Three Ways to Protect Passengers and Staff Using AI and Analytics

This blog looks at how AI and data analytics can help deliver an accurate and real-time view of what is happening to safeguard passengers and staff 24/7.

1. Live incident detection and prioritised alerts

With AI and data analytics as part of your command, control and communication solution, you’ll always have eyes on the situation. Continuous processing of huge volumes of data means you’ll instantly be alerted to security and safety risks - from trip or slip hazards and lone children to license plates linked to persons of interest on police watchlists.

The system also knows how to prioritise these positive alerts and who to notify based on the incident category. For instance, if a security guard spots an abandoned bag, they can press an emergency panic (shortcut) button on their mobile device which:

• immediately sends a notification to multiple control room operators
• automatically raises the system threat level with appropriate audio and visual queues

2. Spotting suspicious activity that warrants attention

AI and video content analytics also help identify behaviours that may indicate a potential safety or security risk to passengers or staff.

The system can spot suspicious or anomalous activity that warrants operator attention by applying algorithms and rules to live camera footage. Examples may include a person loitering (perhaps near concourse cash points), erratic driving in car parks/repeat circling of station approaches, or individuals waiting too close to the edge of platforms.

Notifications are prioritised and automatically issued to relevantly qualified and/or authorised staff while also ensuring control room teams receive real-time updates.

3. Alerting you to things that haven’t happened yet

Because AI is so good at spotting risk potential, your system can also issue ‘pre-warning’ notifications to help you avoid or better prepare for incidents.

Take overcrowding at platform barriers. This can lead to problems such as social distancing issues, passengers missing their trains, and potential aggression towards staff. Better to avoid this scenario all together. With AI/analytics, you can. Your system will spot when the number of people present is reaching a pre-specified threshold, and correlate this with train arrivals to assess impact on passenger flow. It knows when to automatically dispatch more support staff and/or update passenger info points to ensure wait times for both boarding and alighting passengers is minimised.

Smoke detection is another good example. Here the pre-warning notification can alert relevant personnel to the presence of smoke, prioritise footage from cameras to help confirm the presence of fire, and line up fire response protocols should positive confirmation be received.