Abstract
The massive scale of concrete construction constrains the raw materials’ feedstocks that can be considered – requiring both universal abundance but also economical and energy-efficient processing. While significant improvements– from more efficient cement and concrete production to increased service life – have been realized over the past decades through traditional research paradigms, non-incremental innovations are necessary now to meet increasingly urgent needs, at a time when innovations in materials create even greater complexity. Data science is revolutionizing the rate of discovery and accelerating the rate of innovation for material systems. This review addresses machine learning and other data analytical techniques which utilize various forms of variable representation for cementitious systems. These techniques include those guided by physicochemical and cheminformatics approaches to chemical admixture design, use of materials informatics to develop process-structure-property linkages for quantifying increased service life, and change-point detection for assessing pozzolanicity in candidate supplementary cementitious materials (SCMs). These latent variables, coupled with approaches to dimensionality reduction driven both algorithmically as well as through domain knowledge, provide robust feature representation for cement-based materials and allow for more accurate models and greater generalization capability, resulting in a powerful design tool for infrastructure materials.
Highlights
The increasing rate of concrete placement in recent decades – at nearly 3 tonnes/per person/year – has intensified pressure to further reduce the environmental impacts of this essential material while providing growth in global infrastructure necessary to meet the needs of a growing and increasingly affluent population [1]
This paper reviews emerging Machine learning (ML) and statistical approaches for predicting performance of cementitious systems relying on smaller experimental data sets and feature representation informed by domain knowledge of underlying chemistry, physics, and engineering, which represent a significant advantage to training over existing approaches
It had an intrinsic viscosity that was nearly 20 greater than PCE, suggesting its mechanism of plasticization was through solution‐based forces, in contrast to conventional water‐reducers developed for ordinary Portland cement (OPC) systems. This application demonstrates that while Hierarchical Machine Learning (HML) can provide novel predictions and physical insight in optimizing complex systems based on small datasets, it requires analytical representation of all latent variables that drive system responses and surrogate physical measurements to estimate them for the model to be effective
Summary
The increasing rate of concrete placement in recent decades – at nearly 3 tonnes/per person/year – has intensified pressure to further reduce the environmental impacts of this essential material while providing growth in global infrastructure necessary to meet the needs of a growing and increasingly affluent population [1]. In 1998, Yeh et al [3] applied an ANN on a collection of 1030 concrete samples from 17 data sources, where the compositional proportions of cement, supplementary cementitious materials (SCMs), aggregate, water, age, and admixture were features used to predict compressive strength.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
