Computational & Technology Resources
an online resource for computational,
engineering & technology publications
PROCEEDINGS OF THE FOURTH INTERNATIONAL CONFERENCE ON SOFT COMPUTING TECHNOLOGY IN CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING
Edited by: Y. Tsompanakis, J. Kruis and B.H.V. Topping
Vulnerability Analysis of Road Networks using Nodal Entropy Metrics and Simulation of Threats
S.E. Christodoulou and L. Dimitriou
Department of Civil and Environmental Engineering, University of Cyprus, Nicosia, Cyprus
S.E. Christodoulou, L. Dimitriou, "Vulnerability Analysis of Road Networks using Nodal Entropy Metrics and Simulation of Threats", in Y. Tsompanakis, J. Kruis, B.H.V. Topping, (Editors), "Proceedings of the Fourth International Conference on Soft Computing Technology in Civil, Structural and Environmental Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 22, 2015. doi:10.4203/ccp.109.22
Keywords: road networks, network vulnerability, stochastic simulation, network entropy.
In this paper, a framework for investigating and identifying the locations and areas that affect the overall roadway network vulnerability is proposed. The unbiased metric that is used here for identifying vulnerable locations stands for Shannon's entropy of the network's nodes, since these (contrary to the network links) 'summarize' the users' paths and possibly depict better the path redistribution in cases of network deterioration. The proposed vulnerability framework is composed using a stochastic simulation feature that imposes randomly selected link failures (or closures) and a demand (re)assignment mechanism that provides the users reactions to network alterations. Then the nodal entropy metric is used for the unbiased identification of the most severely affected parts of the network since nodes with increased entropy relates to locations that play an important role in the overall network performance. The proposed framework can provide a comprehensive network mapping of network locations with increased importance, here depicted in network 'heatmaps'. Results of the proposed framework application in a suitable network of realistic size are presented, giving insights on the use of entropy in network vulnerability analysis.
purchase the full-text of this paper (price £20)