Com­puter Sci­ence News

From re­search pro­ject to start-up: AI mar­ket­place op­tim­ises pro­duc­tion pro­cesses

 |  ResearchEIM-NachrichtenCS-Nachrichten

Reducing production times, saving costs and identifying errors - artificial intelligence (AI) is an important key to product development. In order to bring users, providers and experts in AI together, eighteen partners from science and industry have developed a digital platform as part of the ‘AI Marketplace’ research project. During the three-year project period, numerous solutions for specific industrial applications have already been developed. The project, in which the Heinz Nixdorf Institute at Paderborn University played a leading role, has now resulted in a digital start-up that provides companies throughout Germany with easy access to AI applications from third-party providers.   

Promoting competitiveness with AI

‘Engineering is becoming increasingly digital and AI is becoming more and more crucial for the economy. Many companies see AI as an important technology of the future and are increasingly developing strategies to utilise its potential in a targeted manner and integrate it into their business processes,’ says Ruslan Bernijazov, a member of the Advanced Systems Engineering department at the Heinz Nixdorf Institute in Paderborn and co-founder of the AI Marketplace. ‘Identifying new technologies, optimising design data or analysing market opportunities - AI offers great potential for increasing productivity and economic growth. However, manufacturing companies often lack sufficient expertise to tap into the diverse possibilities. This is where we come in with the ‘AI marketplace’.’ Together with Prof Dr Roman Dumitrescu from the Institute of Computer Science at Paderborn University, Bernijazov founded the start-up following the successful completion of the research project. In addition to free web checks and other consulting formats, they now also offer an app and service store that gives users easy access to AI applications from third-party providers.

Successful pilot projects implemented

During the three-year project period, the scientists have already successfully implemented several pilot projects with the platform. For the internationally active agricultural machinery manufacturer ‘CLAAS’, for example, a knowledge database has been created that can be used to find identical and similar components in stock. Against the backdrop of constantly increasing product complexity and variant diversity, the reuse of components offers an important opportunity to save on production, development and storage costs. The ATM manufacturer ‘Diebold Nixdorf’ was able to reduce the processing time of service cases through the project, as the AI localises the cause of a fault before technicians even arrive at the machine.

The company ‘düspohl Maschinenbau’ wraps wooden, metal or plastic profiles using the ‘RoboWrap’ machine. While employees used to set up the machine themselves, the company now relies on AI. The result: the time required to set up the machine has been reduced from up to 16 hours of manual work to just 10 minutes In addition, there are significantly fewer errors than with manual set-up - the scrap rate can be reduced by around 50 per cent.

Thanks to the ‘AI marketplace’, ‘westaflex’ was able to significantly reduce its production time. The company now uses AI software to plan the sequence in which orders are produced, replacing analogue planning boards and spreadsheets. In a pilot project, ‘Hella Gutmann’ has developed ‘Automated Diagnostics’, in which AI identifies potentially defective components in the vehicle based on error codes and sensor values. Thanks to the large amount of data available, defective components can be identified with a high probability in more than 80 per cent of diagnoses.

Ubermetrics Technologies GmbH analyses and processes digital feedback to find out how certain products are perceived by customers. In collaboration with the experts from the ‘AI marketplace’, the company was able to automate its analyses. Among other things, AI applications now help to extract and analyse relevant product information from texts such as press releases or social media posts.

‘The project has enabled us to promote the competitiveness of the industries and give Germany visibility in the global AI market,’ summarises specialist group leader Dumitrescu. The project was funded by the Federal Ministry for Economic Affairs and Climate Protection (BMWK) with a total of 11.5 million euros.

Further information can be found on the AI Marketplace website.

Photo (it's OWL): Partners from science and industry have developed a digital platform for optimising production processes as part of the "AI Marketplace" research project.
Photo (Paderborn University, Tanja Dittmann): The AI marketplace provides companies throughout Germany with easy access to AI applications from third-party providers.