Making the (safety) case with the Goal Structuring Notation

Event details

The concept of a safety case – a structured argument, supported by evidence, intended to justify that a system is acceptably safe for a specific application in a specific operating environment – has origins in the nuclear and offshore industries, but is now adopted much more widely, including in the rail sector. In practice, a safety case often contains or references large volumes of documentation, e.g., design files, safety analyses, human factors assessments, and it can be hard to see the argument – to “see the wood for the trees”. The Goal Structuring Notation (GSN) was developed (in the 1990s) as a way of making arguments explicit by providing a graphical structure linking the top-level claim of “safe … in a specific operating environment” to the supporting evidence.

This talk. by Assuring Autonomy International Programme Director, Professor John McDermid, will briefly introduce the concepts of safety cases, explain the concepts behind GSN and outline the notation. It will then go on to explore some principles for constructing good safe cases and illustrate this with some examples taken from a range of industries.

Safety cases are not always effective and have been criticised by some, for example, because they can give rise to “confirmation bias” and the talk will describe a range of problems seen in real-world safety cases. Despite the problems, there is growing interest in applying safety cases to address the risks in a range of applications including autonomous systems and for high-risk residential buildings, following the Grenfell fire. The talk will conclude by discussing these prospects and setting out some of the issues that need to be addressed to make future safety cases fit for purpose and to add value in the management of risk.

This will be an online presentation using GoToMeeting. Details will be sent before the meeting to those that register. Please register before 1700 on 22 July 2021.