which have to change through the year, especially in the Nordics. Another is adding buffer or longer briefing time for new students.
In Michael's words, schedulers will spend less time building the schedule and more time approving it.
Suggested scheduling will be shown publicly for the first time on November 4 at EATS in Portugal, with a reveal and client event where FlightLogger will gather first-hand feedback from schools.
Suggested scheduling is a FlightLogger feature that creates a flight training schedule from the rules a school has set, including prerequisites, aircraft model requirements, availability and preferences, while prioritizing students who are furthest behind. A scheduler reviews and approves the result.
No. It is deterministic: it enforces the rules the academy has set, with no AI judging or guessing. AI may later assist by flagging patterns and risks, but a human always makes the decisions.
Most FlightLogger clients schedule flight training about four days ahead. Suggested scheduling aims to help schools plan at least a week and a half ahead.
Set up prerequisites, start using combined syllabus if you run ground and flight training, set up aircraft model requirements, and follow release announcements for new input fields.
Michael Rasmussen is Product Manager at FlightLogger and owns its product roadmap. He previously onboarded flight schools of every size on FlightLogger's client success team.
In the next episode, Kasper and Michael talk about how training and maintenance stay in sync, and about FlightLogger's new maintenance hub: what it is and where it's heading.
Watch or listen to the full conversation with Michael Rasmussen, Product Manager at FlightLogger, on The Climb. Want to get your school ready for smarter scheduling? Talk to your account manager or book a demo.
The Climb — conversations from the world's most ambitious flight schools.