The TU Delft Safety & Security Institute aims to strengthen the position and increase the impact of Safety & Security research and education by providing seed funding for relevant transdisciplinary collaborative activities at TU Delft.

The following projects have been granted seed funding in the third call for proposals.


1. Coupling climate, economy and water models

Dr. Jazmin Zatarain Salazar     Multi-actor systems, TPM    

Dr. Pradeep Murukannaiah    Interactive Intelligence, EEMCS

Climate policy and negotiations are being informed by global integrated assessment models (IAMs), which link economic and climate processes into a single framework. These models, however, fail to capture the distribution of risks at regional scales and rely on coarse aggregations of the total cost of climate change.  In order to support a disaggregated analysis within IAMs the applicants are working together to couple global models to regional ones to provide decision support at the relevant temporal and spatial scales.  The goal is to expand the set of test cases to start bridging the gap between global and regional insights relevant to the water domain. This seed funding will contribute to the long-term ambition of developing a suite of open-source modeling libraries for decision support in the context of climate change mitigation and adaptation in water systems. 

The funding will be used to develop an open-source python repository for the development of global integrated assessment models, along with links to the water sector at the river basin level.  Selected case studies in Eastern and Subsaharan Africa will be built for this initial phase as they exemplify hotbeds of climate challenges and water conflicts.  We will build upon the economy-climate modules within integrated assessment modeling to expand the analysis alongside the water modules to assess impacts at a river basin level.   The goal is to make the modules extendible, easily generalizable to other case studies and to enable collaborative development and maintenance.

2.    Assessing the Recoverability of Critical Infrastructural Networks (ARCIN)

Dr. N.Y. (Nazli Yonca) Aydin                     Systems Engineering, TPM                

Dr. O. (Oded) Cats                                   Transport & planning, CEG Robert Kooij                              Network Science, EEMCS

Nowadays, our society depends critically on the proper functioning of various complex networks, such as the internet, the power grid, water distribution systems, and mobile communication networks. Therefore, quantifying the robustness and resilience of such complex networks is of utter importance for managing and ensuring the functionality under disruptions. The framework, recently developed by the Network Architectures and Services Section, Faculty EWI for computing the resilience and recoverability of complex networks using Network Science approach, has only been applied to synthetic instances using pure topological metrics. 
However, this analysis framework has the significant potential to expand on some of the vulnerability analysis techniques deployed in the investigation of transport service and infrastructure and water distribution networks. 
Through the ARCIN project, we aim to jointly supervise students to apply the recoverability framework to two application domains, namely public transportation networks and water distribution networks, using spatial networks. Furthermore, the seed funding will be used for consultation with stakeholders, workshops, and outreach activities with research institutes such as the European Risk & Resilience Institute, Stuttgart, and Humboldt University Berlin to seek collaboration in a joint European project.


3.    GAZE behaviour of road users when inTeracting with an autOmated Vehicle at an intersection (GAZETOAV) H. (Haneen) Farah                    Transport & Planning, CEG

Dr. A.P. (Amir Pooyan) Afghari           Safety & Security Science, TPM

Dr. E. (Eleonora) Papadimitriou         Safety & Security Science, TPM J.C.F. (Joost) de Winter             Cognitive Robotics, 3mE

Recent field experiments and driving simulator studies have shown that road users might adapt their behaviour when interacting with automated vehicles (AVs), known as behavioural adaptation. However, the mechanisms behind such behavioural adaptation are not yet well understood. Studying road users’ gaze behaviour using recent technological advancements and data collection tools can provide additional insight into those underlying mechanisms. In this project we will conduct an experiment to study the gaze behaviour of road users when interacting with an AV at an intersection, using an eye tracking system. We will investigate the impact of several external independent variables (such as external Human–Machine Interfaces displayed on the AV, the recognisability of the AV and its current driving mode, and the presence/absence of a driver in the AV) as well as internal behavioural factors (such as trust, attitudes and receptivity) on road users’ gaze behaviour and their crossing decisions. In addition to the gaze behaviour, the participants will be asked to think-aloud during the experiment to probe what they are thinking, and will be asked to fill in questionnaires related to e.g., their personal characteristics and trust in technology. These factors will then be used to study the association between individual characteristics and the road users’ gaze behaviour and crossing decisions. 


