Reduction of the undulation in Steel sheet

With data collected directly from the factory floor, Clarcat has been able to carry out Data Mining with the aim of identifying conditions under which defects are more likely to appear throughout the production process.

To carry out this task several techniques were used among which are: data cleaning methods, analysis of trends, clustering and the designing of “decision trees”.

With the aid of tools such as Rapid Miner and Pentaho we have been able to process and analyse large volumes of data.

Thanks to our work our client is now able to act according to a set of parameters and reduce the error rate, obtaining an in-depth view into the processes they carry out.