Use Case
Cement Industry

Challenge

The customer aimed to maximize the mill feed rate while ensuring stable mechanical conditions and minimizing the risk of equipment stress. Traditional manual adjustment was too slow and inconsistent to react to the dynamic behaviour of the mill.

To overcome these limitations, the customer sought an AI-driven recommendation system capable of continuously suggesting optimal combinations of three critical set points under varying process conditions.

Approach

Wizata collaborated with the plant’s process engineers to build a data-driven closed loo automation system.

First, historical process data was cleaned, removing period of undesired or unstable operating conditions, to train a reliable AI model. Using this refined dataset, machine learning algorithms were developed to:

➜ Identify valid operating clusters of set point combinations.

➜Recommend in real-time the most effective parameters for current process conditions.

➜ Detect mechanical anomalies in hydraulic and lubrication systems before failure could occur.

➜ Finally automate the full system

All recommendations and alerts were integrated into a real-time dashboard, allowing operators to visualize status and automatically apply AI-driven optimizations.

Outcome and Results

➜ Real-time AI recommendations for mill set points.

➜ Fully automated closed-loop optimization, reducing operator intervention.

➜ Early anomaly detection preventing costly mechanical failures.

➜ Centralized dashboard for continuous visibility and control.

The implementation delivered 10% increase in yield, improved process consistency and stability and reduced downtime.

The solution has been successfully scaled to 5 more mills across different plant locations, confirming robustness of the solution and ROI within a year.