Investigation and artificial intelligence-assisted optimization of geopolymer-improved soils' mechanical properties with waste tire additives

Geopolymers have recently become recognized as an eco-friendly alternative material for soil improvement. In this thesis, the optimum mixing ratios of geopolymer-improved soils were obtained by an artificial intelligence-supported algorithm, and the effects of waste tire additives on the mechanical properties of soil samples were investigated. A series of experimental applications were carried out to provide data for the optimization function of geopolymers for soil improvement. Considering international guidelines, various chemical and fly ash ratios were used to achieve the target strength, and the investigated soil was brought to different liquid contents. Specimens were prepared and cured for 28 days before conducting unconfined compressive and flexural tests on all obtained combinations. After the curing process of the specimens was finalized, unconfined compressive and flexural strength values were attained. An artificial intelligence-assisted optimization study was performed to reach the target strength considering these values, and the optimum mixture ratios for the soils improved with geopolymer were determined. After the samples were produced at the optimum mix ratios and the optimization process was verified, the waste tire was added to the optimum mix at various ratios to examine its effects on the soil's flexural and unconfined compressive strength. It was observed that using geopolymer and waste tires significantly improved the mechanical properties of the soil. The optimum mixing ratios and the effects of waste tire additives on the mechanical properties of the investigated soil, which can guide the project implementers with an environmentalist approach in soil improvement works using geopolymer, were presented in detail.