Bruno Valat, General Manager, and Julia Beumer, Head of Sales for DILAX

The Business of Mobility is an Urban Mobility Company series highlighting some of the most successful new businesses in the mobility sector. Featuring a closer look at the way in which companies stand out, CEOs, Directors and other c-level executives elaborate on what it takes to turn a great idea into a great company.

Collecting passenger data starts with smart sensors, a field DILAX has pioneered since 1988. Bruno Valat and Julia Breumer discuss how DILAX is helping create better cities with their innovative contribution to automatic passenger counting (APC) systems.

Sign Up


By signing up to the Autonomy newsletter you agree to receive electronic communications from us that may sometimes include advertisements or sponsored content.

The challenge of developing a sensor as accurate as the human eye (while still protecting privacy)

An APC system consists of two main components: a counting sensor that collects the data, and a master unit that stores and transmits the data to a central computer where an API turns it into useable information for the client. It’s a tricky technical problem to develop a sensor that matches the detail and accuracy of the human eye but at the same time protects passengers’ privacy in compliance with GDPR. And that’s why – somewhat incredibly – we still see people with a handheld ticker counter, counting shoppers in Paris. It’s also why traditional people-counting technology – like ‘3D stereo vision’ – was designed to mimic human sight. It’s called ‘stereo’ because it uses two cameras (positioned above where people enter), which offers similar depth perception as the human eye. It’s a good solution, and with the help of AI can be around 97% accurate.

However, the challenge for us is to get as close as possible to 100%. There are other technologies, like thermal imaging and photoelectric break beam, but these are not as accurate as 3D stereo vision.

Innovating with laser technology

Our newest sensor departs from the traditional technology. A laser emits a three-dimensional pattern of dots (what we call ‘structured light’) onto a targeted area where passengers enter and exit the vehicle. A receiver picks up the pattern of interference created by human bodies – and other large objects – giving us 99% accuracy and assuring privacy by design. The technology also identifies bikes, wheelchairs and strollers, and distinguishes between adults and children.

The importance of passenger counting

DILAX is dedicated to improving the offering of public transport operators, thus encouraging citizens to make the switch from car ownership. That’s why it was so important for us to count bicycles, strollers, wheelchairs and children. This information is useful for operators to understand their market and what sort of lifestyle people lead to meet their needs.

What’s more, thanks in part to smartphones and apps, operators can now communicate in real-time about the occupancy of an arriving vehicle (a key advantage given the pandemic) and whether, for example, there is available space for a stroller. We’re even getting to the point where passengers will be able to view, while waiting on the platform, which wagons will be less occupied than others. We will also be able to see which seats are available and book them ahead of time.

But we don’t just count passengers. Our systems also monitor arrival and departure times for trains and buses. As research proves, nothing irritates customers more than lack of punctuality. In this regard, APC data is also helping public transport companies meet KPIs, manage their fleets, and improve the service and network.

Business intelligence based on counting

DILAX employs 160 people – mainly based in Berlin. When it comes to product, we more-or-less follow the Apple model, designing all hardware in-house (with production partners in Berlin and Brandenburg), and focusing on integrating the hardware and software from the ground up. Our customers interface with our APC systems through a dashboard called Citisense, which delivers sophisticated business intelligence for public transport. Citisense generates dozens of dynamic reports across a range of variables. The tool can even run forecasts and simulations of changed routes and stops. Cities can also open the database up to third-party mobility providers to help integrate modes and enhance the overall customer experience.

In one project, our data for a smaller city in the south of France helped the authorities to make the choice to invest in 24-meter-long busses. The decision was made after we had been collecting data for ten years; it helped the operator reduce costs significantly while improving the service. There are countless other examples of how routes, timetables and vehicle size have been optimized.

Next steps

DILAX is not only involved in mobility but in optimising city space generally, be it stations, malls or workspaces. However, given the disruption in the industry, mobility is an exciting space for us and one that fits neatly with our brand slogan: ‘local data for better cities’. In terms of the refinement of counting methods, we are arguably reaching some sort of endpoint, with the hardware (in tandem with sophisticated AI algorithms) delivering close to 100% accuracy. Indeed, we are almost there; therefore, the real advances will come from the data science and how we can process the data to deliver a better service for passengers and give richer insights to planners. There will also be great advancements in MaaS as operators use the data to encourage multi-modality.

Don’t miss DILAX at Autonomy Digital 2.0, where they will be present as an Industry Partner! Catch their panel on May 19 at 9:30 CEST – « How Local Data Drives Accessibility in Public Transport. »