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

2026-03-22

AI Summary

This study addresses the challenge of multi-objective co-optimization of strength, workability, durability, and sustainability in concrete mixture design, as well as the limited reproducibility of existing AI-driven approaches. To this end, we propose an open-source probabilistic modeling and Bayesian optimization framework based on Gaussian process regression. Leveraging the first publicly available dataset of concrete mixtures and compressive strength—comprising 123 distinct mix designs and over 500 strength measurements—the framework achieves high-accuracy prediction of compressive strength development curves (R² = 0.94, RMSE = 0.69 ksi) and enables multi-objective optimization for low-carbon, high-strength mixtures. This work presents the first reproducible, open-source dataset and AI-based optimization pipeline for concrete design, integrating uncertainty quantification and explicit trade-offs between strength and embodied carbon to advance green and intelligent concrete formulation.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Constraint Satisfaction and Optimization: Mixed Discrete/Continuous Optimization
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation

Abstract

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.

Problem

• concrete mix design
• compressive strength prediction
• sustainability
• open-source AI
• multi-objective optimization

Innovation

• Bayesian Optimization
• Gaussian Process Regression
• Concrete Mix Design
• Open-Source AI
• Multi-objective Optimization