- Project Dates: From April 1995 to March 1997
- Sponsors: ESPRIT Copernicus
The project aims to provide techniques for the safety analysis of software, based on formal models and traditional safety analysis techniques, that provide qualitative evidence that the risk associated with the software is acceptable. This evidence must be in a form that can be used in the preparation of a safety case for the software. The migration of traditional safety analysis techniques into software development has the potential to provide a good basis for an integrated approach to the overall safety analysis for both the application domain and the software domain. However, the application of traditional techniques tends to be less effective, and more complex, when applied to software. Examples are the need for templates in software fault trees, and the "modified" keywords of extended HAZOPS techniques. This project will adopt a more fundamental approach and first understand the basic (semantic) notions of the safety analysis techniques, and then investigate how such notions can be combined with the formal techniques currently proposed for the software domain.