Every Minute in a Queue Counts: The Race for Context in Airports

With passenger counts and sales reports, the question has no answer. The data shows that more people traveled and that the commercial area sold less per passenger. It does not show that the security queue grew during the morning peak, or that check-in for a given airline took twice as long.

The friction points of a terminal are where the passenger's available time is lost. And every minute in a queue is a minute less in the commercial area.

In an airport, the queue is not only an operations metric. It is the first variable of non-aeronautical revenue.

The old model: spot measurements and dedicated sensors

Sensors placed at a few checkpoints, or manual samples, give a partial snapshot. At check-in they often fail: counters change by airline and queues form chaotically, so a fixed sensor cannot measure them reliably.

The trap: AI devices per zone

Installing dedicated AI equipment in each checkpoint or waiting area repeats the hardware model at terminal scale. The number of points makes it unviable, devices become obsolete in a few years and each vendor stays limited to what its own equipment can measure.

What the leading airports do

They use the cameras the terminal already has. AI runs as software over the CCTV to measure queues and waiting times in real time, follow passenger flow between zones anonymously and connect available time with the commercial area. Agents alert when a queue exceeds its threshold and support service agreements with airlines and providers.

In the published case below, one camera covered 500 m², with 10x fewer monitoring points than traditional sensors. In real projects, KSI has validated a difference of 10x or more compared with dedicated hardware: lower infrastructure cost and more analytics coverage for the same budget.

The main airport of a country, run by one of the world's leading private airport operators, needed to improve the passenger experience and increase sales in its commercial areas. The diagnosis was clear: the stress generated by friction points reduced passengers' willingness to buy.

«At the Check-In area we encountered a complex technical problem, as conventional sensors could not measure queues because airlines have variable counters and service, and lines form chaotically. That's when KSI appeared and was able to solve the problem.»

What was implemented with the cameras of the terminal:

• Monitoring of all waiting queues and friction zones.
• Wait time measured individually per passenger and, at check-in, segmented by airline and flight.
• Automatic notifications when the number of people in a queue increases.
• Automated passenger experience SLA reports for the airport and the airlines.

With precise data, the airport set service standards for each congestion point and agreed a data-based continuous improvement program with the airlines. The link between queues and revenue stopped being a hypothesis.

With KSI data connected via MCP (the open standard that lets assistants such as Claude, ChatGPT or Gemini query external data), the AI assistants used by operations and commercial teams will be able to answer terminal questions in natural language and alert when something needs action:

• Which checkpoints exceeded their SLA this week, and with which airlines?
• How much time did passengers have left for the commercial area after security, by hour?
• Where should staff be reinforced in the next peak, based on the queue history?
• Real-time alerts when a queue exceeds its threshold, sent to the team that can act.