|
Trends such as globalisation and
shorter lifecycles place great demands on the flexibility of the software
industry. In order to compete on an international scale, a decreasing time
to market and an increasing level of quality are essential. However, the
complexity of software systems is increasing, making quality assurance a
difficult task since complex systems present unpredictable global behaviours
that cannot be easily explained from the characteristics of their
components. Testing is at the moment the mostly used quality assurance
technique. However, the development of cost-effective and high-quality
complex systems opens challenges that cannot be faced only with traditional
testing approaches. New techniques for systematisation and automation are
required. Automation of difficult and time-consuming tasks like test case
design seems to be the only way to master the complexity, and develop
quality systems within a competitive amount of time.
Continued below .......
EvoTest is a multidisciplinary project that combines the power of
evolutionary adaptive techniques, inspired by biological evolution, with
software engineering techniques like slicing, program transformation and
reliability analysis in order to find solutions to the problems of testing
software systems and deal with its complexity. Since evolutionary algorithms
are themselves complex systems, exhibiting emergent and adaptive behaviour,
they are ideally suited to the nature of the problem. EvoTest's results will
be evaluated using real case studies from the automotive, telecommunications
and communication industries. This is highly significant since the next
generation of automotive and telecommunications systems will support
communication and negotiation between vehicles and communications devices.
This will lead to complex emergent behaviour. EvoTest is highly timely in
seeking to build upon the lead European science and technology in
evolutionary testing and in applying this to the problems of testing complex
systems.
|
