AI system
- System
-
Automated Intelligent Video Review (AIVR) front-of-train camera system with machine-learning analysis
- Model
-
Not assessed
- Release
-
Not assessed
- Application developer
-
One Big Circle
- Role
-
Not assessed
- Configuration
-
Not assessed
- Inputs and outputs
-
Not assessed
- Autonomy
-
Not assessed
- Workflow
-
Front-of-train cameras on scheduled passenger services covering the same track daily. Machine-learning review that raises possible defects to the infrastructure manager. A maintenance organization able to schedule overnight possessions.
- Human oversight
-
Not assessed
- Relation to AI
-
AIVR's analysis raised the fault report that Network Rail engineers acted on. No human had reported it. Whether the minor fault would have escalated is LNER's claim. Detection rates and false-positive rates are not published.
- AI contribution
-
AIVR's analysis raised the fault report that Network Rail engineers acted on. No human had reported it. Whether the minor fault would have escalated is LNER's claim. Detection rates and false-positive rates are not published.
Benefits
- Outcome identifiers
-
Not assessed
- Benefit
-
an overnight repair with no delay to passengers and zero delay minutes
- Category
-
Improved Organizational Efficiency
- Status
-
Not assessed
- Beneficiary groups
-
Organizations. Society. People.
- Beneficiaries
-
Not assessed
- Reach
-
Not assessed
- Distribution and access
-
Not assessed
- Duration
-
Not assessed
- Expected?
-
Yes
- Costs and harms
-
Not assessed
Measurement
- Studies
-
Not assessed
- Study design
-
Not assessed
- Measurement
-
Not assessed
- Setting and sample
-
Not assessed
- Measures
-
an overnight repair with no delay to passengers and zero delay minutes
- Comparison
-
a track defect reported by a train driver a week earlier (January 2026, Cambridgeshire)
- Results
-
Reference: over 10,000 delay minutes, multiple cancellations, and led to a full day of disruption for customers. Result: an overnight repair, with no delay to passengers, and zero delay minutes caused.
- Additional value from AI
-
Not assessed
- Limitations
-
Not assessed
- Denominators
-
Not assessed
- Materials for replication
-
Not assessed
Money
- Kind of value
-
Not assessed
- Amount
-
Not assessed
- Currency and prices
-
Not assessed
- Period and unit
-
Not assessed
- Realization
-
Not assessed
- Recipients and payers
-
Not assessed
- Gross or net
-
Not assessed
- Included costs
-
Not assessed
- Comparison
-
Not assessed
- Method
-
Not assessed
- Return and payback
-
Not assessed
People and places
- Participants
-
Developer: One Big Circle. Operator: LNER (on-train cameras), with findings passed to Network Rail.
- Jurisdictions
-
Benefit: Retford,
United Kingdom. Developer: United Kingdom. Operator: United Kingdom. - Ownership and control
-
Not assessed
- Dependencies
-
Not assessed
- Cross-border connections
-
Not assessed
- Geopolitical implications
-
Not assessed
Follow-up
- Conditions for replication
-
Engineers triage AI reports promptly. False positives (such as the snake) do not swamp the reporting channel. The fault is visible from a train-mounted camera.
- Replication attempts
-
1 follow-up entries. Benefit did not recur: 1.
- Later findings
-
6 July 2026. The system was still running on LNER trains in July 2026, when it raised a snake crossing the line as a potential track defect; Network Rail said raising it automatically was standard practice.
- Availability
-
Not assessed
- Institutional decision
-
Not assessed
- Grounds
-
Not assessed
- Action and result
-
Not assessed
- Success criterion
-
Not assessed
Evidence
- Source numbers
-
AIBD-0255-source-0001–0002
- Citations
-
Numbered references below.
- Language
-
Not assessed
- Source relationships
-
Independent, Operator
- Original evidence
-
Not assessed
- Access and versions
-
Not assessed
- Supported findings
-
Not assessed
- Evidence links
-
Not assessed
- Evidence strength
-
Not assessed
- Missing evidence
-
Not assessed
Grades
- Assessed benefits
-
Organizations. Society. People.
- Depth
-
Organizations: 2. Society: 1. People: Not assessed.
- Reasons
-
Organizations: An overnight planned repair with zero delay minutes. The comparison is a defect the same month that cost over 10,000 delay minutes and a day of disruption. Society: Passengers were spared delay from this fault. The scale of avoided disruption is not quantified. People: No individual outcome is documented.
- Observed or intended
-
Not assessed
- Rubric
-
Not assessed
- Assessor and date
-
Not assessed
- Alternative assessments
-
Not assessed
- Fill
-
Independently reported
Entry details
- Kind
-
Not assessed
- Related entries
-
Not assessed
- Field
-
Railway safety. Track inspection / defect detection.
- Purpose
-
AIVR is built to capture and analyze front-of-train video with machine learning. The aim is to find and fix track and lineside defects before they cause disruption.
- Dates
-
Date basis unassessed: 2026 (precision unassessed). Updated: 21 October 2026.
- Stage
-
Not assessed
- Version
-
SHA-256: c075807381a9…
References
- Other identifiers
-
Not assessed
- Category mappings
-
Not assessed
- Brief
-
Not assessed
- Revisions
-
5 revisions. Latest: 21 October 2026.
Sources
-
News report Independent author
Pashby T (6 Jul 2026). AI system on LNER train misidentifies snake as track defect. New Civil Engineer.
a week later, the AIVR system identified a minor fault which had the potential to turn into a larger issue near Retford.
-
Press release Deployer
Network Rail (31 Oct 2022). Future Rail – how machine learning is improving your journeys.
Automated Intelligent Video Review (AIVR) – developed by Bristol-based tech company One Big Circle – gathers a wide range of data from high-definition video around the railway.