Integrating Model-Driven Development Practices into Agile Process: Analyzing and Evaluating Software Evolution Aspects

Software use is increasing in different areas of society, and new proposals of development processes have been presented to support this demand focusing on increase productivity and reduce time to market. In this context, some software development processes emphasize source code production, such as agile processes, others focus on modeling, such as Model-Driven Development (MDD). ScrumDDM is a hybrid metaprocess that integrates MDD practices into the SCRUM method aiming to specify software processes instances which models can be used in the agile development context. This paper presents a controlled experiment conducted to analyze the effectiveness of a ScrumDDM instance of its ability to support the agility and the evolutionary aspects of this software process. The results of the experiment showed that the models used in ScrumDDM gave extra support for evolution without compromising the development agility by executing a set of model transformations while preserving project code and documentation updated to support future software maintenance.