© unsplash / Scailable

To get a good picture of the IT & Data Science sector and Brabant’s strengths in this sector, the Brabant Development Agency (BOM) conducted an in-depth study. In an almost 300-page report, the study provides an overview of the size of the IT and data science sector in Brabant, with all its trends and opportunities for collaboration. The study shows that it is not only the existing parties that are at the forefront of new developments; much is also expected from start-ups in this sector. One of these is Scailable, the start-up from Eindhoven that has developed software to process large quantities of data into usable information, no longer ‘in the cloud’ but ‘on the edge’. Meaning: hosted locally.

Read more about the nature and results of the study in our overview article.

The idea behind Scailable is that the cost savings, improved security and increased speed achieved by processing data where it is collected will play a major role in accelerating the introduction of Artificial Intelligence (AI). To do this, Scailable compresses high-level AI models into a secure low-level format that runs on a small piece of software on the ‘edge devices’ that collect the data. This requires a minimal amount of memory and can run on a wide variety of chipsets. Which has many advantages. Not only is the response time of the AI models much faster, but the software can also be easily distributed, is compatible with all common tools used by data scientists, and requires minimal server capacity. Moreover, the solution offered by Scailable works even when the internet connection is poor.

“The possibilities of AI are endless,” says founder and CEO Maurits Kaptein, also a professor at JADS and at Tilburg University. “Think about improving safety in factories because cameras detect that someone is standing too close to a machine. Or improving the production process where a robot immediately intervenes if products do not meet the quality standard.”

Subscribe to our Newsletter!

Your weekly innovation overview Every sunday the best articles of the week in your inbox.

    Data traffic

    A lot of AI technology now still uses the cloud environment. There, calculations are made that are then sent back to the device. Kaptein: “A large amount of data traffic is expensive and causes privacy issues. What’s more, it works too slowly for more and more applications. Some 80 percent of AI applications remain on the shelf, unused. Our software ensures that all those wonderful applications can be implemented much more easily.”

    For a very long time, Kaptein and his co-founders worked from a purely content-based, academic perspective. They had come up with an interesting solution and wanted to deliver it as well as possible. But the question was when and to whom. “My network in academic circles is fine, but in the industry a lot less so. With the BOM’s investor readiness program and initial investment from BOM and Volta, this changed. We were able to attract the right people for these tasks and have now also become much more robust as a company.”

    That new network was much needed because in the high-tech manufacturing industry that Scailable focuses on, it is not yet commonplace to use AI in business operations. Kaptein: “That is also completely understandable because AI is a means, not an end in itself. So it is not immediately logical to come to us with a specific task within a production process. We will first have to explain the advantages to such a company.”

    Book of Inspirations

    This now happens not only through company and trade fair visits, but also with a handy ‘Book of Inspirations‘ that Kapstein’s colleague Benedicte Lochtenberg – CCO at Scailable – has compiled for that purpose. The publication contains a wide range of logical applications for the manufacturing industry for which a camera and an industrial gateway or IPC workstation together with Scailables software can provide cost savings, higher safety, or a better product. The examples range from real-time monitoring of products and walkways to crowd and employee detection, face blurring, determination of the right time to harvest, object recognition, and improved level crossing security for railways.

    museum AI
    Unsplash / Scailable

    In all these cases, it is highly questionable whether the “problem owner” thinks at all about a solution via AI, Kaptein knows. “Someone who spends all day reminding his employees to wear a helmet and to follow the correct walking routes in the factory has other things on his mind. With our solution, that all happens automatically. Just like we can tell the rotten apples from the fresh ones, recognize busy zones and empty shelves in stores, and warn when people don’t keep enough distance.”

    Scailable is now deriving its first revenue from a number of specific projects, including at Lely. In addition, the first initial licenses have been sold for the use of the edge software and devices. “Licensing is what we are currently focusing on in terms of our revenue model. We expect a lot from it because the scale is potentially huge; through our software, almost any edge device can run advanced AI models.”

    Collaboration

    This story is the result of a collaboration between Brabantse Ontwikkelings Maatschappij (BOM) and our editorial team. Innovation Origins is an independent journalism platform that carefully chooses its partners and only cooperates with companies and institutions that share our mission: spreading the story of innovation. This way we can offer our readers valuable stories that are created according to journalistic guidelines. Want to know more about how Innovation Origins works with other companies? Click here

    About the author

    Author profile picture Bart Brouwers is co-founder and co-owner of Media52 BV, the publisher of innovationorigins.com.