Managing Quantitative Requirements in System Safety : A Generalization Based on Three Application Domains

Abstract: Combitech is an independent service company providing technical consultancy in system safety and the part commissioning this project. Dealing with system safety is often an issue of providing requirements in order to prevent the system from constituting danger. System safety also handles the issue of defining requirements to ensure that a pre-defined risk level is satisfied. The risk level is often set by legislative organs with the purpose to ensure that the system is safe enough. When the risk level is communicated in numbers i.e. quantitative requirements, a consequential issue is to provide a logic and consistent methodology. The objective of this thesis is to find an adequate and general approach for the management of quantitative requirements in three different domains. This study involves a multiple case study, in which three domains have been taken into account; railway signaling industry, the air traffic management industry and the defense industry. The used material primarily consists of documents, investigations and standards although a small series of interviews is performed. The study resulted in a model partly conveyed in an UML activity diagram. It was also found that most methodologies require substantial data input, which is what seems to be the main problem in the process of managing quantitative requirements. Keywords: system safety, quantitative requirements, UML, Air Traffic Management, Railway signaling, defense industry.