arXiv subjects

Nishant Garg

Publications and source records attributed to Nishant Garg.

5 recordsLinked to original sources

arXiv2026

BOxCrete: A Bayesian Optimization Open-Source AI Model for Concrete Strength Forecasting and Mix Optimization

Modern concrete must simultaneously satisfy evolving demands for mechanical performance, workability, durability, and sustainability, making mix designs increasingly complex. Recent studies leveraging Artificial Intelligence (AI) and Machine Learning (ML) models show promise for predicting compressive strength and guiding mix optimization, but most existing efforts are based on proprietary industrial datasets and closed-source implementations. Here we introduce BOxCrete, an open-source probabilistic modeling and optimization framework trained on a new open-access dataset of over 500 strength measurements (1-15 ksi) from 123 mixtures - 69 mortar and 54 concrete mixes tested at five curing ages (1, 3, 5, 14, and 28 days). BOxCrete leverages Gaussian Process (GP) regression to predict strength development, achieving average R$^2$ = 0.94 and RMSE = 0.69 ksi, quantify uncertainty, and carry out multi-objective optimization of compressive strength and embodied carbon. The dataset and model establish a reproducible open-source foundation for data-driven development of AI-based optimized mix designs.

cs.LG↗

arXiv2023

Sustainable Concrete via Bayesian Optimization

Eight percent of global carbon dioxide emissions can be attributed to the production of cement, the main component of concrete, which is also the dominant source of CO2 emissions in the construction of data centers. The discovery of lower-carbon concrete formulae is therefore of high significance for sustainability. However, experimenting with new concrete formulae is time consuming and labor intensive, as one usually has to wait to record the concrete's 28-day compressive strength, a quantity whose measurement can by its definition not be accelerated. This provides an opportunity for experimental design methodology like Bayesian Optimization (BO) to accelerate the search for strong and sustainable concrete formulae. Herein, we 1) propose modeling steps that make concrete strength amenable to be predicted accurately by a Gaussian process model with relatively few measurements, 2) formulate the search for sustainable concrete as a multi-objective optimization problem, and 3) leverage the proposed model to carry out multi-objective BO with real-world strength measurements of the algorithmically proposed mixes. Our experimental results show improved trade-offs between the mixtures' global warming potential (GWP) and their associated compressive strengths, compared to mixes based on current industry practices. Our methods are open-sourced at github.com/facebookresearch/SustainableConcrete.

cs.LG↗

arXiv2022

Accelerated Design and Deployment of Low-Carbon Concrete for Data Centers

Concrete is the most widely used engineered material in the world with more than 10 billion tons produced annually. Unfortunately, with that scale comes a significant burden in terms of energy, water, and release of greenhouse gases and other pollutants; indeed 8% of worldwide carbon emissions are attributed to the production of cement, a key ingredient in concrete. As such, there is interest in creating concrete formulas that minimize this environmental burden, while satisfying engineering performance requirements including compressive strength. Specifically for computing, concrete is a major ingredient in the construction of data centers. In this work, we use conditional variational autoencoders (CVAEs), a type of semi-supervised generative artificial intelligence (AI) model, to discover concrete formulas with desired properties. Our model is trained just using a small open dataset from the UCI Machine Learning Repository joined with environmental impact data from standard lifecycle analysis. Computational predictions demonstrate CVAEs can design concrete formulas with much lower carbon requirements than existing formulations while meeting design requirements. Next we report laboratory-based compressive strength experiments for five AI-generated formulations, which demonstrate that the formulations exceed design requirements. The resulting formulations were then used by Ozinga Ready Mix -- a concrete supplier -- to generate field-ready concrete formulations, based on local conditions and their expertise in concrete design. Finally, we report on how these formulations were used in the construction of buildings and structures in a Meta data center in DeKalb, IL, USA. Results from field experiments as part of this real-world deployment corroborate the efficacy of AI-generated low-carbon concrete mixes.

cs.AI↗

arXiv2018

Symmetry induced stability in alkali doped calcium-silicate-hydrate

CO$_2$ emissions originating from the construction industry have a significant impact on global warming where the production of ordinary Portland cement clinker is responsible for approximately 8\% of all human-made CO$_2$. Alkali doped calcium-silicate-hydrate (C-S-H) is a critical silicate material since the use of blended cements and alkali-activated materials in construction industry can substantially reduce human-made CO$_2$ emissions. However, the effect of alkali doping (Na and K) on the long-term stability and associated durability of C-S-H remains an open question. Here, using first principles quantum chemistry calculations on the model crystalline phase clinotobermorite, we show that there is a strong interplay between the thermodynamic stability of alkali doped C-S-H and the symmetry of the alkali atoms in the structure. Our results reveal that a symmetrical distribution of alkali atoms leads to a higher stability value such that stable structures with moderate alkali concentrations can be obtained provided that the alkali atoms are allowed to settle into a symmetrical distribution. We investigate the associated structural mechanisms by calculating the migration barriers of alkali atoms within the material, the electronic charge distribution in the material and the variation of basal spacing by using both computational methods and X-ray diffraction analysis.

cond-mat.mtrl-sci↗

arXiv2017

Drying-Induced Atomic Structural Rearrangements in Sodium-Based Calcium-Alumino-Silicate-Hydrate Gel and the Mitigating Effects of ZrO$_2$ Nanoparticles

Conventional drying of colloidal materials and gels (including cement) can lead to detrimental effects due to the buildup of internal stresses as water evaporates from the nano/microscopic pores. However, the underlying nanoscopic alterations in these gel materials that are, in part, responsible for macroscopically-measured strain values, especially at low relative humidity, remain a topic of open debate in the literature. In this study, sodium-based calcium-alumino-silicate-hydrate (C-(N)-A-S-H) gel, the major binding phase of silicate-activated blast furnace slag (one type of low-CO$_2$ cement), is investigated from a drying perspective, since it is known to suffer extensively from drying-induced microcracking. By employing in situ synchrotron X-ray total scattering measurements and pair distribution function (PDF) analysis we show that the significant contributing factor to the strain development in this material at extremely low relative humidity (0%) is the local atomic structural rearrangement of the C-(N)-A-S-H gel, including collapse of interlayer spacing and slight disintegration of the gel. Moreover, analysis of the medium range (1.0 - 2.2 nm) ordering in the PDF data reveals that the PDF-derived strain values are in much closer agreement (same order of magnitude) with the macroscopically measured strain data, compared to previous results based on reciprocal space X-ray diffraction data. From a mitigation standpoint, we show that small amounts of ZrO$_2$ nanoparticles are able to actively reinforce the structure of silicate-activated slag during drying, preventing atomic level strains from developing. Mechanistically, these nanoparticles induce growth of a silica-rich gel during drying, which, via density functional theory calculations, we show is attributed to the high surface reactivity of tetragonal ZrO$_2$.

cond-mat.mtrl-sci↗