Somaie Jolfaei - Department of Civil Engineering, Faculty of Engineering, University of Zanjan
Ali Lakirouhani - Department of Civil Engineering, Faculty of Engineering, University of Zanjan

Directly determining rock compressive strength is both laborious and expensive. As a result, indirectestimation methods, such as artificial neural networks, are employed. Input parameters, such as quartzcontent, dry density, and Brazilian tensile strength, have been used to predict the compressive strength ofsandstone. genetic algorithm (GA) and particle swarm optimization (PSO) were selected to improvenetwork training using evolutionary optimization algorithms' effectiveness. The results demonstrate thatthe PSO model achieved the best estimation performance with an MSE of ۰.۰۲۱۴ and R of ۰.۹۵. Thelinear regression model exhibited inferior performance with an R of ۰.۸۷.

Sandstone compressive strength, Artificial neural network, Evolutionary algorithms, Genetic algorithm, Particle swarm optimization