Skip to the content

Know which spare aircraft pay for themselves in summer 2027.

SmartFleetOptim is a stress test for your spare fleet. It flies your airline through thousands of simulated seasons, groundings drawn day by day, with and without each spare aircraft, and gives each one’s break-even.

Your airline’s stress test is ready

Built from your airline’s public data. Free, no login.

Committee paper · Demo carrier, an invented airline · Summer 2027, June to August

Hold one long-range narrowbody spare at Hub. Not a second.

Chosen before the summer is known: −EUR 659k if calm, +EUR 3.2M if hard.

In a typical season, it pays as long as the rotation taken off the schedule to hold it earns less than EUR 64k a day, after the cost of flying it.

Typical season, average vs holding none	EUR M	EUR k a day
Avoided: cancellations from 144 to 72, long delays, minus ferry flights	5.9	64
Given up: what the rotation earns, 92 days	4.7	51
Net, ahead in 71 of 100 summers drawn	1.2	13

Overstates the spare, assumed to reach every grounding.

Break-even (dashed: hard season) and what the rotation earns, EUR k a day

Long-range narrowbody, HubPays

pays below 64

51 we assumed86 in a hard season

A second spareDoes not pay

Long-range narrowbody, North

Widebody, Hub

0100

What your rotation earns: EURk a day

Calm: typical aircraft-days lost ×0.6, failures unclustered.

Hard: ×1.6 those days, groundings ×1.25 longer, more clustered. Up to EUR 909k of the +EUR 3.2M assumes overnight ferries.

Our picks, not percentiles.

Before the committee, ask

Operations control: The share of uncovered flights you cancel; ours, 85 to 100%. A quarter lower, holding none does better.
Network planning: Which rotation comes off, what it earns.

6,000 simulated summers · run 07563456c5ce · smartfleetoptim.com

From the run behind an invented airline’s page and memo; yours are built the same way.

This paper as a PDF The full memo (PDF)The method

How to read this paper

An invented airline’s committee paper, read with you, one question at a time.

Which spare do we hold, and at which base?
The decision, in one line
What the committee signs: a type, a base, and where to stop. Every plan met the same simulated summers, so the gap between two plans is the spare’s, not luck.
On the paper: the title.
What does a spare cost us, against what it saves?
Where the trade-off breaks even
Holding a spare costs a rotation; holding none, cancellations and long delays: EUR 64k a day in a typical season. If the rotation earns less, after the cost of flying it, the spare pays.
On the paper: the break-even in blue, ending its first paragraph.
Does it pay at our numbers, and by how much?
Break-even against what it earns
The bar runs to the break-even, EUR 64k a day; the dot is what we assumed the rotation earns, EUR 51k. Bar past the dot, the spare pays: by EUR 13k a day on average.
On the paper: the yardstick’s first row, its dot and its bar’s end.
Why not a second spare, say a widebody?
The next spare does not pay
A second long-range narrowbody, at North, avoids EUR 31k a day of cancellations and long delays; its rotation earns 51k. A widebody at Hub, in addition: 49k against 96k. Neither bar reaches its dot.
On the paper: the two rows under “A second spare”.
What is it worth over the summer, in euros?
The money, averaged over summers
Averaged over typical summers: cancelled flights from 144 to 72, fewer long delays, EUR 5.9M avoided net of ferry flights; EUR 4.7M of rotation given up. Net EUR 1.2M, ahead in 71 of 100 summers.
On the paper: the ledger’s last line, the net.
And in a bad summer? And in a quiet one?
Three kinds of summer, one plan
Across a hard season and a calm one, the figures change and the title stays: on 6,000 simulated summers of each kind, calm and hard our picks, no plan has a smaller worst shortfall.
On the paper: the line under the title.
What could overturn this, and who checks it?
What to check before the committee
Operations control knows what share of flights left without an aircraft you cancel; we assumed 85 to 100%. At three quarters of that, holding none does better. Network planning knows what the rotation earns.
On the paper: operations control’s ask, at its foot.

One thing flatters the spare: the run lets it reach every grounding. Yours is built the same way, from your airline’s public data.

How it works

We build your page.
From public data: your fleet, your bases, the shape of your summer schedule. You send nothing.
You correct it.
Every assumption is one line: what a day of flying earns, how often each sub-fleet is grounded, the spares you hold. Change any number and run it again; what you enter stays private.
Your committee gets a memo.
One printable paper: the plan to defend, what each spare avoids, and the break-even below which it pays.

Three weeks to reproduce your cancellations. If we don’t, you don’t pay.

Before you rely on it, we replay your last twelve months on your own records: your technical cancellations and ours, month by month, on one chart. The test is written in the order, before you send any data.

Then the decision, on your own costs and records. Your data refine exactly this decision.

Pierre Chambet

Data Scientist in Air France's Network Program (2024–2025), founder of SmartFleetOptim (April 2026)

I build each study myself and answer every correction you send.

LinkedIn pierre@smartfleetoptim.com

Questions

What data do you need?

None for the page: it is built on your public fleet, and you correct it yourself. For the replay: twelve months of AOG days and technical cancellations by sub-fleet, your lines by base, the spares you hold and how you recover without one. For the decision, your own costs, entered by you. Nothing to install. Your data refine exactly this decision.

What does it not model yet?

An aircraft stuck at a remote outstation: the run covers every grounding as if a spare could reach it, so it overstates what a spare avoids and gives a break-even that is too high. A spare engine, or an engine lease, in place of a spare aircraft. Crews as a roster: finding a standby crew is a probability for the day. The method.

Is our data safe?

It comes under an NDA, yours or ours, signed before you send anything. Aggregated, no personal data, no tail numbers unless the study needs them. Kept on our server in the European Union (Hetzner, Helsinki) and the founder’s encrypted computer, on no third-party service without your written consent. Used for your study only, never pooled, never used to train a model, and deleted within thirty days of the study’s end, the deletion confirmed in writing.

What does it cost?

Nothing to start: the page on your public fleet is free. The replay and the decision on your own data are fixed-price studies, written in the order before you send anything; the replay is invoiced only if it reproduces your cancellations. Re-runs on the next seasons are proposed at delivery.

Start on your public fleet. Nothing to send.

Open the page built on your airline, correct any number and run it again.

Your airline’s stress test is ready