Replan Academy · Planning with Replan
From planning objectives to network alternatives
Animation paused (reduced motion). Press play to watch.
In one sentence
Objectives in, alternatives out — the engineer decides what goes forward.
Why it matters
The planner writes the brief on the map: which LRT stations to connect, which corridor to protect, maximum route length, minimum coverage, no parallel running, a fleet limit. The generator proposes candidate networks; twenty routes are generated, the planner removes and moves routes, ten are retained, local requirements add thirteen more — a 23-route engineered variant that is then simulated and compared. Not "AI designs the network".
Most of what a transit planner does before a model run — drawing, scheduling, checking reach and cost — is where time is lost between the big model iterations. Seeing that step on a map, with the numbers following the sketch, is what makes the workflow believable.
Sources
- Objectives and constraints are typical of a network-redesign brief (illustrative)
- Schematic, synthetic city (not a real network or tile map); values are illustrative
What you see
- 00.7The brief on the map
- 05.0Generate
- 06.8Three candidates
- 11.320 generated
- 12.6Planner removes routes → 10 retained
- 15.8Local requirements → 23 routes
- 19.4Simulate & compare
Use this explainer
Poster (PNG) · MP4 / GIF: rendered on request
Embed: <iframe src="https://www.replan.city/explains/embed/objectives-to-network-alternatives.html?lang=en" width="900" height="360" style="border:0;max-width:100%;aspect-ratio:5/2" loading="lazy" title="From planning objectives to network alternatives"></iframe>
Paste into any page; the explainer plays as live SVG (no video file).
Licence: CC BY 4.0 — reuse freely with the line “Replan GmbH · replan.city/explains”. The Replan mark in the corner stays.
See it on your own network
Book a demoRelated explainers

This is Replan: from the first line sketch to a modelled scenario
One workspace from the first line sketch to a modelled scenario.

The MATSim loop
The converged state is what scenarios are compared in; calibration tunes the parameters so that it matches observation.

Testing a feeder change before implementation
Both cases sit side by side in the same workspace.
