The voltammograms of lead and tin are strongly overlapped due to their similar 8 properties in electroanalysis, making it difficult for the individual quantification of each analyte. A 9 square wave anodic stripping voltammetry (SWASV) method for the simultaneous determination 10 of the two metals at the bismuth film electrode (BiFE) was proposed. The experimental conditions 11 were investigated and optimized. The linear dynamic ranges and limits of detection (LOD) were 12 determined as 4-150 µg/L and 1.13 µg/L for lead, and 4-110 µg/L and 0.74 µg/L for tin. Radial 13 basis function artificial neural network (RBF-ANN) was introduced to solve the resolution 14 problem, and its performance was compared with the conventional chemometrics methods 15 including principal component regression (PCR), partial least squares (PLS) and least-squares 16 support vector machines (LS-SVM). By analyzing a set of synthetic mixtures of the two metals 17with the proposed methods, the lowest root-mean-square error of calibration (RMSEC) and 18 root-mean-square error of prediction (RMSEP) were achieved with RBF-ANN and the method 19 was applied to real tap water samples spiked with known amounts of lead and tin with satisfactory 20 recoveries of 105% and 102% respectively. Therefore, SWASV coupled with RBF-ANN is an 21 effective alternative used for the simultaneous determination of lead and tin. 22 23