Software Requirements Quality Evaluation: State of the art and research challenges

Abstract

Quality models are important tools for quality management. In software development projects, they are useful as predictive tools for assessing the state of the product being developed and the process used. In order to achieve software quality, a high-quality Software Requirements Specification (SRS) is required. This document is generated at the beginning of a software development project, and is used in all stages. Thus, it is essential to evaluate the quality of the SRS in order to be able to take early corrective and enhancement actions. However, assessing the quality of a SRS is not a simple process, mainly by the multitude of proposals, often contradictory, of the attributes to be evaluated and the methodologies used for that purpose. Thus, it is mandatory to consider proven quality models for guiding this evaluation process. Related to this, this work performs an exploratory analysis of various quality models proposed in this area which can be used as a basis for SRS quality evaluation. Moreover, the work is intended to be a compendium of the most important tendencies and strategies in the field that serves as a starting point for developing comprehensive models and tools for quality attributes evaluation in a SRS.