ANALYTICAL STRUCTURE OF A FUZZY LOGIC CONTROLLER FOR SOFTWARE DEVELOPMENT EFFORT ESTIMATION

The major prevailing challenges confronted by Software Projects in cost estimations are due to inconsistent, incomplete, unclear and uncertain project data available in the primary stages of the project. Numerous models have been proposed for constructing a relationship between software size and effort for estimating the cost of project at an early stage. The main task is to find the effort estimates first, that can later on be factored with the salaries of employees to find cost estimates. The literature shows various algorithmic cost estimation models. These techniques do have their own pros and cons. The need for accurate cost estimation for software projects is still a challenge. Most recently, attention has turned towards Machine learning techniques to predict software development cost. It is proved that a few of the problems associated with previous models are addressed by soft computing techniques. The soft computing techniques were more apt when vague and inaccurate information is to be used.. This paper presents an analytical structure of a Takagi-Sugeno fuzzy logic controller with three inputs and one output for software development effort estimation. NASA 93 dataset is considered for validating the model. The analytical study with three inputs has been presented. The Fuzzy Controller developed with triangular membership functions has been proved to give better results for effort estimation. Simulation has been carried out using MATLAB and the results were obtained for various criteria for assessment. The results produced by Fuzzy Model with triangular membership functions with three inputs were very near to actual effort.