At the moment, it is mainly off-the-peg goods that fill department store shelves. In the future, however, products will become more individualized, and in the long term, some industries will move towards batch sizes of one - in other words, individual items. In automotive production, this goal is not far off; vehicles are already being assembled according to customer requirements. This poses many challenges for production, which can only be mastered with technologies from the field of cognitive sensor technology: Components must be identified and localized, machines and systems must communicate with each other and with their human colleagues. In addition, suitable application logistics are required to control decisions and processes.

The Fraunhofer Institute for Integrated Circuits IIS offers technologies to meet these challenges. The solutions range from localizing industrial trucks to improve warehouse processes and delivering the engine to the correct station, to supporting assembly with intelligent tool tracking, intelligent containers and picking systems, to monitoring the condition of machines.

The data-based optimization of work processes is the basis for increasing efficiency throughout the entire company. For example, data collected using cognitive sensor systems can also be used to automatically control and monitor the supply chain using predictive analytics - a forecasting process used to determine future events.

All of the technologies presented have already reached such a high level of maturity that they have been tested and optimized in industrial pilot projects. Pilot projects are currently underway with BMW and research and development projects with Siemens AG and other partners who are equipping parts of their production and logistics with digital technologies in order to provide their employees with targeted support through assistance systems in their interaction with the machine.

Nerve cells of the industrial Internet of Things

"With cognitive sensor technology, the digital transformation can be put into practice," explains Prof. Dr. Albert Heuberger, Managing Director of Fraunhofer IIS. "Cognitive systems are the nerve cells of the Industrial Internet of Things, or IIoT for short. They not only record the measured values, but also evaluate them directly, make decisions based on intelligent conclusions and forward them as required. Fraunhofer IIS provides concrete solutions for wireless communication and localization in the entire IIoT and IoT environment. Cognitive sensor technology also relies on the use and integration of machine learning to provide the right data in the right place at the right time for the right application."

Intelligent containers order supplies

In assembly, for example, it is important that the workers have all the parts to be installed to hand at all times and that the assembly line does not come to a standstill due to a lack of material replenishment. On the other hand, more parts than necessary should not be kept in stock, as this drives up storage costs. The researchers have developed a tailor-made solution for this: Intelligent containers that know where they are, what their condition and the condition of the parts is, and that automatically order replenishment as required depending on the fill level.

The containers communicate via s-net technology, which also originates from Fraunhofer IIS. The bins communicate wirelessly with each other and with the infrastructure to form a network. They pass on their information to the employee via a dynamic display - for example, they inform them when the full container they have reordered will arrive. The data recorded by the bins is collected in a cloud, where it is available for big data analysis. Inductive near-field positioning can be used to check whether the worker has actually reached into the right box and where they can find the next part to be installed.

The fitter not only needs small parts such as nuts and bolts, which can be stored in such containers, but also larger components, which are stored in shelving systems. In order to keep the storage space as small as possible, the space is used flexibly. This means that the components are always in a different place. A light signal therefore guides the fitter and shows him where he can find the required product in the supply rack. However, such conventional pick-by-light systems are either wired and therefore difficult to install or they have a short battery life. In the Pick-by-Local-Light (PbLL) project, the researchers are therefore developing a new type of picking system based on wireless sensor networks. The basic technology here is also s-net. as