Dr. Daniel Summerbell – Co-founder and CSO
Carbon Re
Cement is essential for modern life – it quite literally forms the foundation of our homes, cities, and infrastructure. As the largest source of industrial greenhouse gas emissions – accounting for 8% of global CO₂ emissions – decarbonising cement production is also crucial for tackling the climate crisis.
The case for early action
While cement companies acknowledge this challenge and are eager to innovate for improved sustainability, many rely heavily on Carbon Capture and Storage (CCS) technologies as the primary pathway to decarbonisation. However, widespread implementation of CCS is unlikely before 2040–2050.
This overlooks a critical concept: the “Time value of carbon”. The emissions we produce today have a greater impact on longterm global temperatures than those released in the future. This means that the most effective way to limit warming is to cut emissions as soon as possible. Early action is not just beneficial – it’s essential.
The three most impactful levers cement companies have to impact emissions today are:
• Increasing the use of alternative fuels;
• Increasing the use of substitute cementitious materials (SCMs);
• Process optimisation
AI is ideally placed to help with all these objectives. For example, AI enables higher alternative fuel use by ensuring that input variation does not impact clinker quality. And by improving clinker quality, AI facilitates more aggressive SCM mixes.
The role of AI
With the introduction of digitalisation infrastructure such as sensors and cloud historians, plant operators can monitor and store data in real-time. The data collected on variables like temperature profiles inside kilns or power consumption during grinding processes provides operators with a holistic view of their plant and begins to facilitate the identification of inefficiencies before they become costly problems.
The complexity of the cement pyroprocess is particularly wellsuited to AI and in particular machine learning, a subset of AI, due to the speed and accuracy with which machine learning algorithms can comprehend and identify patterns in data. Analysing the sheer number of data points per second produced and the complex interdependent process signals without machine learning is constrained by the manpower required and a limited number of trained process engineers across the cement industry.
With machine learning, a plant’s fuel consumption can be modelled and optimised to unlock previously out-of-reach efficiencies.
The benefits of AI-optimised combustion At its core, AI optimisation enables proactive control based on real-time data. By combining human expertise with the computing power of AI, Carbon Re has developed physics-aware machine learning models that continually assess all variations in plant conditions and predict changes in the kiln environment to recommend the optimal oxygen target, enabling more efficient combustion.
What does this look like in practice? Typically, operators run the fan slightly higher to avoid incomplete combustion and because it is simpler to operate in a ‘safe zone’ that requires less hands-on management. With Carbon Re integrated in closed loop, optimal operating parameters are identified and set point changes are automatically updated avoiding incomplete combustion, minimising fuel waste, reducing the specific heat consumption and saving on fan power and electricity costs. All while operators and process engineers focus on higher-value projects rather than the minutia of kiln control. The predictive power of our AI is transitioning plants to be proactive instead of reactive.
With specific AI-optimised models for each kiln, operators can be confident in their ability to increase alternative fuel usage and expand the variety of alternative fuels used. Our models are adaptive to condition changes such as changes to fuel mix and even hardware upgrades. Unlike expert systems, our models automatically retrain, so you don’t need to organise a service to retune them; they are designed to maintain accuracy as plant conditions change.
Improving product quality and reliability
Consistency in product quality is crucial for cement manufacturers to meet customer expectations, and with a digitalised cement plant, operators can make adjustments in real time to optimise the blending of raw materials or adjust quality indicator targets like C3S. This level of precision ensures that the final product meets the desired specifications consistently.
Additionally, the integration of AI models enables predictive maintenance practices that help reduce downtime due to unexpected breakdowns. This results in improved overall equipment effectiveness and reduced maintenance costs.
Managing blockages and fuel-mix variability challenges
With the increased use of alternative fuels, new complications emerge – notably, buildups within cyclones and kilns. These can impede airflow and material movement, leading to decreased efficiency and potential shutdowns.
Carbon Re has developed AI models that use real-time data from pressure, temperature, and gas analyser sensors to detect early signs of build-up and other operational anomalies. Our model provides earlier warnings compared to existing systems, allowing for preemptive measures to prevent potential blockages and avoid costly unplanned shutdowns – ultimately enabling even greater use of alternative, lower-carbon fuels.
