Neuro-Fuzzy Model for Project Status Evaluation

Abstract:

The article presents the fuzzy logic as a tool for support to the evaluation of project status. The major goal of this paper is to present the neuro-fuzzy model for evaluating of project status. This model is based on the EVM method. The ANFIS Editor in the Fuzzy Logic Toolbox in MATLAB software was used to create the neuro-fuzzy model. The model consists of two input variables, one rule box and one output variable. The input variables are represented by SPI deviation (SPI) and CPI deviation (CPI). The output from the rule box and the output variable is PS deviation (PS). The advantage of the model is the ability to transform the input indices SPI and CPI into linguistic variables, as well as linguistic evaluation of overall project status (output). With this approach it is possible to simulate the risk and uncertainty that are always associated with real projects. The case study contains real data on the development of values of indices SPI and CPI for one project in the field of IT. The indices SPI and CPI were obtained from control project milestones. There are 5 control milestones in total. ANFIS Editor generated model parameters (parameters of membership functions, rules). These are adjusted on the basis of the data file.