4.    Build that wall! Full-scale testing of a brick wall during floods Davide Wüthrich              Hydraulic engineering, CEG Robert Lanzafame          Hydraulic engineering, CEG        

Ir. Stephan Rikkert                  Hydraulic engineering, CEG Bas Jonkman          Hydraulic engineering, CEG

Dr. Bas Kolen                         Safety & Security Science, TPM Pieter van Gelder    Safety & Security Science, TPM

Climate change and sea level rise will result in more frequent water hazards, affecting coastal and riverine communities. Adaptation to climate change is therefore a key priority and existing structures should be tested under extreme hydraulic conditions. Particularly in the Netherlands, where many buildings are located in the proximity of levees and hydraulic structures, an investigation of the fluid-structure interaction and building response in case of dike overtopping is critical for the development of resilient infrastructure and efficient evacuation plans to guarantee safety to local communities.
This proposal conducts short, but innovative experiments to test the performance of large-scale brick walls under unsteady hydrodynamic loads, typical of flood events. Tests will be conducted in the Polder2C’s facility in Hedwige Prosper Polder, where a variety overtopping tests are executed on an 8 m tall levee with discharges up to 5500 l/m per wave. Results will provide: 1) quantitative measures of the hydraulic and structural performance of a masonry wall under unsteady flow conditions; as well as 2) qualitative information on the effect of the earthen embankment at the soil-structure transition. Both information are critical for assessing the safety and performance of current and future infrastructure. In addition, a bulk part of this project will assess how these technical results on building stability can be integrated in large-scale urban planning, implementing Safety-by-Design approaches and improving our disaster management strategies, with a focus on safety response in case of extreme events.


5.    Towards a multidisciplinary approach for flood safety Martine Rutten                     Water management, CEG      

Davide Wüthrich                          Hydraulic engineering, CEG Bas Jonkman               Hydraulic engineering, CEG Fransje Hooimeijer             BK J.S. (Jos) Timmermans      MAS, TBM Pieter van Gelder        Safety & Security Science, TPM

In a world influenced by climate change, extreme water flows represent a serious threat for coastal and riverine communities. Recent floods in Germany, Belgium and the Netherlands proved conclusively the destructive nature of these events, with severe damages to buildings and critical infrastructure. Unfortunately, similar events are expected to become more frequent in the future, exposing more people to these catastrophes. Therefore, climate adaptation, mitigation measures and flood protection are closely related and pressing challenges that society needs to face at once.
Despite their tragic nature, these floods allowed to conduct fact-finding missions in the Ahr Valley (Germany) and in Limburg (Netherlands), providing a clearer understanding of the dynamic of these events. In particular, the data collected represent a source of knowledge that needs to be further elaborated to provide an understanding of how damages and fatalities can be reduced during future events. Moreover, the multidisciplinary nature of these challenges makes integrated approaches pivotal to explore opportunities and solutions in their full complexity. Thus, within the DeltaFuturesLAB, multiple teams of MSc students are working on different aspects linked to these recent floods. To further enhance this collaborative research and bring it to a larger level, this proposal establishes a framework to support current and future students/researchers on themes related to flood safety, climate adaption, mitigation measures and crisis response. Altogether, the results are expected to provide a more detailed understanding of these recent events through a combined and multidisciplinary approach in assessing the effectiveness of flood protection measures.


6.    Semantic Web for Chemical Process Safety

Dr. Artur M. Schweidtmann   Chemical Engineering, TNW

Dr. Ming Yang                       Safety & Security Science, TPM

Dr. Christoph Lofi                  Web Information Systems, EEMCS

The improvement of safety of (bio)-chemical process is an urgent problem. However, ensuring safety via inherently safer design principles and operational risk assessment of processes is extremely difficult due to increasingly complex interactions between processes, equipment, chemicals, external stimuli (e.g., electricity markets), and operators. Systematic risk analysis methods can derive accident scenarios, estimate accident probability, assess potential consequence, characterize the process risk to support safety-by-design and risk management. However, previous development often relied on expert systems requiring significant manual work.  
Knowledge graphs (KGs) are a promising technology to represent the complex interactions in modern production facilities.  KG combine implicit engineering knowledge (in form of an ontology) with the actual process topology and data allowing for automated reasoning. This has great potential for semi-automated safety analysis. However, process topology information is currently communicated in P&ID diagrams in PDF format. This information is not machine-readable. This lack of structured data is hindering progress in KG-based process risk analysis. Our long-term goal is to digitize existing process P&IDs in order to set up semantic KGs for operational risk assessment.


7.    Roadmap towards Climate & Infrastructure Resilience

Dr. S. (Saba) Hinrichs-Krapels                  Multi-Actor Systems, TPM

Dr. S.M. (Samantha) Copeland                 Philosophy, TPM Bas Jonkman                             Hydraulic engineering, CEG    

Over the past years, amplified by Covid-19, resilience has risen to the top of policy agendas. Correspondingly, there has been an increased interest from non-academic decision-makers in strategic evidence, methodologies, and safe policy test beds for maintaining cities and regions safe and secure from shocks and stresses, including disasters and pandemics. Within the Faculty of Technology, Policy and Management (TPM) and the TPM Resilience Lab we have developed strategies, tools, and models to help decision-makers deal with the disruptions they are facing, working alongside collaborators in the public sector (examples include the UN World Food Programme, Partos, The Hague Municipality, GGDs & Safety Region Haaglanden, and the Ministry of Defence). 
However, despite scientific breakthroughs achieved in these projects, sustained societal impact with these non-academic partners requires further work. Our tools and methods are relatively new, often technical in nature and their uptake takes time. Furthermore, decision-makers are not always aware of when to use which method, nor what our tools can do for them in practice. Therefore, instead of using state-of-the art methods, there is a tendency to fall back on familiar methodologies even when they have known pitfalls. 
Our vision with this seed funding is to enable proactive, collaborative and brokering engagement with these decision-makers, in two sets of activities: 

