Sirris is a private collective research centre founded in 1949 by Agoria, the multi-sector federation for the Belgian technology industry. Sirris has about 2500 member companies, which range from small and medium-sized businesses to multinational companies (¿95% SMEs), active in 13 technological sectors including Information & Communication Technology. Its mission is to improve the competitiveness of its member companies through technology and innovation. Sirris has currently about 140 employees focusing on knowledge acquisition (applied research) and knowledge transfer (industrial projects). The subdivision ICT & Mechatronics of Sirris will participate in MANTIS.
Relevant Expertise for the project:
Within Sirris ICT & Mechatronics, the expertise with relation to advanced data processing is bundled in the Data Innovation team, a small group (6 team members, all with PhD degree) of highly qualified experts who participate in industry-driven research projects in collaboration with national and international partners (e.g. 6 ARTEMIS projects). The team has set up several industry-driven research programs through which it has created a broad network of international research and industrial partners. Through these projects, the team has acquired extensive knowledge in various areas related to topics such as scalable data processing, complex event processing, pro-active and context-sensitive decision support, user profiling and modelling. Through participation in these projects, the team has created a broad network of international research and industrial partners, such as for example Fundaci´on Tekniker, ST Microelectronics, Tecnalia, ATOS, Tampere University of Technology, Philips, Thales, Institut Polytechnique de Bordeaux, Technical University of Sofia and TNO. On national level, the team has established very successful collaborations with several different academic organisations in Belgium e.g. iMinds, ULB, VUB, UGent, KULeuven, KAHO St. Lieven, etc.
Role in the project:
In MANTIS, Sirris will:
- contribute to the definition of the service platform architecture (WP1) and to the development of several of its components (WP2), most notably:
- distributed data storage components for storing large amounts of (historical) data obtained from heterogeneous sources
- distributed stream-based data processing for supporting (near-)real-time anomaly detection
- batch-oriented data processing for building predictive models based on trend and pattern detection in historical data.
- explore the use of semantic technology for integration, harmonisation and standardisation of heterogeneous data, e.g. operational data, maintenance and event logs, external data sources such as ERP systems, etc. (WP4)
- apply advanced data processing algorithms for knowledge discovery, e.g. temporal data mining algorithms (e.g. sequential pattern mining, temporal association rules, …) for discovering root cause failures and events that co-occur in a certain order, probabilistic modelling approaches (e.g. Bayesian networks) for discovering complex relationships between the different events and for predicting asset failure, multi-criteria decision analysis for collaborative decision making, etc. (WP4)
- validate the developed approaches, methodologies and resulting models in specific use cases