Command CenterLive · 284 Trains · 2.4M Passengers/Day
On-Time Performance
92.4%
↑8pts AI disruption mgmt
Passenger Satisfaction
4.3/5
↑1.1pts AI service
Infrastructure Failures
−47%
AI predictive maintenance
Energy Cost
−22%
AI traction optimisation
🤖 Agent Status
Real-time across all capabilities
Train Performance AI284 trains · delay prediction
Infrastructure Intelligence AITrack · signalling · −47% failures
Passenger Intelligence AI2.4M/day · service recovery
Disruption Management AIOTP 92.4% · recovery planning
Energy Intelligence AI−22% · traction optimisation
Safety Intelligence AISMS · SPAD · near-miss analysis
📡 Live Intelligence Feed
Real-time AI activity
Why RailOS
🚆 Performance: Every Delay Has a Cascade
Rail delay is exponentially costly — one delayed train affects hundreds of connections. AI train performance intelligence predicts delays 20-40 minutes ahead — giving control rooms time to intervene before cascades develop.
🔧 Infrastructure: Track Failures Disrupt Thousands
Unplanned track failures cause major disruption at enormous cost — passenger compensation, alternative transport, emergency engineering. AI infrastructure intelligence predicts failures weeks ahead from sensor data.
⚡ Energy: Rail Traction Is the Largest Cost
Rail traction energy is 30-40% of operating cost. AI optimised driving advisory reduces energy consumption 22% through eco-driving profiles, regenerative braking optimisation, and real-time speed advice.
All AI Agents
🚆
Train Performance AI
Delay prediction, real-time performance, headway management, service recovery.
284 trains
ReAct + Telemetry
🔧
Infrastructure Intelligence AI
Track, signalling, OLE monitoring. Failure prediction 2-4 weeks ahead.
All infrastructure
ReAct + Sensor
👥
Passenger Intelligence AI
Real-time information, crowd management, accessibility, complaint prediction.
2.4M/day
ReAct + NLP
Disruption Management AI
Scenario recovery, timetable adaptation, resource re-planning, passenger rebooking.
All disruptions
Planning + Scenarios
🌿
Energy Intelligence AI
Eco-driving advisory, regenerative braking, depot energy management, carbon reporting.
All trains
Planning + Optimisation
🦺
Safety Intelligence AI
SPAD monitoring, near-miss analysis, SMS evidence, safety trend detection.
All operations
Sequential + Rules
📊
Revenue Intelligence AI
Yield management, pricing optimisation, season ticket analytics, revenue protection.
All routes
ReAct + Pricing
📡 Live Agent Trace
All decisions logged
🛡 AI Governance
Advisory intelligence — humans decide
No autonomous consequential actions: All significant decisions require human approval. AI recommends — authorised professionals decide.
Full audit trail: Every AI output logged with source data, reasoning, and confidence. Explainable to regulators and auditors.
Continuous evaluation: RAGAS faithfulness >0.92 required. Quality degradation detected within 24 hours. Human override always available.