Identifying Conflicts in Security Requirements with STS-ml

Requirements are conflicting when there exist no system that satisfies them all. Conflicts often originate from clashing needs of different stakeholders. Security requirements are no exception to the rule; moreover, their violation leads to severe consequences, such as privacy infringement, which, in many countries, implies burdensome monetary sanctions. In large (security) requirements models, conflicts are hard or impossible to identify manually. In these cases, automated reasoning is necessary. In this paper, we propose a reasoning framework to detect conflicting security requirements as well as conflicts between security requirements and business policies. Our framework formalises the STS-ml requirements modelling language for socio-technical systems. These systems consist of mutually interdependent humans, organisations, and software. In addition to presenting the framework, we apply the it to a case study about e-Government, and we report on promising scalability results of our implementation.