Direct measurement of uniaxial compressive strength (UCS) is time-consuming, expensive, and difficult despite high accuracy. This research aims to indirectly determine the uniaxial compressive strength by statistical and intelligent methods. For this purpose, physical properties, point load index, compressional wave velocity, Schmidt hardness, and uniaxial compressive strength of sandstone samples from Siah Bisheh dam site and Damavand city were measured. Then the performance of different models to estimate UCS based on compressional wave velocity, Schmidt hardness, point load index, and density using multivariate linear regression, random forest algorithm (RFA), K-nearest neighbor (KNN), artificial neural network (ANN) and the adaptive neural fuzzy inference system (ANFIS) were investigated. The results showed that the point load index has the greatest effect on compressive strength among the influencing factors. The results of UCS estimation using intelligent methods showed that RFA provides more accurate results than other investigated methods. The results of this method showed that the value of the model performance index, error (RMSE), and coefficient of determination (R2) are 1.96, 0.02, and 0.99, respectively. The percentage difference between the measured UCS and the predicted average is equal to -0.18%, which shows that the average of the four methods used shows less than one percent difference from the measured average value, and indicates a very high efficiency of these methods for estimating UCS. The results of the Kruskal-Wallis test showed that there is no significant difference between the measured and predicted UCS values.