Visualization of Software Quality Trends
Ladda ner
Typ
Examensarbete för masterexamen
Master Thesis
Master Thesis
Program
Computer systems and networks (MPCSN), MSc
Publicerad
2012
Författare
Hachilif, Narjis
Pejhanmanesh, Behnoush
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
Background: Designing large and advanced software products puts specific requirements on software quality, in particular in multi-team and agile settings. One challenge is to manage and make decisions on quality efforts when the relevant data have large volume and thus becomes hard to overview, interpret and analyze. Involved staff might not only waste time, if they miss important software build and test outcome information this can result in delays in delivery as well as lost improvement opportunities. If the information can be collected and presented in a coherent and intuitive way it can support software engineers in spotting overall patterns and trends and thus support organizational learning. Problem: A case company, Ericsson AB, was unclear on how to dynamically visualize and present information related to software quality. The objective is to investigate both how data can be intuitively represented and how it can improve efficiency of the employees using it. Previous studies have already investigated these aspects separately, but there is a lack of real-world case studies combining them. Method: Through iterative design research we extended an existing project information portal at the case company to visualize test-related data. In order to achieve this, requirements were collected from initial interviews and workshops with users of the existing system. This data was analyzed along with results from the literature to come up with a proposal for a new visualization method. This led to the design of several prototypes and their evaluation by end users. A set of eight interviews was used to evaluate the final result. Results: The selected prototype was then implemented as a web page displaying both an overview and a history of quality data. The integration of the new page in the existing portal was considered as an important issue to address in order to properly evaluate how this new page could improve the work’s efficiency. Conclusions: We conclude that the proposed implementation provides an intuitive visualization of large quantities of software quality data and improves the current visualization method. It helps visualizing software quality trends by displaying the evolution of results over time. The new visualization can be used both as a tool for quality data analysis at Ericsson AB, but it also provides guidelines for developing efficient software visualization tools in general. Moreover, we conclude that there is a need for further improvements in order to display data in the most efficient and intuitive way possible.
Beskrivning
Ämne/nyckelord
Data- och informationsvetenskap , Computer and Information Science