faculty: "FNWI" and publication year: "2005"
| Author||Jonne Kats|
|Title||The Software Project Leader's Cockpit|
|Supervisors||P. Klint, L. Hofland, R. Krikhaar|
|Faculty||Faculty of Science|
|Institute/dept.||FNWI: Instituut voor Informatica|
|Programme||FNWI MSc Software Engineering|
|Abstract||The software project leaders at the Philips Medical Systems - Magnetic Resonance development department collect numerous of measures. However, from one of their most important business assets: the code base, they do not collect much. This thesis describes the results of the definition of a measurement programme, which goal is to define measures that can be extracted from the code base and can help the software project leaders in managing their projects.|
Implementing the measurement programme will cost effort and therefore it is important to know why there is being measured. The Goal-Question-Metric approach helps in defining these goals. The requirements on the measurement programme have been gathered by the use of interviews, the results of these interviews have been structured using this Goal-Question-Metric approach. This resulted in measures that have the following goals, improving project control and reducing the number of faults discovered in the field.
During the project research was done to measures that are being used by other companies. The most of these measures focused on bug reports, time sheets, effort estimations, etc. Other research was done to support ideas resulting from the Goal-Question-Metrics analysis. The research was related to the results from the Goal-Question-Metrics analysis in a GAP analysis, most of the goals were supported by the literature except for two goals. However, these goals could be supported by normal logic and did not need additional literature. Prototyping was used to determine the feasibility and usefulness for some of the measures, a metrics called entropy appeared not to be applicable at the MR organization. A merge estimation measurement and an activity parameter appeared to be useful.
Implementing the measures at the MR department will require effort, because it has to be made part of the daily process. However, implementing the measures resulting from this project require minor effort compared to the benefits they deliver. Additional research should be done to ideas that could possibly improve the measures that resulted from this project, but a working tool implementing the measures should not take to much effort.
|Document type|| scriptie master|
Use this url to link to this page: http://dare.uva.nl/en/scriptie/411024
Contact us about this recordNotify a colleague
Add to bookbag