An Approach to Improve Quality of Software Using Metrics and Tachnical Dept- A case study within Model- Driven Environment
Ladda ner
Typ
Examensarbete för masterexamen
Master Thesis
Master Thesis
Program
Software engineering and technology (MPSOF), MSc
Publicerad
2013
Författare
Petersson, Björn
Zhang, Sheng
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
BACKGROUND: As software products play vital roles in embedded systems, software quality has raised much attention in the software engineering field, especially those using modeldriven development. However, software doesn’t have physical features, so it is hard to measure and monitor its quality. OBJECTIVE: To cope with the difficulty of measuring and monitoring software quality, a framework was developed and proposed to Volvo Group Truck Technology (GTT). After studies, analyses and discussions, software metrics and technical debt have been chosen and applied in this framework. METHODS: The framework was developed based on research result from surveys, interviews, workshops and literature reviews. Software metrics were picked and applied in the framework to get basic measurements, the raw data was transferred, analyzed and presented with a modified form of technical debt so to fit this development team’s requirements. Finally, the validity of the study was confirmed by another survey and historical data analysis. The framework is built to fit a model-driven development environment. RESULTS: Team members - including developers, testers, architects and product managers - gave positive feedback to this framework after it was applied in the development environment. Furthermore, statistical analyses carried out on historical data supported the correctness of the framework. CONCLUSIONS: Software metrics can help to analyze, measure and monitor software quality. Furthermore, potential risks could be reduced by improvements suggested in the report, such as i.e. splitting a method because it has too many lines of code or too high complexity. It can be found in this study that by combining software metrics and technical debt, the framework in the project proved to be an efficient support-tool to improve software quality. Furthermore, the framework has the potential to be adopted in other model-driven development teams and environments
Beskrivning
Ämne/nyckelord
Data- och informationsvetenskap , Computer and Information Science