Machine learning applied to chemical synthesis optimization
Our client commissioned MP DATA to apply an innovating approach in order to study potential non-quality root causes and to develop, subsequently, a process parameter recommendation engine which aim is to maximize output quality depending on the production context.
Following an initial analysis, we implemented an analytical approach based on explainable predictive models capable of predicting the risk of non-quality and prescribing the corrective actions to be taken to minimize It from the early stages of the synthesis. The explainability of the models not only facilitated the adoption of the solution, but also allowed us to complete the root cause analysis with targeted process studies, which confirmed the conclusions of the Machine Learning approach.
The solution was directly included on specific screens in the control room.
-20% of waste
compared to the worst situations the site had encountered
>250k€ of yearly savings