Model-driven engineering (MDE) is a software engineering paradigm based on the systematic use of models. Over the past few years, engineers have significantly increased the use of MDE, which has been reported as a successful paradigm for developing industrial software. Recently, there have also been remarkable advancements in the Artificial Intelligence (AI) domain, with a significant increase in advanced Machine Learning (ML) techniques. The advances in both fields have led to a surge in works that dwell within the intersection of ML and MDE. This work places the focus on systematically reviewing works that leverage ML to solve MDE problems. We have reviewed a total of 9,194 papers, selecting 98 studies for further analysis. The results of our Systematic Literature Review (SLR) bring light to the current state of the art and trends in the field, discussing the drift in the usage of the different available ML techniques along with the remaining research gaps and open challenges. Our SLR has the potential to produce a positive impact in the research community by steering it towards ML techniques that have been successfully applied to solve MDE challenges.

In industry, software projects might span over decades, with many engineers joining or leaving the company over time. In these circumstances, no single engineer has all of the knowledge when maintenance tasks such as Traceability Link Recovery (TLR), Bug Localization (BL), and Feature Location (FL) are performed. Thus, collaboration has the potential to boost the quality of maintenance tasks since the solution advanced by one engineer might be enhanced with contributions from other engineers. However, assembling a team of software engineers to collaborate may not be as intuitive as we might think. In the context of a worldwide industrial supplier of railway solutions, this work evaluates how the quality of TLR, BL, and FL is affected by the criteria for selecting engineers for collaboration. The criteria for collaboration are based on engineers’ profile information to select the set of search queries that are involved in the maintenance task. Collaboration is achieved by applying automatic query reformulation, and the location relies on an evolutionary algorithm. Our work uncovers how software engineers who might be seen as not being relevant in the collaboration can lead to significantly better results. A focus group confirmed the relevance of the findings.

Feature Location (FL) is a common task in the Software Engineering field, specially in maintenance and evolution of software products. The results of FL depend in a great manner in the style in which Feature Descriptions and software artifacts are written. Therefore, Natural Language Processing (NLP) techniques are used to process them. Through this paper, we analyze the influence of the most common NLP techniques over FL in Conceptual Models through Latent Semantic Indexing, and the influence of human participation when embedding domain knowledge in the process. We evaluated the techniques in a real-world industrial case study in the rolling stocks domain.

Clone-and-Own (CAO) is a common practice in families of software products consisting of reusing code from methods in legacy products in new developments. In industrial scenarios, CAO consumes high amounts of time and effort without guaranteeing good results. We propose a novel approach, Computer Assisted CAO (CACAO), that given the natural language requirements of a new product, and the legacy products from that family, ranks the legacy methods in the family for each of the new product requirements according to their relevancy to the new development. We evaluated our approach in the industrial domain of train control software. Without CACAO, software engineers tasked with the development of a new product had to manually review a total of 2200 methods in the family. Results show that CACAO can reduce the number of methods to be reviewed, and guide software engineers towards the identification of relevant legacy methods to be reused in the new product.

Software maintenance and evolution activities are responsible for the emergence of a great demand of feature location approaches that search relevant code in a large codebase. However, this search is usually performed manually and relies heavily on developers. In this paper, we propose a feature location approach that, instead of searching directly into code from a natural language query as other approaches do, transforms a natural language query to a query that is made up of the requirements that are located as relevant. Furthermore, our approach limits the scope of the code search space by selecting only the code of those products that hold relevant requirements. We evaluate the overall effectiveness of our approach in the industrial domain of train control software. Our results show that our approach improves in 18.1% the results of precision with regard to searching directly into code, which encourages further research in this direction.