Filter results

47626 results

Young Scientists on stage

Martijn Nagtegaal, PhD student @ImPhys participated in the "Hoezo? Show!". The Hoe?Zo! Show is an interactive science game show in which young scientists answer questions from curious children live on stage in a race against time. Will the scientists be able to answer children's questions quickly enough? Or do the children's questions prove too difficult for these clever minds?

Cyber Security Webinar by Roland Kromes MSc - Modeling a low-power IoT architecture for Blockchain and Smart Contract applications

Cyber Security Webinar by Roland Kromes MSc - Modeling a low-power IoT architecture for Blockchain and Smart Contract applications 23 November 2021 12:00 till 12:45 - Location: Zoom meeting Join Zoom Meeting https://tudelft.zoom.us/j/93459336716?pwd=VFk4N3dCQTF2TUlHYnBLejk0dFVPUT09 Meeting ID: 934 5933 6716 Passcode: 389654 Abstract Nowadays, most IoT applications are based on a centralized system in which all of the system participants have to rely on a central entity. In such a system, data immutability, data traceability, and transparency cannot be provided. Blockchain technology is an entirely decentralized system in which the third trusted party (central entity) is removed. Contrarily to centralized systems, blockchain technology provides data immutability, traceability, and transparency. Most modern blockchains also allow the deployment of smart contracts, which are digital programs that can be read by all participants and executed automatically according to an event on the blockchain. These advantageous features of blockchain technology show a clear interest in the integration of IoT with blockchain technology. The contribution's main objectives are the study of the integration possibilities of IoT with blockchain technology and the development of a model of dedicated low-power-consumption IoT hardware architecture that enables communication with multiple types of blockchain. The proposed blockchain APIs (Ethereum, Hyperledger Sawtooth) are written in C++. These APIs can create valid transactions. The IoT architecture model's CPU is emulated with QEMU and the dedicated cryptographic hardware accelerators are modeled in SystemC-TLM high-level hardware description language. As QEMU and SystemC-TLM work on different time environments, they must be synchronized by a co-simulation platform. The CPU of the proposed IoT architecture executes a Linux Operating System, which runs blockchain APIs. These APIs cannot have direct access to the hardware accelerators. In order to handle access to the hardware accelerators and the orchestration of dedicated power management, Linux Kernel device drivers were developed. The results represent that a significant reduction of the overall energy consumption can be achieved when the elliptic curve point multiplication and hash operations are hardware accelerated. Bio I am Roland Kromes, a PhD candidate at the Université Côte d'Azur in the Electronics, Antennas and Telecommunications Laboratory (LEAT-CNRS). In January, I will join the Cybersecurity group at Delft University of Technology as a postdoctoral researcher. I obtained my diploma in Electronics, Systems and Telecommunications with the specialty of embedded systems at the Université Côte d'Azur. My thesis focuses on the possibilities of IoT integration with Blockchain technology and the modeling of a specific low power IoT architecture for Blockchain and Smart Contract applications. This thesis is part of the Smart IoT for Mobility multidisciplinary project (ANR-National Agency for Research) in which I could also work on acceptability issues of Blockchain technology with fellow management researchers. My research interests: Blockchain, IoT, applied cryptography, secure data sharing.

Cyber Security Webinar by Dr. Yunpeng Li - Uncertainty quantification and propagation in high-dimensional spaces

