The launch window does not move.
JAXA's MMX mission to the moons of Mars is due to lift off on the evening of 19 October, on a date set by orbital mechanics rather than a project plan. Miss the window and the next one is roughly 26 months away. Most of our installation work runs on the same rule, with smaller numbers.
At 20:41 UK time on 19 October, if the weather holds, an H3 rocket is due to leave the Tanegashima Space Center carrying JAXA's Martian Moons eXploration spacecraft, MMX. The plan is to reach Mars orbit, land on its moon Phobos in 2027, scoop at least 10 grams of surface material and bring it back to a landing site in Australia in 2031. A small rover called Idefix, built by the French and German space agencies CNES and DLR, rides along.
The interesting date is not the 19th. It is 7 November. JAXA has published a backup period running from 21 October to 7 November, and after that the door closes. Earth and Mars only line up for an efficient transfer about once every 26 months. A rocket that is ready on 8 November is a rocket that waits until the next window comes round.
When the date is set by physics, the plan bends around it. Nobody argues with Mars.
Fixed date, flexible everything else
MMX has already lived this once. It was planned for 2024 and was moved back two years to 2026. That is what a missed window costs at planetary scale: not a week of overtime, but the next alignment. So the teams behind missions like this plan backwards from a date nobody can renegotiate, and they decide early which parts of the mission are allowed to flex.
We do not launch anything to Phobos. But a surprising amount of our work has the same shape. A trade stand has to work when the doors open at nine on the first morning. A brand installation has a press night. A store opening has a date on a billboard. If the system is not ready, there is no slipping it to Tuesday, because the event will not be there on Tuesday.
What a window changes about how you build
- 01Lock the date, then argue about scope. The list of features is the thing that moves. Agree on day one what gets cut first if the build runs late, so the cut is a decision and not a panic.
- 02Design the backup period in. JAXA did not hope for good weather on one night, it published eighteen spare days. We hold spare days before every install for the same reason: shipping delays, a venue that gives you power four hours late, a screen that arrives cracked.
- 03Know the cost of a miss in plain numbers. For MMX it is roughly two years. For a three-day event it is the whole event. That number should decide how much rehearsal, spare hardware and on-site cover the job gets.
- 04Separate what must be right at launch from what can be fixed in flight. Spacecraft routinely take updates on the long cruise out. A good installation is the same: the hardware and the core flow have to be right on the day, but content and tuning can keep shipping after the doors open.
- 05Rehearse the whole thing, end to end, somewhere that is not the venue. A full dress run in the workshop catches the problems a checklist never will.
Should clients care about a sample return mission? Only as a clear example. When a deadline is truly fixed, the honest plan says so at the brief and builds slack into everything except the date. When we quote an install against an event, that is the conversation we have first: what the window is, what it costs to miss, and which parts of the build are allowed to give.