With SAFETY4RAILS coming to an end, it is time to take a brief look at the scientific perspectives on and contributions to the project. The partners have conducted several studies and analyses related to the resilience of railway infrastructure against cyber-physical threats, and to the development of the SAFETY4RAILS Information System (S4RIS) platform from various perspectives.
Some of the research topics and findings that contributed to the project are briefly described in this blog post.
At the beginning of the project, articles and studies were published on the risk assessment state-of-the-art and risk assessment approaches to lay the foundations for the development of the S4RIS platform. In addition, individual tools contributing to the platform were analysed and described in a couple of studies.
Pasino et al. (2021) analysed and compared state-of-the-art risk assessment methodologies for critical infrastructure, such as mathematical and statistical methods, machine learning techniques, graph and network methods protection, to find the pros and cons of each of them. Their results show that “statistical and mathematical methods provide the most accurate results but need a large amount of data and execution time, while machine learning and complex network approaches work well even if the data are scarce and have a lower computational cost. In addition, the graph and network approaches tend to be the most flexible, able to adapt to every data availability condition and to deal with multiple hazards contemporarily.”
Rajamäki (2021) introduced a conceptualization for resilience management of cyber-physical systems (CPS) to bring the lessons to be learned from earlier projects to SAFETY4RAILS. Five principles for resilience management of CPS are proposed in the article: (1) design and implement a security management plan, (2) employ all appropriate security technologies, (3) ensure the adequacy and quality of security information, (4) make sure that situational awareness is always up to date, and (5) design and implement a resilience management plan that covers all four event management cycles (plan/prepare, absorb, recovery, adapt) and interdependencies with other systems. Furthermore, Rajamäki discussed the meaning of the principles for the rail transportation sector in the article.
Then more detailed approaches to specific tools were presented by different partners. Thomopoulos et al. (2021) contributed to the project by testing and analysing the tool iCrowd to be utilized in the S4RIS platform. At the early stages of the project, they presented testing and demonstration results of the use of iCrowd.
Miller et al. (2021) discussed risk and resilience assessment approaches specific to railway networks. Furthermore, hardware-based counter measures were discussed in detail, also including the hardware security module that was later applied in SAFETY4RALS. The article introduced the S4RIS and a few of the different tools that were planned to be implemented within the S4RIS platform.
As the project proceeded in conducting simulation exercises, Bonneau et al. (2022) further introduced the S4RIS platform and the evaluation and validation methodology that was applied during the simulation exercises in operational environment.
Crabbe et al. (2022) then continued from the work of Bonneau et al. by describing the S4RIS platform progress in their article. It described the architectural solution implemented for S4RIS during 2021 and the demonstration of representative capabilities from the first simulation exercise with Madrid Metro at the beginning of 2022. They stated that, for all tools and resilience phases, the initial end-user evaluation was very positive after the simulation exercise.
Srivastava et al. (2022) focused on modelling of transportation networks with interconnections for criticality and resilience analysis. Specifically, they focused on implementation of mitigation measures and rating them to understand their effectiveness. The analysis was performed using an agent-based simulation tool called CaESAR (Cascading Effects Simulation in Areas for increasing Resilience) in the SAFETY4RAILS integrated platform with detection messages being received from other partners using DMS.
Köpke et al. (2022) assessed the collaboration of two tools in the SAFET4RAILS toolkit: CuriX, a tool for monitoring and detecting abnormal behaviour of infrastructure in the presence of threats, and CaESAR, a tool for assessing propagation of performance losses over distributed systems. The results show that the combination of the main functionalities anomaly detection, cascading effects analysis and an Agent-Based Model (ABM) can contribute to resilience assessments of critical infrastructures.
The increased knowledge and results of the studies and analyses conducted by the partners will support the increase in security and resilience of rail and metro services and transportation services. The risk assessment approaches provided framework for the development of the S4RIS platform, and for further investigation of assessing risks in transportation sector.
The analyses and evaluation of different tools contributing to the S4RIS platform will increase the understanding of the functionalities of technical solutions and tools that could be useful in supporting the security of railway sector. In addition, the results will provide insight and topics for future studies in the resilience assessment of critical infrastructures.
Even though the SAFETY4RAILS project has ended, the partners will continue work on the S4RIS tools development. Also, the project results and further studies will be disseminated in conferences and academic journals.
Crabbe, S., Roß, K., Köpke, C., Faist, K., Villamor Medina, E., Siebold, U., Cazzato, E., Mádi-Nátor, A., Ben-Yizhak, E., Peled, I., Kanak, A., Ugur, N., Halit Ergun, S., Ergun, S., Tiemann, M., Bonneau, M-H., Bourdache, K., Thomopoulos, S. C. A., Kyriakopoulos, C., Panou, K., De Santiago Laporte, A., Matsika, E., David, R., Costa, E., Siino, G., Setunge, S., Mahmoodian, M., Naderpajouh, N., Ottonello, D., Silva, T., Georgakopoulos, A., Giannopoulou, E., Mitrou, M. & Stavroulaki, V. (28 August – 1 September 2022). SAFETY4RAILS Information System platform demonstration at Madrid Metro simulation exercise [Paper presentation]. 32nd European Safety and Reliability Conference (ESREL).
Köpke, C., Walter, J., Cazzato, E., Linguraru, C., Siebold, U. & Stolz, A. (26-30 September 2022). Methodology for resilience assessment for rail infrastructure considering cyber-physical threats [Paper presentation]. CPS4CIP workshop at ESORICS conference.
Miller, M., Satsrisakul, Y., Faist, K., Fehling-Kaschek, M., Crabbe, S., Poliotti, M., Naderpajouh, N., Setunge, S., Ergün, S., Kanak, A., Tanriseven, S., Lekidis, A., Matsika, E., Sick, P. & Cazzato, E. (19-23 September 2021). A Risk and Resilience Assessment Approach for Railway Networks [Paper presentation]. 31st European Safety and Reliability Conference (ESREL).
Pasino A., De Angeli S., Battista U., Ottonello D., Clematis A. (2021). A Review of Single and Multi-Hazard Risk Assessment Approaches for Critical Infrastructures Protection. International Journal of Safety and Security Engineering.
Rajamäki, J. (16-17 June 2021). Resilience Management Concept for Railways and Metro Cyber-Physical Systems [Paper presentation]. 21st European Conference on Cyber Warfare and Security (ECCWS)
Srivastava, K., Köpke, C., Faist, K., Walter, J., Marschalk Berry, J., Porretti, C. & Stolz, A. (26-30 September 2022). Modelling and Simulation of Railway Networks for Resilience Analysis [Paper presentation]. CPS4CIP workshop at ESORICS conference.
Thomopoulos, S.C., & Kyriakopoulos, C. (2021). Anomaly detection with noisy and missing data using a deep learning architecture, Proc. SPIE 11756, Signal Processing, Sensor/Information Fusion, and Target Recognition XXX, 117560R (16 April 2021); https://doi.org/10.1117/12.2589981
Laurea, October 2022