Prediction of ERP outcome measurement and user satisfaction using adaptive neuro-fuzzy inference system and SVM classifiers approach

Abstract

Nowadays, ERP (enterprise resources planning) system is one of the very crucial and costly projects in the field of information systems for business investment. We report a practical approach which applies both the fuzzy logic analytical model and an expert judgment method support vector machines (SVM) classifier to predict whether the ERP software implementation project succeeds or fail. Here we develop an ANFIS model and SVM model approach, where ANFIS method uses the concept of fuzzy logic to predict the key ERP outcome “user satisfaction” using causal factors during an implementation as predictors which gives the prediction result 5.0000 which is an accurate result in comparison to existing prediction techniques such as MLRA and ANN, where SVM is a binary classifier model which tells about the prediction of good and bad performances of ERP project by dividing the whole dataset into two classes by user-defined condition, where the values of user satisfaction below 5 sets the output value 0 and the user satisfaction value above 5 sets the result 1 which gives the successful prediction results. The main objective of this research is to give satisfaction with user for better prediction results of ERP implementation success.