Knorr-Bremse, the world's leading manufacturer of braking systems, stands for precision and accuracy, a quality that must be 100% right, down to the smallest component. This commitment to perfection naturally necessitates a high level of quality control throughout the entire production process. Even with the highest standards, defective components can emerge, requiring a robust rework process. To maximize efficiency and reduce errors in this essential quality control step, Knorr-Bremse partnered with STIWA, transforming their rework into a self-learning system.
Gerhard Schwarz, Plant Manager at the Aldersbach plant, highlights Knorr-Bremse's long-standing strategy: "Since 2011, we have been building on structured, data-based process analysis with professional data-based rework. Through the targeted networking, visualization, and analysis of data, we can keep our processes stable in the long term. This allows us to identify and cushion outliers earlier." Armin Eisenreich, Quality Planning Team Leader, adds, "This helps us to avoid rework as far as possible."
Equipping Staff With AI-Supported Rework
Knorr-Bremse’s pursuit of excellence has led to another important milestone with STIWA’s AMS Analysis-CI Rework. This self-learning tool, which is integrated into the ‘AMS ZPoint-CI’ machine control centre, enables machine operators to carry out rework more quickly and effectively. Thomas Langthaler, Engineering Team Leader and technical contact, explains: “This gives machine operators access to process analysis data directly at the machine. Armed with the knowledge of how faults have been correctly fixed in the past, they can react immediately and swiftly with the correct rework measures.”
Automating Knowledge for Faster Rework Decisions
Until now, specialists have analyzed the causes of errors in the production process when components were faulty. In doing so, they relied heavily on their years of experience, which meant that the success of the root cause analysis and the resulting corrective actions depended heavily on the knowledge of individual people.
The rework analysis tool automates a large part of these processes and continuously expands the information base for the self-learning system. As Armin Eisenreich of Knorr-Bremse emphasizes, this means: “If faults do occur, we now have the tools to get non-conforming parts back under control.”
This solution creates a fully standardized structure with a shared database and a user-friendly interface. Langthaler highlights its importance: “Linking the data shows how defects have been successfully rectified in the past, enabling the correct rework measures to be taken immediately and swiftly.” If no specific rework action can be derived from the fault report, the fault is forwarded to the quality laboratory for a more detailed examination, with the results being fed back into the system.
Predicting and Preventing Defects
“In future, the machine control center will be able to predict, based on all the data available to it, when a particular process will start producing faulty parts. Appropriate measures can then be taken in good time,” explains Armin Eisenreich. This foresight enables Knorr-Bremse to tackle potential problems proactively, rather than merely reacting to them. The data also provides a solid foundation for future machine learning models
A 30% Error Reduction
Many rework loops are eliminated, and new employees quickly achieve rework success. Gerhard Schwarz confirms the significant impact: "We expect an improvement of the functional-test-failure-rate by 30 percent from it." This translates directly into lasting decreases in production costs.
Building on this success, what was successfully launched in Aldersbach will now be implemented throughout the entire corporation. AMS Analysis-CI Rework will be rolled out at the Hungarian Kecskemét location and at the North American subsidiary Bendix. The data from ZPoint-systems across all sites will be transferred into a cloud solution, allowing operators in Hungary and North America to leverage the comprehensive practical knowledge from Aldersbach from day one, further increasing product quality and sustainably reducing production costs across the globe.