Can Reclaimed Asphalt Pavement Aggregates be Effectively Utilized in Designing Paving-Grade Geopolymer Concrete? An Experimental Study with Strength Prediction Using Machine Learning Algorithms

Geopolymer concrete (GPC) has emerged as a prominent choice in the construction industry as a sustainable alternative binder. This research delves into the performance of paving-grade GPC by recycling reclaimed asphalt pavement (RAP) as a replacement for natural coarse aggregates. The experimental framework comprehensively examines its effect on the strength and elastic modulus (E), with results portraying more than a 20% reduction in 28?days of compressive strength with a 50% RAP inclusion. Similarly, complete replacement of RAP (100%) significantly affected the E value, revealing a substantial decrease of 46.67%. However, contrary to the existing literature, a Fourier transform infrared spectroscopy study indicated no notable change owing to RAP integration, suggesting a physical interaction at the molecular level. The acquired mechanical properties were used to design pavement thickness, which indicated that, even with a 50% RAP incorporation in the mix, the thickness could be reduced by approximately 9% compared to traditional cement concrete. In addition, an analysis of its impact on the stress ratio revealed an increase with the increment of RAP proportion. To understand the effect of relevant parameters such as NAOH molarity, RAP proportion (%), and curing age on the compressive strength of RAP–GPC, an in-depth analysis leveraging six supervised machine learning algorithms was carried out. The gradient booster validated with a k-fold cross-validation technique demonstrated the highest accuracy for both training and test data sets (89.5% and 96.6%, respectively). The minimal root mean square error (4.18%), mean square error (26.9%), and mean absolute error (4.16%) further substantiated the accuracy of the developed model.

• Record URL:
  • https://doi.org/10.1177/03611981241253612
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/03611981
• Supplemental Notes:
  • Praveen Kumar https://orcid.org/0000-0002-2463-5532© The Author(s) 2024.
• Authors:
  • Ghosh, Ayana
  • R.N, G.D. Ransinchung
  • Kumar, Praveen
  • 0000-0002-2463-5532
• Publication Date: 2024-12

Language

• English

Media Info

• Media Type: Web
• Features: References;
• Pagination: pp 1905-1922
• Serial:
  • Transportation Research Record: Journal of the Transportation Research Board
  • Volume: 2678
  • Issue Number: 12
  • Publisher: Sage Publications, Incorporated
  • ISSN: 0361-1981
  • EISSN: 2169-4052
  • Serial URL: http://journals.sagepub.com/home/trr

Subject/Index Terms

• TRT Terms: Algorithms; Compressive strength; Geopolymer concrete; Machine learning; Modulus of elasticity; Pavement design; Reclaimed asphalt pavements; Thickness
• Subject Areas: Highways; Materials; Pavements;

Filing Info

• Accession Number: 01925969
• Record Type: Publication
• Files: TRIS, TRB
• Created Date: Jul 29 2024 8:13PM