Cyber Security Webinar by Dr. Yunpeng Li - Uncertainty quantification and propagation in high-dimensional spaces 28 September 2021 12:00 till 12:45 - Location: Zoom meeting Join Zoom Meeting https://tudelft.zoom.us/j/99314357068 Meeting ID: 993 1435 7068 Passcode: 678062 Abstract: Deep learning models, which are cornerstones in the success of modern machine learning applications, often lack representation of uncertainty or produce overconfident predictions that can lead to costly consequences. The optimal transport theory has found it application in quantifying uncertainty in large-scale machine learning problems by measuring the Wasserstein distance between data distributions. Separately, for real-world sequential inference applications, differentiable filters provided a mathematically principled framework to propagate model uncertainty. This talk will hence be divided into two parts. In the first part, I will discuss a new family of slice-based Wasserstein distance metrics, called augmented sliced Wasserstein distances (ASWDs), with a novel incorporation of injective neural networks. ASWDs learns nonlinear projections that can capture the complex structure of the data distributions which improve their projection efficiency. In the second part, I will discuss our recent efforts in constructing more expressive dynamic model and proposal distributions in the differentiable particle filtering framework through normalizing flow. In addition, I will introduce an end-to-end learning objective based upon the maximisation of a pseudo-likelihood function which can improve the estimation of states when large portion of groundtruth information are unknown. Bio: Dr Yunpeng Li is Senior Lecturer in Artificial Intelligence in the Department of Computer Science at the University of Surrey in the UK. Before joining Surrey as a Lecturer in AI in 2018, he was a postdoctoral researcher in the Machine Learning Research Group in the Department of Engineering Science at the University of Oxford and was a Junior Research Fellow at Wolfson College at Oxford. He received a PhD in Electrical Engineering at the McGill University in Canada in 2017. His research interests are in the areas of statistical machine learning and signal processing, particularly Bayesian inference techniques and the optimal transport theory. He has broad interests in applications of machine learning, e.g. breast cancer detection, dental disease detection, and environment acoustic sensing. His work has won the best paper award in the NeurlPS Workshop on Machine Learning for Developing World in 2018 and 2019.

Cyber Security Webinar by Martin Fejrskov MSc - Using NetFlow to measure the impact of deploying DNS-based blacklists

Cyber Security Webinar by Martin Fejrskov MSc - Using NetFlow to measure the impact of deploying DNS-based blacklists 12 October 2021 12:00 till 12:45 - Location: Zoom meeting Join Zoom Meeting https://tudelft.zoom.us/j/95465881342 Meeting ID: 954 6588 1342 Passcode: 808322 Abstract To prevent user exposure to a wide range of cyber security threats, organizations and companies often resort to deploying blacklists in DNS resolvers or DNS firewalls. The impact of such a deployment is often measured by comparing the coverage of individual blacklists, by counting the number of blocked DNS requests, or by counting the number of flows redirected to a benign web page that contains a warning to the user. This paper suggests an alternative to this by using NetFlow data to measure the effect of a DNS-based blacklist deployment. Our findings suggest that only 38-40\% of blacklisted flows are web traffic. Furthermore, the paper analyzes the flows blacklisted by IP address, and it is shown that the majority of these are potentially benign, such as flows towards a web server hosting both benign and malicious sites. Finally, the flows blacklisted by domain name are categorized as either spam or malware, and it is shown that less than 6\% are considered malicious. Short bio: Martin Fejrskov is an Industrial Ph.D. student at Telenor Denmark and Aalborg University focusing on detecting cybersecurity threats with data from Internet Service Providers. Prior to the Ph.D. studies, he was a Solution Architect at Telenor Denmark, focusing on security services and the national backbone network. Before that he was a Product Manager at the Danish company formerly known as ETI A/S, heading the development and architecture of one of the core products. He received his Master of Science from Aalborg University in 2005 with excellent grades, studying primarily within the fields of network protocols, traffic analysis and security.

Cyber Security Webinar by Dr. Shihui Fu of the University of Waterloo - Transparent Succinct Zero-Knowledge Arguments for R1CS with Efficient Verifier

Cyber Security Webinar by Dr. Shihui Fu of the University of Waterloo - Transparent Succinct Zero-Knowledge Arguments for R1CS with Efficient Verifier 09 November 2021 12:00 till 12:45 - Location: Zoom meeting Join Zoom Meeting https://tudelft.zoom.us/j/95108026354?pwd=RVhXS3EwQjIyZnVzNk1ScWNzQmtTdz09 Meeting ID: 951 0802 6354 Passcode: 378365 Abstract We propose Polaris, a zkSNARK with quasi-linear prover time and both polylogarithmic proof size and verification time in the size of the arithmetic circuit representing the statement. By instantiating with different commitment schemes, we obtain several zkSNARKs where the verifier's costs and the proof size range from $O(\log^2{N})$ to $O(\sqrt{N})$ depending on the underlying polynomial commitment schemes. All these schemes do not require a trusted setup. It is plausibly post-quantum secure when instantiated with a secure collision-resistant hash function. Our experimental evaluation demonstrates that Polaris offers a much lower verification time than Ligero and Aurora for instances with large sizes as we reduce the complexity from linear to polylogarithmic. Short bio Shihui Fu received his Ph.D. degree in Mathematics from Academy of Mathematics and Systems Science, CAS in 2018. Currently, he is a post-doc in University of Waterloo. His main research interests include cryptographic protocols and zero-knowledge proof.

