Knowledge-Driven Scenario Development for Critical Infrastructure Protection

Abstract

Scenarios play a central role in developing novel modelling and simulation based systems for Critical Infrastructure Protection. Well-developed scenarios can be used to accelerate system development in different phases including requirements engineering, functional testing and system validation. Developing scenarios, in particular cross-border scenarios, is however a time-consuming task requiring a significant amount of contributions from end-users, system developers and domain experts. Existing approaches for scenarios development are mainly ad-hoc and heavily dependent on texts and diagrams scattered in various documents and data sheets. This introduces more vagueness and ambiguities which are intrinsic in natural languages. Furthermore to fully benefit from modern simulation technology, an extra modelling step is often needed to translate the text-based scenario descriptions into a computer model for domain-specific simulators. Most of the translation work needs intensive interaction between system developers and domain experts and hence can only to be done manually in a tedious and error-prone way. In this paper a knowledge-driven scenario development method is proposed to tackle the problem mentioned above. The core component is a formal knowledge base CISO constructed with Description Logics ($\mathcal {DL}$), which covers the essential elements and their relations of a crisis scenario. Based on CISO a new scenario development workflow is proposed together with a software tool suite to facilitate scenario generation and visualisation. A concrete crisis scenario is presented based on the proposed method to demonstrate its potential for generating precise scenario definitions.