Document Type : Original Article
Department of Civil Engineering, Iran University of Science and Technology, Tehran, Iran
The complex characteristics of cohesionless soil texture necessitate an examination of settlement simulation in cohesionless materials, making it a fundamental area of inquiry. This article discusses the development of integrated data intelligence algorithms with the objective of improving the reliability and accuracy of estimate results for shallow foundations (S_m) on cohesionless soils. The proposed models integrate the reptile search algorithm (RSA) with support vector regression (SVR) analysis and adaptive neuro-fuzzy inference system (ANFIS). Based on the findings of the research, it can be seen that the RSSVR and RSANF systems have shown adept capabilities in the domain of estimate. During the training stage, the RSANF simulation gained the smallest performance index (PI) value of 0.0668, which was smaller than the PI of 0.0993 for RSSVR. Similarly, during the test stage, the RSANF simulation acquired a PI of 0.0904, which was fewer than the PI of 0.1038 related to RSSVR. The constituents mentioned above offer a new approach to enhance the precision of forecasting models and further our comprehension of predicting the S_m.