Requirements analyses workshops 

Four ‘requirement analyses’ co-design workshops with members of the Municipality of The Hague and Rotterdam to (i) identify key challenges they faced in recent disasters, (ii) co-design questions they would want answers in upcoming research projects, and (iii) promote ownership for future projects. These elicited requirements will feed back into discussions with the TU Delft Safety and Security community about the available approaches and tools that best fit the municipalities’ needs. Our combined expertise in risk management, policy implementation, and social resilience will be used to synthesise workshop findings into tangible future research projects that build on existing state-of-the-art research, but also cater to non-academic stakeholder needs. 

Safety and resilience guide

Create a user-friendly, visually-appealing online and printable guide that can help to identify a ‘which method when?’ for communicating the available tools, models and frameworks for approaching safety and security challenges in times of disasters. The first above workshops will inform the content of the guide, and latter ones be used to validate its utility in communicating its content. 

● Workshops (including logistics, event costs, catering) & promotional activities (costs not requested - funded through the TPM Resilience Lab) 

Aims & Audience 

These activities are a crucial first step in creating a stronger alliance with existing networks of project partners, specifically the Municipality of The Hague and Rotterdam, and enable them to become a ‘living lab’ for safety, security and resilience research. The requirements and insights gleaned from these workshops will inform future student projects and more ambitious research proposals (including an ambitious 10 million EURO Horizon Europe 2021 proposal on resilient cities and infrastructures, led by Hinrichs-Krapels and Comes), while the guide will be an engagement tool for bringing in new partners interested in safety, security and resilience research, such as the international community in The Hague, or the global resilience initiatives such as Arup’s Resilience Shift or the Coalition of Disaster Resilient Infrastructures

8.    TU Delft Evacuation Performance: Standardising Assessment

Natalie van der Wal                 Multi-Actor Systems, TPM

Sebastiaan Star                      Manager Campus Safety, Directie ICT & FM

Serge Hoogendoorn               Transport & Planning, Faculty CEG

There currently exists no central detailed reference source for evacuation data; therefore, it is difficult for safety professionals and policymakers to learn from and disseminate best practices and ensure that drills are designed and executed consistently and effectively. Researchers and practitioners need to agree on standard metrics and data structures and establish a common data-sharing platform, ensuring that confidentiality and ethical requirements, regional contexts, and commercial sensitivities are observed. 

Aims and objectives
The project aims are: (1) to achieve consensus on standard metrics and a protocol for evidence-based assessment of occupant evacuation performance at TU Delft Campus (during drills and real-life evacuations) and (2) to share information and best practices on drills and evacuations via an online repository. The objective is to design evacuation drill metrics and data structures for TU Delft Campus. 

This will include: (1) interviewing the emergency response teams of each building on current best practices and bottlenecks; (2) designing standard metrics and data structures for all campus buildings to record evacuations and drills, preparing for data sharing with researchers and safety practitioners, (3) either using existing devices or equipping one to three buildings with movement tracking devices (such as footfall sensors/RFID trackers on people + beacons) in combination with existing or new camera’s; (4) tracking different data and metrics of evacuations and evacuation drills over a period of 6-12 months.

9. SpeakUp! Conversational Agents for Mental Health and Wellbeing

Dr. ir. Marcia Baptista           Air Transport & Operations, AE

Dr. A.Y. (Aaron) Ding            Engineering Systems and Services, TPM Daan Schraven            Design & Integrations, CEG Marijn Janssen      Engineering Systems & Services, TPM 

One of the most important challenges of the Aviation industry is how to combine innovative technologies from Artificial Intelligence (AI) to make everyday life easier. What could AI do to make flying more Efficient, Safe, and Sustainable? Innovative initiatives to maintain aircraft have social, economic, and environmental effects, such as changes in labor productivity, inventory reduction, decrease in flight delays and cancellations due to improved turnaround process, and fuel savings, a byproduct of having a better upkeep fleet. The vision of the Airborne Safety Lab is to help develop the future Edge Computing platform of Prognostics services, connecting engineering devices, processes, and people through advanced AI technologies to collect, share and analyze data. The aim is to turn data into intelligent decisions in an unique way, making maintenance operations smoother than ever. 

The Airborne Safety Lab will strengthen the collaboration of AE, TPM and CiTG to advance the state-of-the-art in aviation safety. Interest is in developing cyber-secure and data-protected distributed computing solutions that can better support the deployment of Artificial Intelligence services for aircraft safety and maintenance. It is also a goal to promote sustainability practices in aviation maintenance. The Airborne Safety Lab builds on active work in predictive maintenance, edge computing, and sustainable practices. The impact of this work will follow from improving the safety and reliability processes of aviation and from raising awareness to the topic. This research can be extended to space applications and air mobility solutions.

/* */