Cyber Security Webinar by Rui Wang MSc - Privacy-preserving Byzantine-Robust Federated Learning

Cyber Security Webinar by Rui Wang MSc - Privacy-preserving Byzantine-Robust Federated Learning 07 December 2021 12:00 till 12:45 - Location: Zoom meeting Join Zoom Meeting https://tudelft.zoom.us/j/93851661903?pwd=UzhXVlAxellBUDhRRXdyNWYyOGdlUT09 Meeting ID: 938 5166 1903 Passcode: 931813 Abstract Privacy and Byzantine-robustness are two major concerns of federated learning (FL), but mitigating both threats simultaneously is highly challenging: privacy-preserving strategies prohibit access to individual model updates to avoid leakage, while Byzantine-robust methods require access for comprehensive mathematical analysis. Besides, most Byzantine-robust methods only work in the honest-majority setting. In this webinar, I will introduce previous works about this topic and presents some expected experimental results about Privacy-preserving FL to resist > 1/2 Byzantine attacks.

Cyber Security Webinar by Daniël Vos Msc - Training Robust Decision Trees Against Adversarial Examples

Cyber Security Webinar by Daniël Vos Msc - Training Robust Decision Trees Against Adversarial Examples 14 September 2021 12:00 till 12:45 - Location: Zoom meeting Join Zoom Meeting https://tudelft.zoom.us/j/94484315617 Meeting ID: 944 8431 5617 Passcode: 049126 Abstract Decision trees and tree ensembles are among the most popular machine learning models for tabular data but they suffer from adversarial examples. When we make specific changes to a model's input, the resulting prediction can be completely wrong. Since this finding, researchers have proposed different methods to train trees that are robust so such attacks. However, these methods can take hours to run and offer no performance guarantees. In this seminar we will look at how to speed up training by orders of magnitude and explain how we can prove the optimality of robust decision trees.

Research lines

To serve society, the focus of the group is on two main research lines: assessment of existing concrete structures The group is famous for the research on failure mechanisms, especially shear failure, development of non-destructive techniques such as acoustic emission for damage detection (monitoring), and on assessing and developing guidelines for proof loading of existing structures. upscaling of innovative types of concrete Group activities are focused on developing safe procedures for the structural application of new types of concretes in order to serve societal demands. For example, using Ultra High Performance Fibre Reinforced Concrete (UHPFRC) for developing a fast and safe procedure for the replacement of existing bridges, using UHPFRC and bendable concrete (SHCC) for strengthening and repair or “green” (geopolymer) concrete as more sustainable building material. Current interest of the group is to investigate and develop a guideline for a safe structural application of these materials in reinforced and pre-stressed elements and structures.

Research

The mission of the group is to perform the research directly serving societal challenges. Two main challenges in the area of concrete structures are (firstly), how to maintain and deal with the existing structures, and (secondly), how to keep up with the recent developments and apply innovative concretes in structures in a safe and sustainable way. Therefore, the research activities of the group are mainly focused on two research lines: assessment of existing concrete structures, and upscaling of new concrete materials. Most research projects are performed in the Stevinlab by graduate students or PhD-students in combination with a technician and a scientific staff member. Research lines PhD candidates and Postdocs Laboratory

Half Height Horizontal

2nd round of the 2024 SPRINT funding call with São Paulo, Brazil

Cutting-Edge Multi-Camera System for Multi-Modal Data Capture

Researchers from the Intelligent Systems department have built an advanced multi-camera capture system designed for high-quality data collection from objects and people. By incorporating existing audio devices, this system enables the recording of multi-modal data from human social interactions including both video and audio, offering invaluable resources for understanding human behavior, modeling physical objects, and analyzing human-object interactions. 

