Currently, one of the topical areas of application of machine learning methods in the construction industry is the prediction of the mechanical properties of various building materials. In the future, algorithms with elements of artificial intelligence form the basis of systems for predicting the operational properties of products, structures, buildings and facilities, depending on the characteristics of the initial components and process parameters. Concrete production can be improved using artificial intelligence methods, in particular, the development, training and application of special algorithms to determine the characteristics of the resulting concrete. The aim of the study was to develop and compare three machine learning algorithms based on CatBoost gradient boosting, k-nearest neighbors and support vector regression to predict the compressive strength of concrete using our accumulated empirical database, and ultimately to improve the production processes in construction industry. It has been established that artificial intelligence methods can be applied to determine the compressive strength of self-compacting concrete. Of the three machine learning algorithms, the smallest errors and the highest coefficient of determination were observed in the KNN algorithm: MAE was 1.97; MSE, 6.85; RMSE, 2.62; MAPE, 6.15; and the coefficient of determination R², 0.99. The developed models showed an average absolute percentage error in the range 6.15−7.89% and can be successfully implemented in the production process and quality control of building materials, since they do not require serious computing resources.

The concept of sustainable development provides for the search for environmentally friendly alternatives to traditional materials and technologies that would reduce the amount of CO₂ emissions into the atmosphere, do not pollute the environment, and reduce energy costs and the cost of production processes. These technologies include the production of geopolymer concretes. The purpose of the study was a detailed in-depth analytical review of studies of the processes of structure formation and properties of geopolymer concretes in retrospect and the current state of the issue. Geopolymer concrete is a suitable, environmentally friendly and sustainable alternative to concrete based on ordinary Portland cement (OPC) with higher strength and deformation properties due to its more stable and denser aluminosilicate spatial microstructure. The properties and durability of geopolymer concretes depend on the composition of the mixture and the proportions of its components. A review of the mechanisms of structure formation, the main directions for the selection of compositions and processes of polymerization of geopolymer concretes has been made. The technologies of combined selection of the composition of geopolymer concrete, production of nanomodified geopolymer concrete, 3D printing of building structures from geopolymer concrete, and monitoring the state of structures using self-sensitive geopolymer concrete are considered. Geopolymer concrete with the optimal ratio of activator and binder has the best properties. Geopolymer concretes with partial replacement of OPC with aluminosilicate binder have a denser and more compact microstructure due to the formation of a large amount of calcium silicate hydrate, which provides improved strength, durability, less shrinkage, porosity and water absorption. An assessment of the potential reduction in greenhouse gas emissions from the production of geopolymer concrete compared to the production of OPC has been made. The potential of using geopolymer concretes in construction practice is assessed in detail.

According to the sustainable development concept, it is necessary to solve the issue of replacing fiber from synthetic materials with natural, environmentally friendly, and cheap-to-manufacture renewable resources and agricultural waste. Concrete is the primary material for which fibers are intended. Therefore, the use of vegetable waste in concrete is an essential and urgent task. Coconut fiber has attracted attention in this matter, which is a by-product of the processing of coconuts and makes it relevant. This work aims to investigate the experimental base for the strength properties of dispersed fiber-reinforced concrete with coconut fibers, as well as the influence of the fiber percentage on the mechanical, physical, and deformation characteristics. The samples were made of concrete with a compressive strength at 28 days from 40 to 50 MPa. The main mechanical characteristics such as strength in compression (cubic and prismatic) and tension (axial and bending), as well as the material’s compressive and tensile strains, were investigated. The percentage of reinforcement with coconut fibers was taken in the range of 0% to 2.5% with an increment of 0.25 wt.%. Tests were carried out 28 days after the manufacture. The microstructure of the resulting compositions was investigating using the electron microscopy method. The most rational percentage of coconut fibers was obtained at 1.75%. The increase in mechanical indicators was 24% and 26% for compression and axial compression, respectively, and 42% and 43% for tensile bending and axial tension, respectively. The ultimate strains in compression were raised by 46% and in tension by 51%. The elastic modulus was increased by 16%.

The growth in the volume of modern construction and the manufacture of reinforced concrete structures (RCSs) presents the goal of reducing the cost of building materials without compromising structures and opens questions about the use of environmentally friendly natural raw materials as a local or full replacement of traditional mineral components. This can also solve the actual problem of disposal of unclaimed agricultural waste, the features of which may be of interest to the construction industry. This research aimed to analyze the influence of preparation factors on concrete features with partial substitution of coarse aggregate (CA) with rubber tree (RT) seed shells and to determine the optimal composition that can make it possible to attain concrete with improved strength features. CA was replaced by volume with RT seed shells in an amount from 2% to 16% in 2% increments. Scanning electronic microscopy was employed to investigate the structure of the obtained concrete examples. The maximum increase in strength features was observed when replacing coarse filler with 4% RT seed shell by volume and amounted to, for compressive and axial compressive strength (CS) and tensile and axial tensile strength (TS) in twisting, 6% and 8%, respectively. The decrease in strain features under axial compression and under axial tension was 6% and 5%, respectively. The modulus of elasticity increased to 7%. The microstructure of hardened concrete samples with partial replacement of CA with RT seed shells in the amount of 2%, 4% and 6% was the densest with the least amount of pores and microcracks in comparison with the structure of the sample of the control composition, as well as samples with the replacement of CA with RT seed shells in an amount of more than 6%. The expedient effective replacement of CA with RT shells led to a reduction in battered stone of up to 8%.

Assessment of the mechanical properties of structural steels characterizing their strength and deformation parameters is an essential problem in the monitoring of structures that have been in operation for quite a long time. The properties of steel can change under the influence of loads, deformations, or temperatures. There is a problem of express determination of the steel grade used in structures—often met in the practice of civil engineering or machinery manufacturing. The article proposes the use of artificial neural networks for the classification and clustering of steel according to strength characteristics. The experimental studies of the mechanical characteristics of various steel grades were carried out, and a special device was developed for conducting tests by shock indentation of a conical indenter. A technique based on a neural network was built. The developed algorithm allows with average accuracy—over 95%—to attribute the results to the corresponding steel grade.

Improving the efficiency and quality of construction mainly depends on the cost of building materials, which is about 55–65% of total capital-construction costs. The study aimed to obtain geopolymer fine-grained concrete with improved quality characteristics that meet the construction field’s sustainable development criteria and that have environmental friendliness, economic efficiency, and advantages over competing analogues. The dependences of strength characteristics on various compositions of geopolymer concrete were obtained. It was found that the most effective activator is a composition of NaOH and Na₂SiO₃ with a ratio of 1:2. The increase in the indicators of the obtained geopolymer concrete from the developed composition (4A) in relation to the base control (1X) was 17% in terms of compressive strength and 24% in tensile strength in bending. Polynomial equations were obtained showing the dependence of the change in the strength characteristics of geopolymer concrete on the individual influence of each of the activators. A significant effect of the composition of the alkaline activator on the strength characteristics of geopolymer fine-grained concrete was noted. The optimal temperature range of heat treatment of geopolymer concrete samples, contributing to the positive kinetics of compressive strength gain at the age of 28 days, was determined. The main technological and recipe parameters for obtaining geopolymers with the desired properties, which meet the ecology requirements and are efficient from the point of view of economics, were determined.