Roadmap

The principle

The fastest way to ruin a platform is to build every part of it half-way. VPM takes one capability as far as a customer will pay for it before starting the next. Fuel came first because it is the clearest case; it is not the end of the list.

Now — live

• Next-port prediction, with second and third calls ahead.
• Fuel need, grade, cost and CO₂ to the next port.
• Load-aware arrival and departure windows on real sea routes.
• The Port View — arrivals for any port, on the tracked fleet.

Next

• Wider port coverage. A port becomes commercially useful once enough of its arriving traffic is tracked. Closing that gap, port by port, is the main effort.
• The Port View, deeper. Today it reverses the tracked fleet. Next is live regional traffic — vessels currently underway toward a port — and then full prediction for every vessel in a region, which needs wider data.
• A reporting-shaped emissions output for compliance teams.

Later — the need-layers

Each of these runs on the same prediction engine, adding one "what will they need" layer on top. In rough order of how cleanly they follow from what already exists:

• Lubricating oil — follows almost directly from the consumption model.
• Provisions, water and waste — driven by the call itself.
• Maintenance and repair — driven by running hours and age.
• Dry-dock and hull work — cyclical, needs a docking-schedule signal.

Every port, multiplied by every layer, is a separate market. They get built one at a time, each finished before the next begins.

Other seas

The engine is not tied to the Baltic. Anywhere with AIS coverage and shipping-lane density data can be added as a data exercise rather than a rebuild — the North Sea already is.