Industrie 4.0

Data analytics for the fourth industrial revolution, such as proactive service and maintenance of production resources or finding anomalies in production processes.



Data-driven aspects of medicine are explored, such as the need-driven care of patients or IT controlled medical technology.


Smart Infrastructure

Untersuchung datengetriebener Aspekter städtischen Lebens, bspw. der Verkehrssteuerung, der Müllentsorgung oder der Katastrophenbewältigung, bedarfsgesteuerte Optimierung von Verbrauchsmodellen, basierend auf Daten intelligenter Stromzähler.

Featured Projects

  • BigGIS_logo

    BigGIS: Fusion of geospatially distributed heterogeneous Sensor Data

    Increasing data volumes and increasingly complex calculation models require fast and robust procedures. This is the topic of the BigGIS project, in which integrated procedures for dealing with uncertainty within (geo-)big data are developed. Together with the SDIL, suitable algorithms are implemented, tested and further developed on the basis of temperature data.

    Read more

  • SDSC BW: Sawing with the SDSC experts

    During the planning and production of sawn timber a wide range of data accumulates: from quality data of the wood to the data generated at the saw line and the sales data. In a joint project with the SDSC-BW, the sawmill Karl Streit from the Black Forest optimized its approach to the planning of round timber incisions.

    Read more

  • QuestMiner - Detection and evaluation of anomalies in graphs

    An anomaly is generally defined as a deviation from the norm and from the expected behavior. Such anomalies often indicate incidents and constellations that require immediate attention and reaction. In a social network, an anomaly can indicate spontaneous attractions such as demonstrations. Their early detection is crucial for further management. With regard to graph data, anomalies can be modeled as subgraphs, in which the nodes deviate significantly from the norm attribute values and edge distributions. In the case of a dynamic graphs, historical conditions can also be taken into account.

    The goal of the project is to develop a method for dynamical heterogeneous graphs, which is able to continuously detect and evaluate anomalies. Thereby, the procedure should be designed for real-time service and should permanently be supplied with a stream of new data. In addition, the method should be scalable and able to process large amount of data. For this, not only an algorithm specially adapted to this problem is necessary, but also supporting index and data structures that provide efficient access to historical data. The applicability and practicability of the procedure should be assessed during the course of the project by means of a prototypical implementation for which we want to use the SDIL platform.

    Read more

  • SDSC BW: Precise planning of production processes

    Sedus Stoll AG is a full-service provider of office equipment and workplace concepts. For the analysis project of the SDSC BW, Sedus provided data from the production of office chairs, which are available in a large number of configuration variants.

    Read more

  • SDSC BW: Plant growth by means of image recording

    The da-cons GmbH sells products for the determination of a variety of plant properties. To this end, it has developed the PhenoScreen system, which supports the seed industry in plant breeding and is based on sensors such as cameras, hygrometers and luxmeters. As part of the potential analysis, the SDSC BW examined the question of whether it is digitally possible to reliably detect plants in the image files.

    Read more