The use of neural networks for the prediction of the settlement of pad and one way strip footings on standard penetration test results

The two major criteria that control the design of shallow foundations on cohesionless soils are the bearing capacity and settlement criteria. The settlement criterion is usually more critical than the bearing capacity criterion in the design of shallow foundations on cohesionless soils. Thus, settlement criterion usually controls the design process, rather than bearing capacity criterion.In this study, artificial neural networks (ANNs) were used to predict the settlement of a single or pad footing and one-way strip footings, without a need to perform any manual work, such as; using tables or charts. With this purpose in mind, a computer programme was developed in the Matlab programming environment to calculate the settlement of pad footings from five traditional settlement prediction methods, such as; Burland and Burbidge (1963), Meyerhof (1965), Terzaghi and Peck (1967), Pary (1971), Peck et al. (1974). The footing geometry (length, L, and width, B), the footing embedment depth, Df, the bulk unit weight, ?n, of the cohesionless soil, the footing applied pressure, Qk, and standard penetration test result, SPT-N, varied during the settlement analyses and the settlement value of each pad footing and one-way strip footing was calculated for each method by using the written programme. Then, an ANN model was developed for each traditional method to predict the settlement by using the results of the analyses. The settlement values predicted from the ANN model were compared with the settlement values calculated from the traditional method for each method. The predicted values were found to be quite close to the calculated values. Moreover, several performance indices such as determination coefficient (R2), variance account for (VAF), mean absolute error (MAE), and root mean square error (RMSE) were calculated to check the prediction capacity of the ANN models developed. The obtained indices make it clear that the constructed ANN models have shown high prediction performance. It has been demonstrated that the ANN models developed can be used at the preliminary stage of designing pad and one-way strip footings on cohesionless soils, without a need to perform any manual work, such as; using tables or charts. Sensitivity analyses were also carried out to examine the relative importance of the factors affecting settlement prediction.