STAR: a tool for analyzing imprecise requirements

Requirement analysis is an important activity in a software development process. Customers usually describe requirements in a natural language, which is often vague. Requirements can be rarely satisfied in a black and white sense. That is, a requirement is usually satisfied to some degree. Existing requirement methodologies are limited in specifying requirements and in supporting tradeoff analysis between the conflicting requirements. A tool support for the requirement process is important since a number of tasks in the requirement process are very time consuming. We describe an on-going project for building a tool that supports requirement specification, trade-off, and analysis. In this tool, we use fuzzy logic to capture the imprecise requirements. Once customers identify some requirement relationships, we are able to infer more implicit relationships between requirements. This functionality is crucial since identification of all requirement relationships is tedious. Moreover, the tool provides the capability of making trade-off between conflicting requirements. We illustrate the tool using a textile spinning frame manufacturing example. The goal of developing such a tool is to automate part of the requirement process such that requirement errors can be reduced in the software development process