A.K.M. Moinul Islam; Michael Unterkalmsteiner MSE-2009-16, pp. 176. COM/School of Computing, 2009.
During the last decades, the dependency on software has increased. Many of today’s modern devices embed software to control their functions. The increasing dependency has also taken part in shaping the software development process to produce better quality software. Many researchers and practitioners have spent large investments to improve the software development process. A research area within software engineering that addresses the assessment and improvement issues in
development processes is called Software Process Improvement (SPI).
One of the essential aspects in software process improvement is measuring the outcome of the implemented changes. The measurement and evaluation of software process improvement provides the means for the organization to articulate the achievement level of their goals. Although the importance of measuring and evaluating the outcome of software process improvement is paramount, there exist no common guidelines or systematic methods of measuring and evaluating the
improvement. This condition evokes difficulties for practitioners to implement software process improvement measurement programs. This issue has raised the challenge to develop and implement an effective framework for measuring and evaluating the outcome of software process improvement initiatives.
This thesis presents a measurement and evaluation framework for software process improvement. SPI-MEF provides guidelines in the form of systematic steps to evaluate the outcome of software process improvement. The framework is based on key concepts which were
elaborated in previous work.
In this thesis, a validation of SPI-MEF is also conducted by
involving representatives from academia and industry. The validation is aimed to judge the frameworks’ usability, applicability and usefulness. Finally, a refinement of the framework is carried out based on the input from the validation.