PredictLand has developed, for the multinational automotive company Linde + Wiemann, a solution for the reduction of energy consumption and predictive maintenance while ensuring the quality of work, in short, with the aim of increasing the efficiency of welding processes.
Customer
Linde + Wiemann, based in Wuppertal, Germany, is a leading manufacturer of metal components and assembly solutions for the automotive industry.
With more than 40 factories worldwide, it operates in countries in Europe, America, China and South Africa.
Its annual turnover is in the range of several billion euros, making it a major player in the global automotive industry.
Challenge
One of the biggest challenges in the industry is high energy consumption, in particular sub-optimal welding processes that have a major impact on weld quality and generate high energy consumption. Linde Wiemann is interested in reducing energy consumption while ensuring production quality.
Solution
To achieve this goal, PredictLand develops an Artificial Intelligence system, combining Computer Vision and Machine Learning/DeepLearning techniques. This system works with real-time process data from a robotic cell. This solution will allow its scalability to all robotic welding processes.
Results
With the results, Linde Wieman estimates a reduction of around 53% in energy consumption, while optimizing cycle time in the welding processes, without loss of quality, improving the efficiency and OEE of the stations involved and, consequently, of the entire factory.
The project, called NEOFAR, has been supported by the Ministry of Industry, Trade and Tourism, as well as by the European Union through the Recovery, Transformation and Resilience Plan; and has received funding from the Ministry of Industry, Trade and Tourism within the support program for AEIs to contribute to improving the competitiveness of Spanish industry.