Breaking the bubble in Buitenhof: TU studentexplores the connecting power of a mural

Op 16 september onthulde burgemeester Marja van Bijsterveld officieel de nieuwe muurschildering ‘Cherished’, op de flat aan de Haydnlaan in Buitenhof. TU Student Virginia Facciotto deed voor haar master ‘Design for Interaction’ een half jaar onderzoek naar de vraag wat de waarde van een mural is voor de wijk: “De muurschildering kan zeker bijdragen aan een positief beeld over de wijk, maar dan moet wel het hele verhaal verteld worden.”

Unconscious Bias session during Summer Introduction Programme (18-22 August 2024)

As part of Summer Introduction Programme , our colleagues Fatima Ait Bari and Stavroula Manolaki of the Diversity & Inclusion (D&I) Office gave an engaging presentation on Unconscious bias. The audience consisted of a group of master’s, exchange and bridging students, many from the international community, arriving at TU Delft this academic year 2024-2025. The session had as a goal to take the students along a “journey” of exploring how unconscious biases are formed, how they influence our interactions, and how we can become more aware of them. Participants were encouraged to reflect on their own biases and consider how they hinder their connection with people with diverse perspectives. Through interactive discussions and scientific insights, the presentation emphasized the importance of focusing on shared similarities to foster inclusive relationships. The D&I colleagues guided the students through a thought-provoking journey, helping them take steps toward greater inclusion and understanding in both academic and social environments. The presentation ended with encouraging the students to “Get together on the basis of our similarities; to grow on the basis of our differences”. We wish all new master, exchange and bridging students a successful and enjoyable academic year at TU Delft!

Fresh from Cafe X, not from the delivery service: food deliveries are not allowed

Geen eten bezorgen bij X
We hebben een nieuwe regel opgenomen in de huisregels van X met betrekking tot het nuttigen van afhaalmaaltijden bij X. Het is niet toegestaan om eten te laten bezorgen door een bezorgdienst en dit op het terrein te nuttigen. Je bent natuurlijk wel van harte welkom in Café X voor een hapje en een drankje.

Equity, Diversity & Inclusion Month 2024

In oktober 2024 viert de TU Delft de EDI-maand met als thema: United for Change, Together We Grow. Gedurende de hele maand biedt het Diversity & Inclusion office in samenwerking met de faculteiten, netwerken en serviceafdelingen workshops, trainingen en andere evenementen georganiseerd door en in de meeste gevallen aangeboden door collega's binnen de TU Delft.

NWO grants funding for innovative research on physical experimental environments

How to move from experiment to mainstream? A consortium led by professor Tamara Metze, has been awarded a prestigious grant from the Netherlands Organisation for Scientific Research (NWO). In search for pathways toward more sustainable futures, Metze and her team will explore how various innovations in field labs such as The Green Village, in urban living labs such as Engy Lab South-East in Amsterdam, and in all sorts of citizens’ initiatives, can be mainstreamed and make more impact on sustainability transitions. Pilot paradox The project ‘From EXperiment to sustainable change: TRAnsformative methodologies for Innovation and learning’ (EXTRA) seeks to overcome a persistent “pilot paradox”. In this paradox, much experimentation takes place but long-term systemic impact remains difficult. Researchers together with all sorts of change makers will synthesise existing knowledge on how to mainstream, upscale, spread, broaden and deepen developed innovations. Tamara Metze: ‘I am excited to unravel what are effective ways of cocreation that lead to mainstreaming the positive changes made in experimental environments. We will figure out how learning and innovation can lead to lasting changes in regulations, policies, and financial systems and the biophysical environment.’ Tamara Metze Read the NWO press release Actionable tools The project is crucial for accelerating sustainability transitions. By refining methodologies for mission-driven experimentation and develop hands on tools for all sorts of change-makers, it will be easier to mainstream the sustainable lessons and innovations. ‘These tools will not only aid grassroots innovators but also influence institutional and organisational structures, ensuring that lessons learned from experiments are better anchored in policies, regulations, and organisations’, explains Metze. The project will employ a transdisciplinary action research approach, bringing together knowledge from various disciplines and policy domains. By co-creating solutions with public and private partners, the research will have an immediate impact. In the long term, the project aims to build a more efficient innovation ecosystem, contributing to more impactful and sustainable outcomes for both society and the environment. Projectpartners TU Delft, VU Amsterdam, Wageningen University & Research, Hogeschool van Amsterdam, Erasmus Universiteit Rotterdam, Hogeschool Rotterdam, The Green Village, AMS Institute; PBL Planbureau voor de Leefomgeving, WoonFriesland, Dijkstra Draisma, Provincie Noord-Holland, Ministerie van Binnenlandse Zaken, PRICE / Almere, BouwLab, Alliantie Samen Nieuw-West, Innovation Quarter.

