PLATFORM
Canopy Platform Edge Devices AI Agents Privacy by design
INDUSTRIES
Industries overview Hotels & Hospitality Retail Smart Cities Construction Out-of-Home
PROOF
Case Studies Insights
TECHNOLOGY
Technology overview Edge Compute Data Sovereignty OAIC Compliance
COMPANY
About Us Contact
Sign in Get in touch
CASE STUDY

Passenger flow optimisation at a Sydney transport hub

Anonymous crowd sensing for a busy interchange.

Case Study · Smart Cities — Sydney transport interchange

A Sydney transport interchange had daily passenger counts but no live density figure. Operations could not justify additional platform staff without one.

Busy railway station platform with commuters

The problem

The interchange had passenger counts in the back office, but they were aggregated daily. Operations could not see a density spike until it was already over. They could not ask for more staff without a live number, and they could not install facial recognition — the public would not accept it, and the regulator would not allow it.

Our deployment

We installed a Canopy CU-3 in the station equipment room, fed by existing platform cameras. The edge ran a density agent on defined zones: entry, platform, bus stop. When a zone exceeded its planned density for more than 60 seconds, operations got an alert. The numbers were aggregated, not individualised.

The result

Real-timeDensity visibility, previously daily
0Facial recognition deployed
0Offshore data

Operations staff used the live density view to justify two additional platform staff on weekday mornings. The interchange became demonstrably safer during peak, without any new surveillance capability that identified commuters.

Canopy Console — Zone density Interchange · weekday AM peak · alert after 60s above plan Entry Platform Bus stop 6:00 7:00 8:00 9:00 10:00 Dashed line = planned density · shading = alert window Zones Entry live Platform live Bus stop live Facial recognition 0 deployed Aggregated counts only
Zone density through the weekday AM peak, with the 60-second alert rule applied.

Implementation details

The interchange ran a Canopy CU-3 unit in the station equipment room, fed by 14 existing platform cameras. Density zones were drawn on a station floor plan in the console: entry concourse, platform 1, platform 2, bus stop, taxi rank. The density agent reported the number of people in each zone every five seconds, and triggered an operations alert when a zone exceeded its planned density for more than 60 seconds.

The edge unit ran entirely on the station's internal network. The event stream to our Sydney control plane was HTTPS, on a dedicated VLAN. Operations staff saw density in real time on a wall-mounted dashboard in the station control room. No individual was tracked across zones; the number in each zone was aggregated.

What the data showed

Between 8am and 9am on weekday mornings, platform 1 ran at 140% of planned density for 20 minutes. Two additional platform staff were added for that window. Signage at each camera read "Anonymous crowd counting in operation".

Ongoing support

The deployment is now in its second year. Operations staff use the dashboard daily. The planning team uses the aggregated numbers quarterly, to inform timetable and staffing decisions. We have helped the station's security team run a tabletop exercise around a density event; the dashboard was the single source of truth. The edge unit has survived one network outage of 36 hours, with no lost data.