Focus projects

In our project, we are developing an AI-supported solution for the non-destructive detection of defects in light metal castings, in particular car tyre rims. While conventional quality assurance systems are based on X-ray images and classic image processing methods, our innovative approach offers the opportunity to work more flexibly and efficiently. This is particularly valuable when specialised personnel are scarce or frequent recalibrations are necessary. Together with Fraunhofer IIS, we are endeavouring to enable precise and automated defect detection.

The use of AI enables the processing of large amounts of data and can help to optimise complex procedures and decision-making processes in healthcare. However, there is often uncertainty as to what AI can be used for in healthcare, what risks and potentials exist and how the quality and transparency of AI applications can be ensured.

The »Drive 4.0 Living Lab« project brings together leading research institutions and industry partners to develop intelligent, efficient and sustainable drive solutions. By integrating advanced data analyses and artificial intelligence, we are creating a digital platform that offers cross-manufacturer services and innovative usage options.

Increasing digitalisation is also generating more and more data in manufacturing companies. At the same time, these companies are subject to the challenges posed by current market demands and customised production. Increased demands on production itself and the need for transparent and optimised material flow processes are the result. This raises the question of how operational process data from production can be utilised. The ‘ProCheck’ project therefore aims to use process mining methods to create process transparency and enable automated and continuous analysis, optimisation and review of material flow processes in small-scale production processes.

In industry, numerous processes are recorded manually: Those responsible observe the behaviour of employees, record the times of process steps, identify the weak points of a process and manually design solutions for process improvement. Measures are often identified manually and evaluated based on experience. At the same time, more and more data - error messages, key figures or batch information on parts - is being collected and stored in Industry 4.0. However, this information is usually isolated from one another in data silos, which is sufficient for a single solution in digitalised production. However, the derivation of process models and performance indicators requires a more holistic picture of the available information in order to be able to define predictive measures. The aim is to depict processes so transparently that it is possible to automatically record and analyse process models using cyber-physical (CPS) or Internet of Things (IoT) systems and artificial intelligence. Gleichzeitig werden im Bereich Industrie 4.0 immer mehr Daten – Fehlermeldungen, Kennzahlen oder Chargeninformationen von Teilen – erhoben und gespeichert. Meist sind diese Informationen allerdings isoliert voneinander in Datensilos, was für eine Einzellösung in der digitalisierten Produktion ausreichend ist. Die Ableitung von Prozessmodellen und Performancekennzahlen benötigt allerdings ein ganzheitlicheres Abbild der vorhanden Informationen um daraus prädiktive Maßnahmen definieren zu können. Ziel ist es, Prozesse so transparent darzustellen, dass eine automatische Aufnahme und Analyse von Prozessmodellen durch den Einsatz von cyber-physischen (CPS) bzw. Internet of Things (IoT) -Systemen und Künstlicher Intelligenz möglich ist.

A key element of the quality promise made by many manufacturers is the long service life of the appliances they produce. This includes guaranteeing the availability of spare parts so that repairs and maintenance can be carried out over the long term - even if the appliance is no longer in production. The challenge here lies in the forecast that needs to be made at the end of the regular production run in order to pre-produce and stock the corresponding quantities. The previous manual processes are very time-consuming and harbour many uncertainties. BSH Hausgeräte GmbH has therefore commissioned the Supply Chain Services working group to find a data-driven solution that can provide automated and reliable support for the decision.

In the »KITE - Artificial Intelligence in Transport to Reduce Emissions« project, researchers from the Supply Chain Services working group at Fraunhofer IIS developed a new AI-based route planning process to reduce empty runs and make transport logistics more sustainable.

The »Fluid 4.0« project brings together fluid technology manufacturers, machine and plant manufacturers, users and service providers from the fluid industry for the first time in order to digitalise processes with data and technology support and to make circular economy strategies and technologies usable.

Global supply bottlenecks, rising raw material prices, dwindling resources - sustainable business makes sense not only ecologically, but also economically. And that's not all: The regulatory requirements of the German government or the EU are also demanding more and more commitment from business. For example, companies must continuously demonstrate their CO2 emissions by calculating their product carbon footprints, while at the same time demonstrating suitable measures to reduce their own CO2 footprint and reduce the need for natural raw materials. Circular or circular economy approaches offer high solution potential here and are therefore increasingly coming to the fore in considerations.