Physical and mathematical models widely used in hydraulic engineering are not always effective in practice. Numerical methods such as the adaptive neuro fuzzy inference system (ANFIS) can be quite useful for estimating the key parameters, especially for complex and multivariate mathematical models. The aim of this study is to determine the discharge coefficients of labyrinth side weirs with two cycles using an experiment setup and ANFIS. Estimation of the discharge efficiency of the side weir has been achieved by using 285 laboratory data sets. The results of ANFIS functions and multiple nonlinear regression (MNLR) analysis were compared with experimental results. It is shown that two ANFIS functions, trimf and gbellmf, give the best estimations, and therefore this technique can be used effectively in practice. (C) 2016 American Society of Civil Engineers.