Simulation based assessment of resilience of two large-scale socio-technical IT networks

Most sectors of modern society are currently undergoing rapid digitalization. With the progress of this development IT networks are becoming an ever more crucial asset whose failure are prone to cause economic losses and risks to health and safety of the population. Due to this development, it is increasingly important that IT networks are designed to be resilient, meaning that they either maintain or quickly recover their functionality when exposed to strains of various kinds. Simulation-based methods that consider supply network topology as well as system for repairing the supply network have previously been applied and found to be useful for assessment of resilience of electricity and water networks. A method of this kind is here applied for IT networks. The method, however, considers only hardware and operator fault modes, not software and dependency faults. The IT networks are studied as socio-technical systems, here broadly understood as systems whose functionality depend on technical as well as organizational sub-systems. The aim of the present paper is to investigate if the method is applicable for assessment of resilience of large-scale IT networks that are critical to society, meaning that (1) it is possible to apply for such systems based on available data, in this case gathered mainly through interviews with focus groups, (2) the results are relevant for users/maintainers/owners of these systems and (3) it captures all relevant factors. The method is tested in case studies on the IT network of a municipality and the SCADA system of a wastewater network. Follow-up interviews were conducted with system managers to assess the perceived relevance of the method and results. It is concluded that the method can be used to obtain the resilience metrics sought for and that the method enables system owners to see if and for what levels of strain they are presently reaching their desired targets concerning system resilience. Follow-up interviews also indicate that the method is perceived by the system managers as being able to support decision making concerning improvement in system resilience. (C) 2018 Elsevier B.V. All rights reserved.