Summary Concurrent programming is increasingly present in modern applications. Although it provides higher performance and better use of available resources, the mechanisms of interaction between processes/threads result in a greater challenge for software testing activity. The nondeterminism present in those applications is one of the main issues during the test activity since the same test input can produce different possible execution paths, which may or not contain defects. The test data automatic generation can alleviate this problem, ensuring higher speed and reliability in software testing activity. This paper explores the automatic test data generation for concurrent programs through Genetic Algorithm, a bioinspired optimization technique, and proposes a test data generation approach for concurrent programs, called BioConcST, and a new operator for the selection of test subjects, called FuzzyST, which uses fuzzy logic. The approaches were evaluated in an experimental study towards their validation. The results showed that BioConcST is more promising than the other approaches at all analyzed levels. FuzzyST, together with Elitism and Tournament operators, provided the best results; however, it proved more suitable for concurrent programs of higher complexity.

Software testing activity is imperative to improve software quality. However, finding a set of test cases satisfies a given test criterion, is not a trivial task because the overall input domain is very large, and different test sets can be derived, with different effectiveness. In the context of structural testing, the non-executability is a feature present in most programs, increasing cost and effort of testing activity. When concurrent programs are tested, new challenges arise, mainly related to the non-determinism. Non-determinism can result in different possible test outputs for the same test input, which makes the problem of non-executability more complex, requiring treatment. In this sense, our project intends to define an approach to support automatic identification of infeasible testing requirements. Hence, this proposal aims to identify properties which cause infeasible testing requirements and automate their application. Due to complexity of the problem, we will apply search-based algorithms in the automation of concurrent and sequential programs treatment.

The architectural properties of product‐line architecture (PLA) design have been successfully optimised by multi‐objective genetic algorithms (GAs). Memetic algorithms (MAs) extend GA by adding a local search after the global search process. MA outperformed GA in the context of class modelling, software testing and the next release problem. However, no studies on the application of MAs for PLA design optimisation were found in the literature. In light of this, the authors performed an exploratory study, where MA was used to apply design patterns for a search‐based PLA design (SBPD) and achieved promising results. From the obtained results, they adjusted the MA‐based implementation. This study aims at investigating if the MA is more profitable to SBPD than GAs. Two empirical studies that involve four PLA designs were carried out for this task by varying the pair of objective functions. Empirical results show that MA achieved satisfactory solutions, despite being influenced by the original PLA design.

The Product Line Architecture (PLA) is one of the most important artifacts of a Software Product Line (SPL). PLA design can be formulated as an optimization problem with many factors. In this context, OPLA-Tool was developed to automatically identify the best alternatives for a PLA design using multi-objective evolutionary algorithms, based on genetic algorithms (GA). From an original PLA, OPLA-Tool obtains alternative designs to improve the original one in terms of the objectives selected for optimization. In a recent study, we extend OPLA-Tool to add a memetic algorithm (MA) and promising empirical results were obtained. However, the results allow us to hypothesize that including feature modularization as an objective to be optimized by MA could obtain even better solutions. In addition, it is interesting to know the experts' opinion about PLA designs automatically obtained, what has not been done yet. Thus, the objective of this work is twofold: to investigate the aforementioned hypothesis and to conduct the first qualitative evaluation of the PLA design solutions automatically obtained by search algorithms. An empirical study involving MA and GA was carried out with four different PLA designs. MA presented the best solutions according the quality indicators used in the quantitative analysis. The results of the qualitative evaluation showed that the optimized solutions are well evaluated by the experts. After both analyzes, the hypothesis could be confirmed.