A software system for the management of Software Engineering families of experiments: turning a prototype into a fully operational system

Abstract

The software system studied during this thesis was proposed and a prototype was created during the Riofrio’s PhD Thesis “Proceso para la Agregación de Familias de Experimentos en Ingeniería del Software” (Riofrío Ojeda, P. 2019 Universidad Politécnica de Madrid) Empirical Software Engineering is the area of Software Engineering whose objective is to obtain evidence about the methods, techniques, processes and tools involved in software development, this evidence is obtained by conducting empirical studies. In Empirical Software Engineering, it is well known that the results of a single experiment show only part of the information. Obtaining reliable evidence requires combining the results of multiple experiments. One of the approaches to compile a set of related experiments is the family of experiments, referring to the need to perform different replications of the same experiment to consolidate a body of knowledge in Software Engineering. In the families of experiments, there is greater freedom to perform statistical analyses, due to the deep knowledge of the configuration of each experiment and access to the data collected, being able to change the experimental configuration in a planned and controlled way to give an global answer to the hypothesis. The main objective of Riofrío's PhD Thesis is to propose a process for the aggregation of families of experiments in Software Engineering and to facilitate the storage of the information associated with the aggregation process, a web application prototype has been proposed and developed. The prototype developed was scarcely documented and a reverse engineering was needed to generated a software architecture document containing the application context diagram, quality attribute requirements classification and tactics, module view describing display logic and data layer, allocation view, sequence diagrams, class diagrams, entity-relationship diagram and Laravel framework architecture diagram. A software requirements specification document was also needed containing the uses cases, system sequence diagrams and the data dictionary. This reverse engineering was the subject of my EMSE Internship “Experimentation in Software Engineering Application Reverse Engineering”. This EMSE thesis is the continuation of the work done during the internship mentioned above and all the knowledge accumulated in the software architecture and software requirements specification documents would be useful in accomplishing the thesis objetive. Chapter 2 lists some of the collected knowledge, as the data dictionary describing the elements of information managed by the software system, architecture diagrams depicting the context, structure, modularity and physical allocation and use cases and sequence diagram describing the software system features.