Software project risk assessment based on cost drivers and Neuro-Fuzzy technique

Risk assessment plays crucial role in the software project management. The critical examination of different risk assessment methods help researchers and practitioners to evaluate the impact of various project related risks. The existing Fuzzy Ex-COM (Fuzzy Expert COCOMO) model is a combination of fuzzy technique and Expert COCOMO. It takes help of expertise and information from earlier activities carried. Moreover it has limitations that it cannot make room for support from other significant risk rules. The proposed work analyzes the impact of the Artificial Neural Network (ANN) technique for software risk assessment process that integrates the non-linear learning features of neural networks with fuzzy logic having capability to deal with sensitive and linguistic data. It also helps to generate risk rules using ANN techniques to improve the accuracy of risk assessment process. The results shows that this technique with available project data and Neuro-Fuzzy Risk model gives improved results in comparison with existing Fuzzy Ex-com model.