Organizations face new challenges every day due to the need to be competitive in their markets and, at the same time meeting the quality of the product, to satisfy the expectations of their most demanding customers.
Our client, an international steel manufacturing company, had different problems with the quality of their products: In some cases, these problems manifested as deviations from the values of the mechanical properties of the product, with respect to its specification.
This fact was only visualized at the end of the production process, where samples of the material are taken and tested in the laboratory to obtain the values of the mechanical properties. If the results showed deviations with respect to the product specification, the material had to be declassed, with the consequent business impacts.
The Solution The mechanical properties of the material are intrinsic variables of the same and can not be measured by direct methods but by means of laboratory tests.
Given the correlation between the mechanical properties of the material and the process variables that produced it, and given the impossibility of measuring them directly, we developed a predictive model based on machine learning algorithms trained with historical data.
The prediction of the mechanical properties in the middle of the production chain helps to detect deviations in an early stage, which allows acting downstream in the remaining processes, to correct the deviations generated.
The predictive model responded adequately, with prediction errors of less than 5% for 90% of the population considered.