evidoria

← Back to browse

Good practice Imported

AI Video Analytics & Digital Twin — Monaco's Monte-Carlo Station Passenger Safety System

Monaco · Monte-Carlo · See the Monaco profile

Evidence: Descriptive / self-reported Top 74% 47/100 · Ask Evidence Copilot about this practice

SNCF Gares & Connexions and Monaco's government paired a digital twin with AI video analytics at Monte-Carlo station to monitor passenger density in real time — anonymised, no facial recognition — for roughly 8 million annual passengers; safety alerts remain in development.

~8,000,000
Annual station passengers
~5%
Vendor-reported counting-accuracy margin

Details

Maturity
Pilot
Promoter
SNCF Gares & Connexions / Principality of Monaco (Extended Monaco digital strategy)
Period
2024-2025 (pilot ongoing)
Keywords
transport, public safety, smart city, computer vision

Context

Monaco's 'Extended Monaco' digital strategy, launched by Prince Albert II in April 2019, set a goal of embedding AI across public services; Monte-Carlo station handles heavy passenger spikes during events like the Monaco Grand Prix.

Activities

SNCF Gares & Connexions, with video analytics from XXII and a digital-twin/orchestration platform from Akila, feeds real-time computer-vision analysis (not facial recognition) into a live 3D digital twin of the station, giving operators per-square-metre passenger-density readings for the station's roughly 8 million annual passengers.

Results

The vendor reports a counting-accuracy margin of about 5%, and video is anonymised on-device to meet GDPR.

Conclusions

Safety-specific features such as flame/smoke detection and track-intrusion alerts are described by the project lead as still in development, so the system's contribution to actual safety outcomes is not yet independently measured.

Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Medium (1–3 years)
Staffing & skills
SNCF Gares & Connexions project team, XXII (video analytics vendor), Akila (digital-twin/orchestration platform vendor)

Conditions for success

  • On-device video anonymisation (no facial recognition) built in from the start to meet GDPR
  • Digital twin designed to handle both routine operations and mass-event spikes (e.g. Monaco Grand Prix)

Common failure modes

  • Safety-specific features (flame/smoke detection, track-intrusion alerts) remain in development, so safety-outcome benefits are not yet demonstrated

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful