Risk Assessment of Software Projects Using Fuzzy Inference System

摘要

Risk management in software projects plays a vital role in the success of the project. Various risk factors in such projects make it difficult to make reliable and quick decisions in order to accept, mitigate, transfer or reject these risks and obtain an overall view of the whole project. In this paper it is introduced a fuzzy expert system which includes expertise to evaluate risk of software projects in all respects. Fuzzy inference has been used because of its capability in dealing with ambiguity and linguistic variables. Risk factors, the probability of failure and the severity of impact, are very close to fuzzy theory concepts. To develop our fuzzy expert system we deal with a rule base with about 17 million rules. Instead of constructing the whole rule base, a heuristic programming was created to infer the inputs without losing any rules. The output of the model is numerical values which present state of risk for each factor as well as the risk of project called the total risk. The results show better performance compared with traditional risk analysis system. The proposed tool can be used as a decision support system for top management to compare different projects or better risk mitigation in these projects.