How do we develop lower-carbon concrete faster, while maintaining performance, quality, and cost?
It’s one of the construction industry’s biggest challenges, and one that increasingly requires a different approach than traditional trial-and-error mix design.
At this year’s fib Congress in Lisbon, one of the world’s leading conferences for concrete engineering and structural design, Ecometrix and RISE Research Institutes of Sweden presented a joint paper exploring how artificial intelligence can fundamentally change the way concrete recipes are developed and optimized.
From Experience-Based Design to Data-Driven Decisions
Concrete mix design has always been a balancing act.
Engineers must simultaneously optimise strength, workability, durability, production cost, and environmental performance, often through multiple rounds of laboratory testing.
As sustainability targets become more ambitious and new materials enter the market, that process is becoming increasingly complex.
The paper presented at fib explores how AI can support engineers by evaluating thousands of potential concrete formulations digitally before physical testing begins, significantly reducing development time while enabling better-informed decisions.
Why Data Matters
Artificial intelligence is only as good as the data behind it.
One of the key findings presented at fib was the importance of building a robust data foundation. The research combines industrial data, published research, and environmental information into a database of more than 45,000 concrete recipes, allowing predictive models to estimate performance, production cost, and carbon footprint from a proposed mix design.
Rather than replacing engineering expertise, AI becomes a decision-support tool that allows engineers to explore more alternatives, faster, and with greater confidence.
From Research to Industrial Application
The research presented at fib builds on the technology behind ACORN, Ecometrix’s AI platform for concrete recipe optimisation.
Initial validation demonstrated that AI-assisted optimisation can significantly reduce the number of physical test iterations while maintaining strong agreement with laboratory results, creating the potential to shorten recipe development by 50–60%.
For an industry facing increasing pressure to reduce embodied carbon while maintaining performance and competitiveness, this represents an important step towards more data-driven product development.
Beyond Concrete
While the paper focuses on concrete, the broader lesson extends far beyond a single material.
Across manufacturing and heavy industry, organisations are discovering that AI creates the greatest value when it is built on trusted data, validated through domain expertise, and applied to real engineering challenges.
That is the philosophy behind ACORN, and behind Ecometrix’s broader approach to industrial AI.
Looking Forward
Presenting this work at fib Congress 2026 was an important milestone, not only for ACORN but for the wider discussion about how AI can support the transition to a more sustainable built environment.
Innovation doesn’t happen through algorithms alone.
It happens when research, industry expertise, and trusted data come together to solve real-world challenges.
We look forward to continuing that journey together with our partners at RISE, the concrete industry, and the wider research community.
About the paper
The paper, ”AI-powered concrete recipe optimization regarding material parameters, cost and environmental footprint,” was presented at fib Congress 2026 in Lisbon by researchers from RISE Research Institutes of Sweden and Ecometrix. The work presents the research and validation underpinning ACORN, Ecometrix’s AI platform for concrete recipe optimisation.