Carbon Re utilises high-resolution data to detect the spikes in a range of sensors, including pressure that precedes a blockage, and predict the severity of the risk based on real-time conditions. This prediction gives operators hours of warning to adjust operational parameters to minimise the formation of a blockage and prevent unplanned outages.
Carbon Re’s approach
We collaborate closely with our customers to deploy advanced AI techniques tailored to each plant’s unique efficiency and decarbonisation goals, integrating directly with the plant’s Advanced Process Control (APC) systems, such as ABB Ability Expert Optimizer and FLSmidth Cement’s ECS/ProcessExpert. With AI-optimised models customised for each kiln, operators can confidently increase the use of alternative fuels and expand the variety of alternative fuels used to cut fuel costs and emissions.
We describe the APC as the precision controller helping the control room operators, and Carbon Re as the adaptive layer that supports process engineers. The results from combined roll-outs with ABB and Carbon Re are:
• 4–6% reduction in specific heat consumption
• 60–80% reduction in process variability
• 10–70 kg less CO₂ per tonne of cement
Together, EO and Carbon Re deliver significantly better performance– faster, smarter, and more adaptive cement production.
Fuel cost savings and carbon emission reductions at Heidelberg Materials Mokra Plant
To measure the results of Carbon Re in closed loop control at Heidelberg Materials Mokra Plant in Czechia, a month on/month off test was conducted. During this period the plant observed:
• 4.1% reduction in fuel cost index;
• 4.5kg/tonne of clinker (~2%) reduction in fuel-derived carbon emissions;
• A 33% reduction in C3S variance in the first month of operation.
These results were validated by comparing fuel inventories from the Carbon Re on month with those from the previous month before the system was implemented. Since the installation of Carbon Re, the plant targets adjusted to increasing alternative fuel use above its current 86% thermal substitute rate. With Carbon Re adjusted to this objective alternative fuel use increased by an additional 3.5%. The plant has continued to use Carbon Re and is continuing to roll out new features on top of the existing models.
Road to decarbonisation
In conclusion, digital technologies are revolutionising cement plants and setting the stage for a more sustainable future. By embracing a combination of digitalisation and AI, the cement industry is transforming the challenge of inefficiencies and high energy consumption into an opportunity to achieve higher levels of product quality and productivity.
Today, Carbon Re reduces fuel-derived carbon emissions by up to 5% – saving up to 30,000 tonnes of CO₂ per year at a 2Mt/year plant. These efforts not only help our customers meet their emissions targets today but also prepare the industry for the future.
There is still time to make a substantial impact in the fight against climate change, to slow its progression, and in some cases, even reverse its effects. The cement industry has a critical role to play; addressing climate change is both a moral imperative and a smart business decision.
Fuel prices are confidential and fuel cost volatility makes estimating overall fuel cost impact difficult. Therefore, the relative costs above per GJ were used to create a fuel cost index to determine bottom-line impact: RDF: 100% (baseline), Tyres: 120%, Sewage Sludge: 120%, Animal Meal: 200%, Coal: 300%, Heavy Fuel Oil (HGD): 500%, Natural Gas: 500%.
Carbon Re is committed to reducing carbon emissions on a gigatonne scale, driven by a passionate, multi-disciplinary team of Process Engineers and AI Experts who possess a deep understanding of cement production processes. Moreover, Carbon Re’s collaboration with industry leaders like FLSmidth and ABB are expanding access to AI-driven operational savings, empowering more cement producers to address the dual challenges of cutting carbon emissions and managing rising fuel costs. As the industry moves forward, it is clear that leveraging these advanced digital tools is not only beneficial for business but essential for a sustainable future.
Looking forward, AI will be essential for making Carbon Capture and Storage (CCS) viable in cement production. AI-driven process optimisation can stabilise and control flue gas emissions, making CCS more technically and economically feasible. This is a crucial step towards a more sustainable and resilient cement industry and supports the broader transition to CCS.
Any way you look at it, cement plants investing in AI are setting themselves up for a net zero industry and future-proofing operations.
