The public defense of Sahar Tahvili's doctoral thesis in Computer Science and Engineering

Doctoral thesis and Licentiate seminars

Datum: 2018-12-21
Tid: 13.15
Plats: room Lambda, MDH Västerås.

The public defense of Sahar Tahvili's doctoral thesis in Computer Science and Engineering will take place at Mälardalen University, room Lambda, MDH Västerås, at 13.15 on December 21, 2018.

Title: “Multi-Criteria Optimization of System Integration Testing”.

Serial number: 281

The faculty examiner is Professor Franz Wotawa, Graz University of Technology, and the examining committee consists of Associate Professor Brian Nielsen, Aalborg University; Senior Research Scientist Shaukat Ali, Simula Research Laboratory; Associate Professor Anne Håkansson, KTH Royal Institute of Technology.

Reserve; Adjunct Professor Johan Åkerberg, Mälardalen University

Abstract:

The role of software is important in our daily lives and to the progress of society in general. Over the past decades, improving the quality of software products has become a unique selling point for software companies. Achieving high quality software products is possible through a balanced integrative approach of design and verification activities during the software development life cycle (SDLC) process. Considering the mentioned facts, software testing becomes a major player in the product development process, which can satisfy the end users’ needs and also ensure high quality of the final product.

Software testing is performed manually or automatically and the transition from manual to automated testing and continuous changes in the testing procedure have quickly become widespread at industry. However, human intuition, inductive reasoning and inference cannot be fully covered by the current form of test automation thus, manual testing still plays a vital role in software testing. Software testing is often performed on several levels, such as unit, integration, system and acceptance levels. Proposing an appropriate testing procedure (either manual or automated) depends on several parameters such as the quality of requirements, the size and complexity of the product and also the testing level.

Integration testing can be considered as the most complex testing phase in some practical scenarios. Integrating the unit tested modules and testing their behavior as an integrated unit can result in huge increase in complexity. Integration testing is usually performed after unit testing (and before system testing), when all modules have been tested and approved separately.

To test a product manually, a set of test case specifications need to be created. A test case specification describes textually a testing goal and how the product should behave at specified input parameters. Usually a large set of test cases is required to test a product. Executing manually all test cases for a product requires time and also testing resources. Therefore, optimizing and improving software testing process has received much attention over the last decade form researchers in academia and also in industry.

In this doctoral thesis we propose a set of optimization techniques for test case selection, prioritization and scheduling for manual integration test cases. All proposed approaches in this thesis have been evaluated on industrial testing projects at Bombardier Transportation (BT) in Sweden.