Aggregating Centrality Rankings: A Novel Approach to Detect Critical Infrastructure Vulnerabilities
Assessing critical infrastructure vulnerabilities is paramount to arrange efficient plans for their protection. Critical infrastructures are network-based systems hence, they are composed of nodes and edges. The literature shows that node criticality, which is the focus of this paper, can be addressed from different metric-based perspectives (e.g., degree, maximal flow, shortest path). However, each metric provides a specific insight while neglecting others. This paper attempts to overcome this pitfall through a methodology based on ranking aggregation. Specifically, we consider several numerical topological descriptors of the nodes’ importance (e.g., degree, betweenness, closeness, etc.) and we convert such descriptors into ratio matrices; then, we extend the Analytic Hierarchy Process problem to the case of multiple ratio matrices and we resort to a Logarithmic Least Squares formulation to identify an aggregated metric that represents a good tradeoff among the different topological descriptors. The procedure is validated considering the Central London Tube network as a case study.
KeywordsCritical infrastructures Criticality analysis Ranking aggregation Analytic Hierarchy Process Least squares optimization
This work was supported by INAIL via the European Saf€ra project “Integrated Management of Safety and Security Synergies in Seveso Plants” (4STER).
- 5.Faramondi, L., et al.: Network structural vulnerability: a multiobjective attacker perspective. IEEE Trans. Syst. Man Cybern. Syst. (99), 1–14 (2018)Google Scholar
- 10.Starita, S., Esposito Amideo, A., Scaparra, M.P.: Assessing urban rail transit systems vulnerability: metrics vs. interdiction models. In: D’Agostino, G., Scala, A. (eds.) CRITIS 2017. LNCS, vol. 10707, pp. 144–155. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-99843-5_13CrossRefGoogle Scholar
- 16.Bozóki, S., Tsyganok, V.: The (logarithmic) least squares optimality of the arithmetic (geometric) mean of weight vectors calculated from all spanning trees for incomplete additive (multiplicative) pairwise comparison matrices. Int. J. Gen. Syst. (2019, to appear)Google Scholar
- 17.Fagin, R., Kumar, R., Mahdian, M., Sivakumar, D., Vee, E.: Comparing and aggregating rankings with ties. In: Proceedings of the Twenty-Third ACM SIGMOD-SIGACT-SIGART Symposium on Principles of Database Systems, pp. 47–58. ACM (2004)Google Scholar
- 24.Oliva, G., Bozóki, S.: Incomplete analytic hierarchy process with minimum ordinal violations. arXiv preprint: arXiv:1904.04701 (2019)