Project Classification Using Soft Computing

Project management has been firmly established as a concept for organizing, innovative as well as strategic endeavors. Project managers play a key role in successful project completion. It´s not novel that same project classified as a `simple´ by one project manager may be classified as a `complex´ by others and there are endless reasons. Soft Computing approaches come into picture in such situation. Now a days, Software engineers emphasis on data repository to extract information in order to better manage projects and to produce higher quality software systems that can be delivered on time and on budget. Improper judgment of project complexity, which results into `delayed project´, will occur because of improper estimation of complexity of individual modules of the project. In this paper, each activity diagram of the project is classified into one of the three classes: Simple, average or Complex. Proposed approach is divided into various phases. Pre- Processing phase removes irrelevant information, Feature- Extraction phase extracts various features, Relevance Analysis phase removes irrelevant features. Relevant features and output variable will be fuzzified and used as input to Fuzzy Inference System during the Fuzzy Classification phase.