This month we are addressing building materials and resources at Reincarnate. After the pre-print paper released by our Data Scientist Researcher Christoph Völker and his colleagues from our partner Bundesanstalt für Materialforschung und – prüfung (BAM) on green building materials and the ecologically informed Sequential Learning approach, we have the honour to interview Dr Völker and find out more about the results, especially the Sequential Learning App for Materials Discovery (SLAMD).
Hello Dr Völker, we are delighted to have this short interview with you today, back to your Convention in Morocco!
Hello, thank you very much for the invitation.
We are going over the topic of building materials and resources this month. Please tell us more about why we need to think out of the box regarding materials in construction.
Well, materials development has not kept pace with the digital revolution that has taken place in the construction industry since the 60s.
This is because of building materials’ complexity and variability, making them difficult to model and analyze. Historically, we have limited the development of new materials to labour-intensive laboratory testing. We see the clear consequences: a significant contribution to CO2 emissions of materials in use today!
The industry is developing alternatives such as secondary raw materials, recycled materials and industrial by-products from various sources and processes. But, using many new raw materials and their combinations, both with each other, with additives and possibly other stimulants, leads to an exponentially growing number of possible compositions. The problem is that their research would exceed the current lab-based experimental exploration capacity!
What is your ecologically informed Sequential Learning approach in all of this?
The approach is a novel way that challenges traditional data-driven materials development practices by prioritizing practical design frameworks over accurate Artificial Intelligence (AI) models.
Well, AI-based materials design methods solve the problem of optimally tuning formulations much more efficiently. This time rather than relying solely on laboratory validation, optimization frameworks such as Sequential Learning and the closely related Bayesian Experimental Design create a data-driven feedback loop: From a variety of possible formulations, the Sequential Learning approach predicts the most promising candidate materials, which are then empirically validated in the laboratory.
So the ecologically informed Sequential Learning approach makes it possible to improve many material properties simultaneously and include critical socioeconomic factors such as carbon footprint, material costs, or resource availability as optimization criteria!
We are curious about the SLAMD application; please tell us more!
Yes, sure. It is the concrete application of the ecologically informed Sequential Learning approach. The objective of this tool is to accelerate materials research in the wet lab through AI. Currently, the focus is on sustainable concrete and binder formulations, but it can be extended to other material classes in the future.
Here you will see a digital representation of resources and processes and their socioeconomic impact; you can calculate complex formulations and enrich them with detailed material knowledge, integrate laboratory data and apply it to novel formulations and tailor concrete to the purpose to achieve the optimal solution.
How does it operate?
In the application, you find a digital lab with three options: the base, the blend and the formulations. The first is the material itself, the second is their combination, and the third is converting your resources into the entire spectrum of possible concrete and binder formulations.
All this data is integrated into the AI Optimization tool that Uses socioeconomic metrics to identify recipes tailored to your requirements. The good news is that the tool is further developed in Reincarnate!
Sounds very promising for the industry! What are the requirements to install this app?
It depends on the operating system, but we have a dedicated section in BAM to guide interested stakeholders in setting it up. So let me share the link here!
Thank you for being with us today.
You’re welcome!
Do you want to learn more about Reincarnate or even contribute to our mission?
Subscribe to our newsletter and follow us on LinkedIn and Twitter, so you don’t miss a thing!