Digitization of the concrete production chain using computer vision and artificial intelligence

The production of concrete currently goes along with pronounced CO2-emissions and an enormous consumption of (mineral) resources. In response to sustainability requirements, concretes thus are increasingly produced using recipes containing six to ten different raw materials including recycled materials and industrial wastes. This increasing complexity results in an increased sensitivity to unpredictable fluctuations in material properties or boundary conditions during the production process. Digital sensor systems and quality control schemes are considered as key to solving this problem, however, digital technologies from other industries have not yet fully established themselves in concrete construction sector, especially in the quality control. Despite the fact that the concrete industry has extremely high repetition factors, big data based quality control is missing, as we currently lack both sensor systems providing data and concrete specific data treatment algorithms. This paper presents an overview on digital methods based on computer vision and artificial intelligence to quantify the properties of concrete raw materials and the fresh concrete along the entire process chain. The methods differentiate between systems that are incorporated into the production process, i.e. in the concrete plant, and systems that are applied after production, i.e. at the construction site. While the first kind enables an online reaction and control of the concrete properties in real-time, namely already during the batch-production, the latter approach allows an offline and, therefore, post-production quality control. All proposed methods eventually contribute to a facilitation of a digital control loop for ready-mixed concrete production. The developed techniques can be easily applied to pre-cast elements production or concrete products.