Kaolin Quality Control


A multinational company that specialises in the processing of industrial minerals, approached PREDICT to provide a solution to identify the process drifts that impact the quality of Kaolin.

Detected process drifts

Anticipated the drifts that impact quality of Kaolin

Optimise process parameters

To improve the operating conditions for calcination

Improve quality control

By suggesting necessary actions at the right time


The client was looking for a solution that could

  • Detect the process drifts

    The entire operational unit was thoroughly analysed and models were formulated to identify the normal operational behaviour which helped in anticipating drifts in the process.

  • Optimise the process parameters

    Analytical algorithms were developed that could take into account various aspects of the client’s installation and optimise the parameters to produce the best quality materials from Kaolin calcination.

  • Improve the quality control

    Understanding the processes and employing methods to detect the drifts helped the client in having a better handle on the overall quality control process.


Kaolin is subjected to Calcination process to alter its chemical and physical properties and produce materials intended for a specific application. To obtain the desired product of the highest quality, it is critical to maintaining the right operating conditions for calcination.

PREDICT was able to provide a solution that could anticipate the drifts in quality control of the Kaolin calcination process by identify the most optimal parameters for process. PREDICT could further alert the client whenever the operating conditions were drifting from the normal expectations.

how can we help you?

Contact us to know how Predictive Maintenance can transform your organisation.

Imerys is fully satisfied by the performance of PREDICT and work regarding the implementation of the Kasem software within the Monicalc project, at the Imerys Minerals Site in the UK.

Imerys Minerals Limited

Looking for a cost-effective Predictive Maintenance solution?