Machine Learning-Based Approach For Optimizing Mixture Proportion Of Recycled Plastic Aggregate Concrete Considering Compressive Strength, Dry Density, And Production Cost
Abstract
The extensive use of plastics has resulted in significant environmental challenges due to non-biodegradable waste. Researchers explore the integration of plastic waste into construction materials, specifically as aggregates in concrete. As the inclusion of plastic aggregates can influence physical and mechanical properties of concrete, a comprehensive analysis of key concrete properties affected by plastic aggregate content should be conducted. Considering the intricate relationship between mixture proportion and properties of plastic aggregate concrete, a random forest (RF) model, a machine learning algorithm, is employed to predict compressive strength and dry density. The RF model, trained on an extensive dataset from diverse sources, demonstrates notable accuracy with low mean absolute percentage errors for compressive strength (7.2 %) and dry density (2.8 %). Moreover, the RF model provides the importance of plastic aggregate content in determining the concrete properties. Subsequently, the trained RF model is applied to optimize mixture proportions, securing compressive strengths of 20 and 30 MPa. The optimization process takes into account factors such as plastic aggregate content, production cost, dry density, and compressive strength. In comparison to experimental results from previous studies, the optimized mixtures exhibit a high strength-to-density ratio, indicating superior structural efficiency.
Recommended Citation
S. H. Han et al., "Machine Learning-Based Approach For Optimizing Mixture Proportion Of Recycled Plastic Aggregate Concrete Considering Compressive Strength, Dry Density, And Production Cost," Journal of Building Engineering, vol. 83, article no. 108393, Elsevier, Apr 2024.
The definitive version is available at https://doi.org/10.1016/j.jobe.2023.108393
Department(s)
Civil, Architectural and Environmental Engineering
Keywords and Phrases
Machine learning; Optimal mixture proportions; Plastic aggregate concrete; Plastic waste; Random forest
International Standard Serial Number (ISSN)
2352-7102
Document Type
Article - Journal
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2024 Elsevier, All rights reserved.
Publication Date
15 Apr 2024
Comments
Missouri University of Science and Technology, Grant None