Complex fuzzy logic reasoning-based methodologies for quantitative software requirements specifications

Abstract

Quantitative software engineering is one of the most important paradigms for software development. That is, Requirements, Analysis, Design, Coding, and Testing. One of the challenges associated with quantitative software engineering is the fact that many of the quantifiable parameters are concomitant with uncertainty. Part of the uncertainty is due to the fact that a significant portion of the software engineering process involves human beings presenting rational, yet difficult to quantify, behavior. Due to this fact, soft computing approaches, specifically fuzzy logic based reasoning, present significant opportunities for constructing sound quantitative software engineering models. This work presents a new and innovative approach for fuzzy logic based quantitative software engineering procedures. We present a complex fuzzy logic based inference system used to account for the intricate relations between software engineering constraints such as quality, software features, and development effort. The new model concentrates on the requirements specifications part of the software engineering process. Moreover, the new model significantly improves the expressive power and inference capability of the soft computing component in the soft computing based quantitative software engineering.