摘要：This study describes a methodology for modeling employee commitment prediction by utilizing the application of the Adaptive Neuro-Fuzzy Inference System (ANFIS). For modeling in ANFIS, first, employee commitment dimensions are identified by literature review; second, dimensions are ranked with Artificial Neural Network (ANN) by the importance; and third, selected important dimensions are used as input variable in ANFIS modeling. Statistical population of this study is engineers working on civil projects in North West Iran. After review, job autonomy, superior support, routinization, distributive justice, procedural justice, emotional exhaustion, pay, role conflict, and role ambiguity selected as input variable for commitment prediction model in civil project. Artificial Neural Network selects role conflict, superior support, routinization, distributive justice, and autonomy as the important predictors of employee commitment with suitable estimate accuracy (87.25%). In the following, these variables were used as input variables to the ANFIS model. The training and check data sets are employed in the ANFIS model. ANFIS is trained with the help of MATLAB version R2009a. In this study, the back propagation method and Gaussian curve membership function were used for ANFIS modeling. The experimental result shows that the RMSE values for checking and training data are 0.7263 and 0.7468, respectively. The extracted ANFIS model integrates both neural network nonlinear capabilities and fuzzy logic qualitative approach and has potential to capture the benefits of both in a single framework. Nonlinear relationships between two predictor variables and output variable can be the basis for prediction of any employee commitment in future. The innovative aspect of this research is that this study handles vagueness and ambiguity in the employee commitment problems and tries to build effective model that not only consider quantitative aspects but also convert human judgments about qualitative criteria into meaningful results
