Machine Learning Aided Design and Prediction of Environmentally Friendly Rubberised Concrete

Use of operational research techniques for concrete mix design: A systematic review

Traditional methods for designing concrete mixtures provide good results; however, they do not guarantee the optimum composition. Consequently, applying operational research techniques is motivated by an increasing need for designers to proportion the concrete's raw materials that satisfy the concrete performance requirements such as mechanical properties, chemical properties, workability, sustainability, and cost. For this reason, many authors have been looking for mathematical programming and machine learning solutions to predict concrete mix properties and optimise concrete mixtures. Therefore, a comprehensive review of operational research techniques concerning the design and proportioning of concrete mixtures and a classification framework are presented herein.

Influence of a Novel Carbon-Based Nano-Material on the Thermal Conductivity of Mortar

The paper presents the results of research work to assess the thermal conductivity of mortar incorporating a novel carbon-based nano-material (CBN). The data from the laboratory tests served as the starting point in training an artificial neural network (ANN) based on the Levenberg–Marquardt backpropagation algorithm that was used to predict the values of the thermal conductivity at later ages. The used CBNs were essential precursors of multi-walled carbon nano-tubes but different from their counterparts in the fact that they were capped at the ends. This configuration should result in lower surface tension and should prevent the bundling even without the use of surfactants and sonication. The obtained results show that the mortar mixes with CBN exhibit higher values for the thermal coefficient at early ages compared to the reference mix, even at very low percentages of CBN by weight of cement. The ANN is able to accurately predict the experimental results both at 28 days and at later ages. The obtained results should serve as the starting point for further investigations into the microstructure of cement-based materials enhanced with CBNs.

Modeling of Compressive Strength of Self-Compacting Rubberized Concrete Using Machine Learning

This paper gives a comprehensive overview of the state-of-the-art machine learning methods that can be used for estimating self-compacting rubberized concrete (SCRC) compressive strength, including multilayered perceptron artificial neural network (MLP-ANN), ensembles of MLP-ANNs, regression tree ensembles (random forests, boosted and bagged regression trees), support vector regression (SVR) and Gaussian process regression (GPR). As a basis for the development of the forecast model, a database was obtained from an experimental study containing a total of 166 samples of SCRC. Ensembles of MLP-ANNs showed the best performance in forecasting with a mean absolute error (MAE) of 2.81 MPa and Pearson’s linear correlation coefficient (R) of 0.96. The significantly simpler GPR model had almost the same accuracy criterion values as the most accurate model; furthermore, feature reduction is easy to combine with GPR using automatic relevance determination (ARD), leading to models with better performance and lower complexity.