Options
2010
Doctoral Thesis
Titel
The SCOPE approach for scoping software processes
Abstract
Having software processes that fit technological, project, and business demands is one important prerequisite for software-developing organizations to operate successfully in a sustainable way. Therefore, process management needs to constantly evaluate the organization's processes to see whether they match the demands. In doing so, not only should the current demands be considered; future demands should also be anticipated and proactive support should be provided. Currently, there is little methodological support for selecting and evolving an organization's processes based on identified current and anticipated future demands. This often leads to unfocused development processes with unnecessary variability that only inadequately support current demands, and whose evolution to meet future challenges is very difficult. In order to address this problem, a solution should actively identify an organization's current and anticipated future demands. It should therefore analyze its products and projects in terms of their process demands, using existing knowledge where available. Based on this analysis, it should provide explicit support for defining and evolving the organization's processes, and be readily applicable in industrial practice. This thesis introduces the SCOPE approach for continuously focusing an organization's processes towards the relevant demands. In particular, the SCOPE approach facilitates (a) the determination of the process demands of current and future products and projects, (b) the analysis of processes for satisfying these demands, (c) scenario-based selection strategies that support decisions as to which processes to keep, which to discard, and how to prioritize process modifications, and (d) a prototype tool supporting the overall approach. The approach was validated by means of one controlled experiment and two industrial case studies. The validation was aimed at showing that SCOPE (a) allows for a greater reduction in unnecessary process variability than ad hoc selection; (b) allows for selecting processes that cover a broader range of demands than ad hoc selection; (c) allows for reducing process management effort compared to ad hoc methods; and (d) is fit for industrial application. The validation showed that (a) the application of the SCOPE approach in a controlled experiment led to a 46% reduction in unnecessary process variability compared to ad hoc approaches and allowed for an 82% reduction in an industrial case study; (b) SCOPE users identified 150% more misalignments between processes and demands in a controlled experiment than when working ad hoc; (c) the application of the SCOPE approach allowed for a reduction in process management effort of 83% in a controlled experiment and of 41% in an industrial case study; and (d) the SCOPE approach and results were accepted by the engineers in two industrial case studies as a means for providing adequate support for process management.
ThesisNote
Zugl.: Kaiserslautern, TU, Diss., 2010