From the Transportation edition of November 5, 2026
Barcelona TMB uses AI on 4,500+ bus cameras for real-time passenger counts and camera health
Transports Metropolitans de Barcelona (TMB) · Transit · Barcelona, Spain
- Publisher
- Barcelona City Council Press Service
- Original publication
- 2026-11-04
- Source retrieved
- 2026-11-06
What happened
At Tomorrow Mobility World Congress 2026, TMB said it now has real-time passenger counts per bus by applying AI to images from more than 4,500 onboard CCTV cameras covering the full bus fleet, fused with validations, surveys, occupancy alarms and video counting. The in-house bus technology team built the system with AI vendors; models are audited and continuously retrained. Outputs are intended for service regulation and passenger information, and the same stack flags failed or misaimed cameras to speed maintenance. A related bus-lane intrusion detection pilot also uses AI on cameras.
Why it matters
A fleet-wide reuse of existing transit CCTV for occupancy AI, with explicit audit/retrain loops and a dual use for camera maintenance—directly comparable to U.S. transit APC modernization.
Evidence and measured results
Official city press note; no published accuracy rates, latency figures or before/after reliability metrics.
Limitations and uncertainty
European operator; Catalan-language primary source; measured accuracy not disclosed; passenger-info use is planned rather than proven in the note.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-11-05; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Transit operators lack reliable real-time per-vehicle occupancy and struggle to keep large CCTV fleets healthy.
- Stakeholders
- Transit COO / service planning
- Bus technology / ITS lead
- Customer information lead
- Maintenance
- Discovery questions
- What APC accuracy and latency do you have today?
- How many onboard cameras and what is their uptime?
- Who may use rider video for analytics?
- Value hypothesis
- Vision AI on existing CCTV fused with fare data may improve occupancy estimates and camera maintenance; unproven quantitatively here.
- Potential engagement
- Bounded corridor pilot comparing AI occupancy to manual counts and APC.
- Unsupported claims
- Published accuracy rates
- Service reliability gains
- Passenger information quality improvements
Pre-sales engineering
Role takeaway
- Fit
- Fits agencies with dense onboard CCTV and digitized fare validations.
- Architecture and integration
- Vision models on fleet video fused with validation and alarm feeds; camera health as secondary signal; details not disclosed.
- Prerequisites
- Fleet CCTV access
- Validation data feed
- Model audit process
- Constraints
- Privacy rules for rider video
- European regulatory context
- Vendor collaboration not specified
- Security
- Segment video access, minimize retention of identifiable frames, log model versions and audit results.
- Proof of value
- Manual occupancy audit sample
- Camera-fault detection precision/recall
- Service-regulation decision latency
Delivery
Role takeaway
- Work
- Deploy counting pipeline, audit cadence, ops dashboards and maintenance tickets.
- Dependencies
- Bus tech team
- AI vendors
- Operations and maintenance
- Ownership
- TMB bus technology team with operations consumers.
- Skills and adoption
- Ops training on using real-time occupancy; maintainers on camera-fault tickets.
- Governance checkpoints
- Model audit before production use
- Privacy review of video analytics
- Retrain trigger policy
- Acceptance criteria
- Proposed: occupancy MAPE vs. manual audits
- Proposed: camera-fault MTTR reduction
- Proposed: documented retrain log
- Risks
- Privacy backlash
- Bias in crowded/lighting conditions
- Over-trust without published accuracy
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Computer vision on fleet CCTV fused with fare validations and other onboard occupancy signals; camera health monitoring is a second output of the same stack.
Governance
Who approves, reviews and stays accountable for outcomes?
Continuous result audits and retraining are cited as quality controls; passenger-image retention and purpose limits are not described.
Security and privacy
What data, permissions and controls need testing?
Video of riders is processed for counts; privacy, retention and access controls are not specified in the release.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Real-time occupancy can improve crowding information for riders; camera-maintenance workload shifts toward AI-flagged faults.
Procurement
What should contracts, pricing and exit terms secure?
In-house build with expert AI vendor collaboration; contract terms not disclosed.
Operating model
Which teams own the service once it runs?
Occupancy AI feeds service regulation and passenger info; same stack supports camera maintenance.
Publication history
- 2026-11-05Transportation · Issue 084 resources
Stable resource ID: tmb-ai-bus-passenger-counting-cameras-2026-11-04