Use case points method of software size measurement based on fuzzy inference

Abstract

Size measurement is the key element in software development costs and schedule estimation, and the success of a software project directly relates to measurement accuracy. This paper addresses the problem of use case complexity weight hierarchies of discontinuity in the traditional use case points (UCP) method and proposes an improved complexity weight calculation method that utilizes fuzzy theory to analyze the complexity of use cases. First, with use case transactions as input and complexity weight as output, this paper is based on a fuzzy inference system. Then fuzzy rules are established based on the relationship between complexity weights and transactions in use cases. These fuzzy rules can be used to compute the complexity weight. Studies have shown that the proposed method can eliminate discontinuity grades of use case complexity and enhance the accuracy of UCP estimation as well.