Edge AI in Cement: Deploying Real-Time Predictive Models for Fuel Savings in Clinker Kilns

Argos’ Digital Transformation to Drive Sustainability

Grupo Argos, a Colombian multinational in infrastructure, is driving a bold transformation toward sustainability. Achieving carbon neutrality demands innovation across all operational levels, and Argos’ Digital Manufacturing department is leading the charge. By centralizing digital initiatives, the department has spearheaded transformative projects to streamline production, improve equipment maintenance, and modernize control systems.

These initiatives target critical areas such as:

1. Fuel Savings in the Pyroprocess: Deploying real-time predictive models to optimize fuel consumption during the energy-intensive pyroprocess phase, unlocking significant cost savings

2. Predictive Maintenance for Gearboxes: Aiming to minimize costly downtime, ARGOS explored predictive models to identify potential failures before they occurred.

3. Aggregate Plant Monitoring: Enhancing visibility across disconnected sites to create a centralized and cohesive system for managing remote operations.

4. Computer Vision for Quality Control: Harnessing advanced AI to detect production faults, ensuring higher-quality outputs while minimizing waste.

5. Digital Twin for Cement Production Optimization: Implementing predictive control algorithms to automate and optimize cement production processes for improved efficiency and consistency.

Revolutionizing Clinker Kiln Operations with Real-Time Predictive Models at the Edge

One of the most significant challenges in cement production is the high cost of fuel, particularly in the pyroprocess stage. Initially, Argos relied on traditional fuzzy logic and rule-based control systems, which required frequent manual operator adjustments. These systems, while better than manual operation, were slow and inconsistent.

Seeking improvement, Argos experimented with cloud-based tools, which improved response times but lacked the real-time precision required for optimal control. To overcome these limitations, Argos developed and deployed real-time machine learning algorithms directly onto its clinker kilns, ensuring split-second decision-making in an environment where efficiency is critical.

Objectives of the Project

Argos focused on three key goals:

1. Real-Time Variable Control: Achieving precise, real-time control over critical variables during the pyroprocess, particularly in its fast-moving initial phase.
2. Fuel Cost Reduction: Optimizing control loops to cut heavy fuel consumption and improve cost efficiency.
3. Scalability: Creating a standardized solution that could be deployed across all Argos plants worldwide.

‍

‍

The Solution: Barbara´s Cybersecure Edge AI platform

To meet these ambitious objectives, Argos deployed a cybersecure edge computing platform. The solution, powered by Barbara, provided the following capabilities:

• Barbara’s Cybersecure OS: Deployed on virtual machines, it ensured secure and efficient operation at the edge.
• Barbara Panel: A centralized, cloud-based console enabling Argos to connect assets, deploy AI models and applications, track versions, and update configurations remotely.  It also provided full control over its edge nodes, enabling full lifecycle management, remotely and privately (via VPN).

‍

As Estefan Wolff, Digital Manufacturing Leader at Argos, noted: "One of the aspects I liked the most about Barbara was its hardware agnosticism, both from the capture side with the wide range of supported industrial protocols, and deployment, covering ARM nodes such as Raspberry Pi and Jetson, as well as Intel CPUs, GPUs, and virtual machines."

‍

Measurable Impact

The results of this transformative project speak for themselves:

• $200,000 in fuel savings annually per plant through optimized fuel consumption.
• 60% reduction in bandwidth costs by preprocessing operational data at the edge, transmitting only critical information to the cloud.
• Scalability across plants: Enabling the deployment of new AI models and applications, including use cases like predictive maintenance, energy efficiency, and quality control with a future-proof infrastructure already in place.
• Overcoming fragmentation and heterogeneity of environments in each plant.

A Path to Sustainable Innovation

Argos’ journey demonstrates how leveraging cutting-edge edge computing and AI technologies can transform industrial processes while advancing sustainability goals. By investing in scalable, efficient, and secure edge solutions like Barbara, Argos is not only reducing costs and improving operations but also taking meaningful steps toward its ambitious goal of carbon neutrality by 2050.

This project serves as a model for how industry leaders can align operational excellence with environmental responsibility, driving innovation for a more sustainable future.

‍

Download the customer story here

‍

‍

‍

‍

‍