Machine Learning-Enhanced Soil Stabilization Using Multiwalled Carbon Nanotubes and Fly Ash for Road Construction

This chapter delves into the innovative use of multiwalled carbon nanotubes (MWCNTs) and fly ash to enhance soil stabilization for road construction. The study focuses on improving the unconfined compressive strength (UCS) of soil, a critical parameter in pavement engineering. By incorporating machine learning techniques, specifically Classification and Regression Trees (CART) algorithms, the research develops predictive models to estimate UCS values accurately. The integration of fly ash, known for its pozzolanic properties, and MWCNTs, renowned for their exceptional mechanical strength, offers a promising approach to soil stabilization. The chapter also evaluates the California Bearing Ratio (CBR) of optimized soil mixes, providing practical insights into the load-bearing capacity of subgrade soils. The findings highlight the significant increase in CBR values with the introduction of MWCNTs and cement, underscoring the potential of these materials in enhancing soil strength and durability. The study concludes with a comprehensive assessment of the suitability of various soil-fly ash mixes for flexible road construction, guided by IRC:37-2018 guidelines. This research not only advances the field of soil stabilization but also demonstrates the potential of machine learning in optimizing geotechnical engineering practices.