Skip to main content

AI and data

The power of Data Innovation and Artificial Intelligence

Sirris experts in data & AI contributing to the Flanders Impulse Program in Artificial Intelligence 

In 2019, the Flanders Impulse Program in Artificial Intelligence was officially launched after several months of preparations and negotiations. Already in this early phase, Sirris was involved and contributed with its deep understanding of the needs of the Flemish industry to define the main axes of this program. 

 

The aim of this program is to strengthen research in the field of AI in Flanders and to ensure the transfer and adoption of this knowledge by the Flemish industry. It focuses on four main challenges in the field of AI: data science for supporting complex decision making (1), delivering artificial intelligence to the edge (2), interactions between independently programmed AI systems (3) and communicating and collaborating seamlessly with humans (4). 

 

Together with the other Flemish partners involved in the programme, the Sirris is participating in this programme by mainly contributing to challenge 1 and 4. The Sirris data & AI experts  contribute to this project with the coordination of proof-of-concepts (PoCs). In its role of coordinator,  Sirris collaborates with the other researchers, ensures that the research results are validated on real-world industrial use cases and datasets and steers the upcoming research activities.  The AI & data experts are in particular working on PoCs around prognostics and health monitoring for ageing wind turbines and around smart environments supporting humans in their tasks.  

Distributed product usage analytics, a research topic highlighted in 2019

In 2019, it was observed that many companies in the Elucidata Lab's network operate large groups of complex, geographically distributed assets/products, which are more and more being instrumented and connected. Even though these companies gather and access large amounts of data, many of them expressed that they were still looking into innovative ways for exploiting that data. 

Analysing data about how a product/assets is used in the field, can for example reduce overdesign and extra costs, avoid long subjective discussion on warranty and liability or help identify in which context failures happen. However, there are still quite some challenges for companies. For example, they still need to look into how they will deal with the absence of continuous connectivity and short offloading opportunities. They might also be confronted with limited on-asset storage, low bandwidth and high communication costs. In some other cases, challenges related to data ownership, privacy and security will come into question. Related to this topic, two projects were initiated in 2019. 

 

  • The Innoviris supported project “Distributed product usage analytics (STUNT)” in cooperation with the company ILIAS Solutions started in April 2019. This project will deal with distributed product usage analytics in the context of asset management of off-road and special-purpose vehicles in mobile operations. 
  • Additionally in 2019, the COOCK project “Artificial Intelligence for Distributed Asset Usage Monitoring (AI4DETAIL)” was set up. With this project, the Elucidata Lab aims at accelerating the introduction of technological and non-technological expertise around distributed product usage analytics by organizing master classes and inspiration seminars, offering a technological experimentation platform, organizing interactive 'Bring-your-own-data' sessions, and setting up company-specific projects. The project is currently supported by about 20 companies. The execution of this VLAIO supported project will start in April 2020. 

AI Starter Kit Portal, enhanced with dedicated video tutorials and training sessions

After having released the AI Starter Kit Portal in 2018, in 2019 the Sirris data & AI experts started to think about ways to make the methodologies and solutions developed in the Starter Kits even more tangible for its users.

 

Therefore, a new concept was elaborated in which the Starter Kits would be accompanied by online video tutorials explaining in detail the solutions and methodologies developed in each Starter Kits.

 

In order for companies to apply to their specific context the solutions developed in the Starter Kits, has decided to offer companies two “Bring your own data” training sessions. In a first session, companies will bring a business problem and a description of their available data, which could be used to solve this problem. Together with the participants and the Sirris  experts a plan to analyse this data is elaborated. In a second session, companies would do a first analysis of this data with the  experts. 

 

The execution of these activities will take place in 2020 in the framework of the ESF, EU & Flemish  funded project SKAIDive (Interactief Starter Kit Platform voor de adoptie van industriële AI en Data Science competenties)

EluciDATA Community counts about 30 members 

In 2019, the EluciDATA Community, created by Sirris, became the place where Belgian companies interested in AI and data innovation, meet to exchange and set up new project ideas around these topics, but also share experiences, discuss emerging technologies and they can rely on the Sirris data & AI experts. In 2019, around 10 project ideas were presented to the Community members during the two-yearly events, out of which 4 were recently submitted. These involved around 9 companies.

 

The official kick-off meeting of the EluciDATA Community together with the first Brokerage event “Data and AI driven innovation” aiming at the set-up of new R&D projects took place in April 2019. During this event, members were informed on how to participate in the AI Impulse programme, different EU, Flemish and Brussels funding modalities, as well as the opportunities and challenges related to the participation in collaborative projects via the testimonial of RSscan international, a research intensive SME and the Sirris experts.

 

In December 2019, the EluciDATA Lab organised its second annual event, “The EluciDATA Tech Talk”, focusing on the application of advanced AI/ML methods in industry. Some concrete applications of advanced AI/ML methods in industry were presented by Materialise and the Lab’s experts presented some of the emerging advanced learning algorithms such as federated learning and reinforcement learning.

 

Elucidata Community members (April 2020)