AI-Based Prediction of Strength and Tensile Properties of Expansive Soil Stabilized with Recycled Ash and Natural Fibers

EasyChair Preprint 9743

Abstract

This study investigated the uniaxial compressive strength (UCS) and split tensile strength of a mixture of soil and recycled ash and natural fibers using two different methods, partial least squares (PLS) and classification and regression random forest (CRRF). The study analyzed a dataset of 20 sets with five inputs and two outputs, and the importance of the input parameters was evaluated. The performance of the PLS and CRRF models was assessed, and it was found that the CRRF model outperformed the PLS model. The study also revealed the most and least important parameters in predicting the split tensile strength and UCS in both models. The findings of this study have implications for the use of soil and recycled ash mixtures with natural fibers in construction applications.

Keyphrases: Artificial Intelligence, CRRF, PLS, Recycled ash, Recycled fiber, Soil stabilizer

@booklet{EasyChair:9743,
  author    = {Abolfazl Baghbani and Firas Daghistani and Katayoon Kiany and Mohamad Mahdi Shalchiyan},
  title     = {AI-Based Prediction of Strength and Tensile Properties of Expansive Soil Stabilized with Recycled Ash and Natural Fibers},
  howpublished = {EasyChair Preprint 9743},
  year      = {EasyChair, 2023}}