Thesis
Design analysis techniques for software quality enhancement
Washington State University
Master of Science (MS), Washington State University
2007
Handle:
https://hdl.handle.net/2376/100900
Abstract
In the software life cycle, early detection and correction of flaws and weaknesses in the design phase can reduce overall costs associated with development and maintenance. Current software development methodologies such as the Model Driven Architecture rely on quality Unified Modeling Language (UML) design models. Often these models are complex and consist of many structural and behavioral views. This can lead to inconsistencies between views. Existing approaches remedy many of these inconsistencies but do not address consistency across design views nor software quality metrics in the design phase. This thesis presents two approaches. (1) The first approach is aimed at detecting and resolving security faults in UML designs. The approach defines the notion of security consistency in designs, analyzes UML views for security inconsistencies, and generates a set of recommended design changes that include Object Constraint Language (OCL) expressions. The OCL can be used as a test oracle in both the design and implementation phases of the software life cycle. This work provides an empirical study that demonstrates that the generated OCL reduces security faults. (2) The second approach evaluates design quality using metrics. During software development it is important for component developers to design components that show high cohesion within a component and low coupling between components. Empirical data shows that software artifacts possessing these properties are easier to develop and maintain. Current practice in design metric evaluation relies on extracting structural metrics from individual UML views. This thesis defines a dynamic approach that collects metrics during execution of a model that integrates both UML Class and Sequence Diagrams. These design metrics are used to evaluate component choices by examining cohesion and coupling properties. The design metrics are based on code metrics that have been positively correlated with maintainability and quality. This thesis provides an empirical study that demonstrates a positive correlation between design and code metrics.
Metrics
5 File views/ downloads
10 Record Views
Details
- Title
- Design analysis techniques for software quality enhancement
- Creators
- Daniel Dee Williams
- Contributors
- Orest Jacob Pilskalns (Degree Supervisor)
- Awarding Institution
- Washington State University
- Academic Unit
- Electrical Engineering and Computer Science, School of
- Theses and Dissertations
- Master of Science (MS), Washington State University
- Publisher
- Washington State University; Pullman, Wash. :
- Identifiers
- 99900525375001842
- Language
- English
- Resource Type
- Thesis