The project has focused on two main research lines. First, we studied the 2016 DDoS attack on Dyn DNS, to understand how domain owners responded to the attack and whether their responses were influenced by specific domain characteristics. We then focused to the geopolitical sphere, analyzing the impact of the Russia-Ukraine conflict on network infrastructures. Secondly, we studied how geo-unblocking - a service used to bypass provider policies based on user location – works. We showed that geo-unblocking services relies on a series of mechanisms (DNS manipulation, TLS proxies) and on a network of residential IPs to bypass the geo-fencing.

Strategic deployment of resources in advance for emergency preparedness is a challenging problem due to inherent uncertainties. A novel approach is proposed, leveraging mathematical optimization and data analytics methods. It is applied to Dutch out-of-hospital cardiac arrest and defibrillator data to reduce response time and increase survival in an emergency.

Many bridges and quay walls in historic city centres, such as in Amsterdam, show signs of overdue maintenance and have reached the end of their technical or functional life. Renovation or renewal is required urgently. This is a large and complex task, not only because of its size, but also because of the impact on the environment and the city. The Loqiquay proposal aims at the development of closed-loop logistics methods and multi-project control solutions to accelerate speed of renovations, increase control, and improving sustainability and circularity including reuse of secondary materials and reduction of transport movements and emissions.

ReAL goes from prototyping to a scalable reinforcement learning platform based on the Open Trip Model (OTM), and made available to the logistics industry (especially SMEs). The platform supports decision-making in logistics processes, is developed/tested by partners from the ICCOS-consortium, and will also be complemented by training materials and workshops.

Synchromodality provides a powerful concept for enhancing the efficiency of freight transportation chains where two or more transportation modes are involved. However, an important factor that complicates the operation of synchromodal chains is the omnipresence of uncertainty in many factors, like weather, traffic congestion, delays or disruptions in any of the (inter-)modal networks, driver behavior, and quality of roads. In this project, the aim is to develop methods and models for exploiting the great potential for synchromodality, addressing the question "what" to transport, "how" and "when". The bundling of shipments so as to balance loads on a corridor in both directions can namely lead to a transition from a chaotic state to streamlined state with more efficiency and at less cost. To this end, we analyze and compare different approaches, including centralized and distributed stochastic assignment problems, and data intensity analysis and their resilience under uncertainty. We evaluate our models in a real-life use case in a Field Lab setting.