The great added value of Generative Artificial Intelligence (Generative AI, or GenAI for short) lies in the fact that new content can be created quickly and without specialist prior knowledge, simply by following straightforward instructions. This technology is being discussed across various sectors and is finding particular application in areas such as customer support and product personalisation.
Often overlooked, but no less relevant, is GenAI’s contribution to science, for example in materials research. Together with TU Berlin, we have developed a construction chemistry assistant based on language models (LLMs): this tool enables the discovery of new materials without the need for in-depth prior knowledge of construction chemistry.
The production of concrete accounts for 8% of all CO₂ emissions worldwide. To efficiently develop environmentally friendly concrete formulations, we have developed the SLAMD application (watch the short video: SLAMD Explained) in collaboration with the Federal Institute for Materials Research and Testing. This is based on machine learning and helps research staff to develop alternative concretes more quickly and thus more cost-effectively. Read the Champion Story
The Technical University of Berlin has set itself the challenge of making this application available to a wider user group. Whereas chemical knowledge was previously required, SLAMD should now also be usable by people without specialist expertise.
In its original version, SLAMD is based on Sequential Learning (SL), a specific type of machine learning. SL is a method that can operate with very small amounts of data and, step by step, identifies predictions for new concrete formulations by enriching practical results from laboratory tests.
Where traditional machine learning methods fall short, GenAI opens up new possibilities for application:
The result is the Design Assistant, an intuitive chatbot that enables users to find environmentally friendly concrete mix designs even without specialist knowledge.
Faster and more cost-effective access to more environmentally friendly concrete alternatives using state-of-the-art LLMs.
From theory to practice: cutting-edge research is made accessible to a wider user base.
Enhanced user experience in natural language through AI-supported interaction.
I am delighted with this successful collaboration! The integration of LLMs into SLAMD makes materials research in civil engineering even more accessible and opens up new avenues for both our students and our collaboration partners. We are receiving many enquiries regarding the joint use of this tool, and I am eager to see what further opportunities will arise from this interdisciplinary collaboration.
TU Berlin
The project was carried out in collaboration with the ‘Non-Destructive Testing of Building Materials’ research group at the Institute of Civil Engineering at TU Berlin. The group’s focus is on researching resource-efficient and low-carbon building materials and identifying effective pathways towards a circular economy in the construction sector.
