Chances of the digitalization in the test field for operations and product development

Authors

Roland Strixner, Kratzer Automation AG

Year

2019

Summary

„Connected, Autonomous, Shared, Electric: Each one of these points has the potential to turn our industry upside down. The real revolution however lies in the connection of all of these“.
The automotive world is in turmoil. This turmoil is being forced unusually fast by different drivers, which has significant consequences on the requirements concerning the development processes of the product Automotive and its components. Digitisation plays a big role in this process and opens up the oppurtunity to confer a major edge.
Innovations in automotive can only be used as a unique selling point if you are the first, or at least one of the first to bring the innovation to series production and sales. These innovations need to be tested however, since malfunction can lead to loss of reputation on the market and may even cause lasting damage to the business strength. Kratzer Automation AG sees a trend in the requested test fields towards more and more largescale test fields. In addition, the test capacities in the industry are being greatly enhanced to be able to handle the increasing number of tests. Appropriate efficiency in the corporate processes around and in the test field are indispensable in order to be able to use the investment material “Test bench” accordingly.
With the growing number of tests the amount of data increases massively as well. Furthermore,
especially with classic drives, one gets further and further to the limits of the technically feasible. The classic analysis of tests where one prototype was evaluated is on a decrease. Often a series of identical prototypes is being subjected to the same tests. Modern software packages used for post-processing are able to recognize outliers and analyse the whole series of prototypes. Defects in the manufacturing of the prototypes can therefore be recognized and filtered in the analysis.
On top of that mathematical algorithms like cluster analysis can be used to scan the tests for tendencies and characteristics that don’t show any obvious correlations. Such machine learning algorithms can thus provide clues for optimizations in design and construction.

Members Login