Title: Vulnerability of Networked Infrastructures: Anomalies, Errors, Interdependencies
Citation: CRIS, Third International Conference on Critical Infrastructures
Publisher: International Institute for Critical Infrastructures (CRIS)
Publication Year: 2006
JRC N°: JRC33408
URI: http://publications.jrc.ec.europa.eu/repository/handle/JRC33408
Type: Articles in periodicals and books
Abstract: The paper looks at one critical aspect of the vulnerability of networks: the propagation of errors across systems–of–systems. It proposes an original and relatively simple approach for the fast evaluation of complex system vulnerabilities based on the analysis of their structural complexity, the functional interoperability and the interdependencies among the component systems. The paper includes an application of the model to a transmission network structure taken from a real grid. An error is considered to be a disturbance in the operability of a component. When the disturbance affects a component deviating it from its correct functioning, one has a service failure. The interactions among components are considered as dependencies when a failure in a component can spread to the others connected to it. In this case, the failure of one component is the input error to the next connected one. The net consequence is a pathological succession of errors degenerating into failures, triggering off new errors, in interconnected components. The “health” of the infrastructure has a static view, dependent on the status of its components, and one dynamic dependent on the propagation of the failures. Two criticality indexes are introduced: (i) the Incapability Index, related to the status of components, and, (ii) the Infrastructure Failure Index to characterize the overall system health state. To simulate the response of the infrastructure to the stress of errors and failures, two models are proposed: an Error Propagation Model, with the main goal of expressing the operability pattern of the infrastructure, given a distribution of error injection; and an Influence Flow Model, for tracking the failure propagation on a step-by-step basis. The system’s state of ‘health’ is given by the compound contribution of its components’ capability of providing the global service. A system is considered fully functional when all components present no deficiency in providing the service they are designed for. Nevertheless, a system obviously might be able to function even in presence of local failures. The system is analyzed at two levels: first, at the level of components, and second at the level of the overall system functionality. The first level regards the health status of each of the system components, and its analytical expression is the Incapability Index. The second is based on the compound operability status of all the system components, and regards (i) the level of propagation of errors within the system, and (ii) the severity of the errors (how much an error affects the overall operability of the system). The severity of the error is expressed by the Infrastructure Failure Index and the Weighted Infrastructure Failure Index. A case study is presented that applies the proposed concepts to a national electric power grid. The results have been obtained by means of a prototype software (Infrastructure Security in Electricity Markets – InSIEME) specifically developed for testing the proposed models. The results are believed to confirm the potential usefulness of the approach, as it will emerge from the following.
JRC Directorate:Space, Security and Migration

Files in This Item:
There are no files associated with this item.

Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.