Meet Stéphanie Joudrier, our OCC Product Line Manager!

What impact will AI have on metro operations?

AI-driven solutions are already a reality. Crowd detection is an example. We brought this technology to maturity and today it is deployed in our projects. Work in this area, computer vision, continues.

But now the picture is broader: what we are doing is to inject mature AI into our solutions for new applications beyond computer vision. The focus is on identifying key pain points for our customers and using AI to solve them. 

Typically, what sort of pain points do your customers face?

Our modular Supervision Solution is built to monitor and control a very wide range of function in OCCs, including traffic management, station management, security and passenger information. One of the pain points for our customers can be information overload – particularly with alarm management, because a single incident can generate, in some specific cases, an avalanche of alarms. 

How can AI help?

We already provide a user-defined alarm grouping function. Now we are taking this a step further using machine learning to automatically identify related alarms – intelligent alarm grouping. So instead of managing multiple alarms, the operator only has to deal with one, with no loss of information. This is a concrete new use for AI. Incident investigation and reporting is another example. Currently, this involves manual transcriptions of recorded audio – a major pain point. Audio-to-text is also a game changer to save operator time. 

Can you also use AI to extract meaning from a large body of text?

Yes, definitely...  We’re all familiar with ChatGPT – everybody is looking at how to use it in their personal and professional lives. Generating text synthesis is one of its usage. The same type of technology could be used to support operators, for instance,  when they need to investigate incident root causes, helping them to extract and synthetize meaning full data coming from all supervision systems. 

What about connectivity & cloud?   

We are developing a new generation of cloud native supervision solution providing full-secured connectivity for the operators. This delivers multiple benefits.

First, OCC operators are no longer tied to a specific workstation – they are able to sit anywhere in the control centre and log in via a secure web-based connection. This provides a whole new level of flexibility. 

Second, connectivity extends collaboration. For example, maintenance teams and station managers – not normally based in the OCC – are able to access OCC data & control commands instantly using a tablet or mobile device. External stakeholders, such as transport authority staff, are also  able to access OCC data remotely.

Third, cloud makes it easier than ever to innovate. More and more of our customers are pushing for AI and digital twin. Our cloud-native digital platform means that software can be scaled and upgraded easily. The objective is to add or remove applications without affecting everything else – and do at any point in the life cycle. So our customers can benefit from continuous upgrades. 

What additional benefits do you get with an OCC digital platform?

An important upcoming benefit is being able to create a “digital twin” of your infrastructure. With digital twin, we can do sandbox simulations and use historical data to explore “what if” scenarios and even make predictions, defining strategies which can be reinjected in real life afterwards. This helps not only to optimise day-to-day operations, but also to make better strategic decisions.

Can digital twin help operators to achieve green goals?

Yes. A typical use would be to reduce train energy consumption. For example, you can simulate how changes to train speed and scheduling reduce power demand. And you can see what impact this would have on your traffic regulation & passenger comfort KPIs. The idea is to provide features that allow you to explore “what if” scenarios so you can make the best trade-offs – for example, between regulation and energy consumption. We call this smart regulation.

What first attracted you to the railway industry?

I came into ground transportation through Thales’ innovation lab. I initially worked on AI and smart sensors for critical infrastructure, urban security and e-health – this is why I joined. We also had a fantastic innovation project to use video analytics to detect things on trains. Ten years on, I am proud that our video analytics technology is successfully deployed to monitor real time passenger density on-board trains. It is the result of a fruitful collaboration grounded in trust between innovation lab and development teams.

What advice do you have for a young engineer  coming into the industry?

Choose a subject and a role that you can be passionate about. Be curious and keep learning continuously since things are evolving fast. Cultivate enthusiasm and team spirit.

What’s your favourite train trip?

There are two that I really enjoy. One is going to London by Eurostar. The other is taking the night train to the south of France. I like the ambience of the train and it saves time, because you can work until the end of the day. And the next morning, you are in front of the sea!