GSS EU Science Funding Gateway

Telecommunication networks, transport networks, aviation, energy, water grids, finance are the backbone of today’s society. Due to their exceptional complexity and size, infrastructure networks pose a specific challenge when it comes to identifying different risks, either cyber or physical. Especially in the cyber-domain, many intrusions or attacks remain unnoticed or are detected relatively late. Technological developments in areas like machine learning for analytics, user interfaces as well as storage applications have the potential to improve related capabilities.

Modern urban environments and interconnected infrastructures create constantly big amounts of data. In addition, other sources can be exploited to support the identification and analysis of risks to infrastructures. Therefore, research on enhanced risk anticipation through real-time data analysis has the potential to lead to useful tools to enhance preparedness (contingency plans, scenario-based exercises, allocation of resources, etc.).

Resilience of smart cities is marked by a set of specific requirements taking into account most notably aspects from the integration considering user centred approaches as well as social and ethical aspects of Industrial Internet of Things (IIoT), AI/ Machine Learning approaches for real-time data analytics, ensuring transparency, sufficient knowledge and their operational challenges in this area.

While the availability of larger amounts of data from different sources offers potential to improve the identification of possible risks to infrastructures, it also increases the demand for fast and resilient analytical tools. There is a need to filter information to identify data that is relevant as an indicator for risks and – given the large number of different forms of cyber-attacks or intrusions – also a need to prioritise and decide according to the degree of danger they present. This implies the need for matching data in the appropriate context and verifying the source with a view of ensuring that only relevant data is analysed, thus avoiding false results. Faster identification and localisation of hazardous agents and contaminants inside the infrastructure networks is a key to allow for quick response, inform and involve citizens and residents as well as avoid large-scale damage of any incident. Such identification capabilities can be deployed as part of the infrastructure and integrate with the systems public authorities use to make sure information is available as soon as possible. Furthermore, it is crucial to develop methods for better cooperation between different actors to ensure a common understanding and interpretation of data and to provide interactive tools for exchange and visualisation for decision support. Cooperation between different public and private actors is essential in this regard.