A Study of Test Case Generation from Safety Analysis Reports
| Autor: | Wen-Shing Wang, 王文興 |
|---|---|
| Rok vydání: | 2010 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 99 In the past few decades, digital controls are increasingly incorporated into Safety-Critical systems. Digital I&C systems have faster performance, simpler hardware, and better modifiability than analog control systems. However, there are human casualties incurred by failures of hardware or software in digital control systems; especially the failures of the unknown defects from the digital controls. This research aims at automatically generating a large amount of test cases from safety analysis reports to identify such unknown defects so as to improve the safety and reliability of a safety-critical computing system. This research first develops a common semantic model extracted from nuclear regulations, Chapter 15 in NUREG-0800(Standard Review Plan).Then, concepts in the model are represented by XML tags to mark contents in the Safety Analysis report.Then test cases can be generated automatically from the marked report. A large number of test cases can thus be produced. The types of the generated test cases range from testing of each single variable, to testing of each of the use cases and scenarios; the testing techniques used include boundary testing and equivalence partitioning for black box testing. The method developed in this research is general and can be applied to test case generation in other safety domains. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|