Selecting engineering techniques using fuzzy logic based decision support

Abstract

The task of selecting software engineering methods, techniques, metrics, and tools is usually performed manually, based on the expertise of individuals. This paper presents a systematic tool supported approach, that bases its suggestions an the technical situation, the existing goals, and constraints of a specific organization or a particular project. A prototype of the decision support system supports the elaboration of test strategies. The approach uses information about the technical situation that is provided by answering predefined questions with fuzzy data. The objective is to assign "adequacy values" to combinations of test methods, techniques, metrics, tools, and quantified test situations. The priorities of goals and constraints have assessed by applying a technique that is based on comparing goals in pairs. This permits to check certain consistency criteria by static analysis. A hierarchy of the importance of goals and constraints is calculated, which provides the basis for the determination of the suitability of test methods; techniques, metrics, and tools with respect to goals and constraints.