SATS · Storm-Adaptive Transit Systems

A digital twin that rehearses the storm before the storm arrives.

The ClearGlass SATS digital twin runs continuous, physics-aware storm simulations over subways, tunnels, streets and drainage — then routes every predicted intervention through governed approval. Below is the live simulation core: the same deterministic model that ships in tools/sats_risk_engine.py, running in your browser.

Live twin: run a storm over the demonstration network

Governance is the product. The twin never acts on its own. Advisories auto-publish; service adjustments and protective closures are drafted, risk-scored, and blocked until a human operator approves — the same read-only analysis → draft → approval → execution invariant that governs every ClearGlass autonomous system.

What the SATS twin is made of

No single vendor covers engineering-grade assets, physics simulation, IoT ingestion and agentic operations at once. SATS composes best-of-breed layers behind one governed control plane:

1 · Data & shared model — Azure Digital Twins / AWS IoT TwinMaker: secure ingestion from weather APIs, USGS gauges, GTFS-RT and trackside sensors into a DTDL-modeled shared state.
2 · Asset core — Bentley iTwin: engineering-accurate models of stations, tunnels and drainage networks, federated from BIM and survey data.
3 · Simulation & prediction — the ClearGlass SATS risk engine (deterministic, stdlib Python) extended with MATSim agent-based mobility and physics-informed models; scales into NVIDIA Omniverse for full physics AI.
4 · Visualization — CesiumJS for web-native 3D geospatial views with live flood-risk overlays; Omniverse for immersive operations centers.
5 · Agent orchestration — LangGraph/Temporal agents read twin state (build_twin_state() JSON) and draft operations and passenger guidance.
6 · Security layer (the ClearGlass differentiator) — zero-trust, role-scoped shared views and append-only audit trails so transit agencies, utilities and emergency services can work from one twin without over-sharing.

Platform fit at a glance

Full analysis, scoring rationale and deployment guidance live in the companion guide: Best Digital Twin Simulation Tools for Storm-Adaptive Transit Systems in 2026.

PlatformBest SATS roleFit
NVIDIA OmniversePhysics-AI simulation + immersive visualization9.5/10
Bentley iTwinEngineering-grade asset core (tunnels, drainage)9.0/10
Azure Digital Twins / AWS IoT TwinMakerIoT data fabric & shared model8.0/10
MATSim + CesiumJSAgile agent-based mobility sim + web 3D8.0/10
Autodesk TandemStation-level operational twins7.5/10

Pilot a storm-adaptive twin

The MVP path is deliberately lightweight: the Python risk engine plus Cesium visualization and a cloud data fabric — live hazard ingestion and 1–3 hour storm simulations in weeks, not quarters, scaling to Omniverse-class physics when the mission demands it.

Book a security engagement See pricing & plans