Center for Applied Data Science Gütersloh (CfADS)

Center for Applied Data Science (CfADS), Bielefeld University of Applied Sciences

The Center for Applied Data Science (CfADS) at the Bielefeld University of Applied Sciences supports companies and institutions on their way to digitisation. It conducts research in the field of artificial intelligence, particularly in the key areas of data science, machine learning and cloud-based automation and optimisation. Its own high-performance research infrastructure enables both scientific and application-oriented research together with partners from industry.

CfADS develops new business models on the basis of learning and intelligent methods or optimises existing processes and products. Together with partners, the concepts and methods are put into practice. The AI developed by CfADS in the form of innovative algorithms can be implemented on its own Data Analytics Cluster and tested at its own IIoT Factory in Gütersloh. In addition, assistance systems that support humans can be developed, examined and evaluated in the human-centred Smart Service Lab. With this extensive and powerful research infrastructure, CfADS can fully support its partners in generating further unique selling points from the raw material "data" by means of artificial intelligence.

Exhibit 1: AI-based model factory

A major challenge in production is to be able to react directly to incoming customer requests and to produce and deliver in lot size "one" and in the required time. This challenge can be met with intelligent processes from the field of artificial intelligence (AI). In this case, AI provides solutions to optimize the value chain.

At the model factory, it is shown how components that are to be manufactured into products with different characteristics and a different depth of production in the factory are fed into the production cycle (outsourced) so that the total production time is minimized. In doing so, new customer requests are dynamically reacted to and further restrictions, such as the respective delivery date, are taken into account.

The optimisation of the value chain in the model factory is based on models, from which a systematic control design is derived. The models are also used to evaluate the sequences determined by the AI, which should represent an optimal sequence.

Exhibit 2: Cloud-based predictive maintenance using the example of an industrial washbasin

The company Bio-Circle Surface Technology GmbH, together with the Center for Applied Data Science, has developed a cloud-based service and implemented it in a washstand, so that the washstand presents itself as an intelligent IoT device. The real data flow between IoT device, cloud and user (customer) as well as the methods used for communication offer an efficient and cost-effective operation of the cleaning system. The data sent from the washstand is processed on an ML/application server and visually processed via a web application. A dashboard view on the client displays graphics on different aggregation levels via a browser. In addition, forecasts are calculated from the data, which can be used to control maintenance and service according to demand and to ensure that consumables (e.g. cleaning liquid) are provided in a precisely fitting manner. The availability of the Bio-Circle system increases, while at the same time operating costs are reduced.