In the past five years there has been a dramatic increase in work on Search Based Software Engineering (SBSE), an approach to software engineering in which search based optimisation algorithms are used to address problems in Software Engineering. SBSE has been applied to problems throughout the Software Engineering lifecycle, from requirements and project planning to maintenance and re-engineering. The approach is attractive because it offers a suite of adaptive automated and semi-automated solutions in situations typified by large complex problem spaces with multiple competing and conflicting objectives. This paper 1 provides a review and classification of literature on SBSE. The paper identifies research trends and relationships between the techniques applied and the applications to which they have been applied and highlights gaps in the literature and avenues for further research.

Abstract. The need for effective testing techniques for architectural level descriptions is widely recognised. However, due to the variety of domain-specific architectural description languages, there remains a lack of practical techniques in many application domains. We present a simulation-based testing framework that applies optimisation-based search to achieve high-performance testing for a type of architectural model. The search based automatic test-data generation technique forms the core of the framework. Matlab/Simulink is popularly used in embedded systems engineering as an architectural-level design notation. Our prototype framework is built on Matlab for testing Simulink models. The technology involved should apply to the other architectural notations provided that the notation supports execution or simulation. 1 Automatic Testing at the Architecture Level Software testing is an expensive procedure. It typically consumes more than 50 % of the total development budget [1]. Failure to detect errors can result in significant financial loss or even disaster in the case of safety critical systems. Complete testing

www.elsevier.com/locate/jss

Regression testing is an important and expensive activity that is undertaken every time a program is modified to ensure that the modifications do not introduce new bugs into previously validated code. An important research problem, in this context, is the selection of a relevant subset of test cases from the initial test suite that would minimize both the regression testing time and effort without sacrificing the thoroughness of regression testing. Researchers have proposed a number of regression test selection techniques for different programming paradigms such as procedural, object-oriented, component-based, database, aspect, and web applications. In this paper, we review the important regression test selection techniques proposed for various categories of programs and identify the emerging trends. Povzetek: Podan je pregled tehnik izbora testov za regresijsko testiranje programov. 1