Interview Eveline Vreede with Digitaal Veilig Den Haag

Eveline Vreede is the Managing Director of our Safety & Security Institute. She is responsible for collaborations, programs and the development of the Safety & Security theme at TU Delft | Campus The Hague. Among other things, she connects the institute with government agencies. Digitaal Veilig Den Haag spoke with her about the institute's activities in the field of digital security.

TU Delft Excellence Fund supports Prof. Stefan de Witte and Dr Anasua Chatterjee

With support from the TU Delft Excellence Fund, two prominent scientists recently started their work at TU Delft. As per September, Prof. Stefan de Witte has been appointed Professor of Optics for Nanoscale Metrology at the department of Imaging Physics.

Researchers Use an EUV Flashlight and Free-Air Diffusion, Plus a Clever Algorithm, to Inspect Chips

Postdoc Yifeng Shao (ImPhys) and his colleagues from Delft and Utrecht University have investigated a new imaging technique using extreme ultraviolet light to study the three-dimensional structures of chips at the nanoscale.

Outstanding Paper Award on Scientific Understanding in RL

M. Suau, M. T. J. Spaan, and F. A. Oliehoek were awarded the Outstanding Paper Award on Scientific Understanding in RL at the first Reinforcement Learning Conference (RLC).

Professor of Urban Design Tess Broekmans: ‘Designers must learn to listen better’

Tess Broekmans, co-owner of Urhahn urban design and strategy, delivered her inaugural address as professor of Urban Design on Friday 6 September. In it, she addressed ‘the complexity of the everyday’. After all, how can designers' visions connect better with people's everyday lives? In doing so, she also sees a new role for area developers.

2 Veni Grants for the ELLIS Unit

In an effort to make new and strengthen existing connections, a delegation of ELLIS Unit Delft visited ELLIS Munich and some of the most renowned AI research institutions in Munich area on the 25th and 26th of January 2024. The delegation consisted of people from TU Delft’s cognitive robotics department, and the sequential decision making group.

Unusual waves grow way beyond known limits

Waves that come from multiple directions are extremer than extreme. These remarkable deep-sea waves can be four times steeper than what was previously imagined, as is shown in research by TU Delft and other universities that was published in Nature today. A long time ago, stories were told of mysterious rogue waves that materialised out of nowhere and could topple even the largest ships. These waves lost their mythical character when the first rogue wave was recorded at the Draupner platform in the North Sea. In 2018, Ton van den Bremer and his colleagues at the Universities of Edinburgh and Oxford managed to recreate the Draupner wave in the lab for the first time ever, and this opportunity to study freak waves closely produced unexpected insights. Multiple waves push up water New research by the research consortium now shows that these remarkable waves do not break when traditional theories hold they should, the secret behind which lies in how they arise. Ton van den Bremer, expert on fluid mechanics at TU Delft and led the study, explains: “When most people think of waves, they think of the rolling waves you’d find on a beach. The type of wave we studied occurs in open water and arises when waves coming from multiple directions come together. When these waves with a high directional spread converge, the water is pushed upwards, forming a partially standing wave. An example of this is known as a crossing wave. How crossing waves arise Under certain conditions at sea, waves from multiple directions occur. This can happen in a place where two seas meet, or where winds suddenly change direction, as in a hurricane. When waves from two directions meet, a cross wave occurs, provided their directions are far enough apart. The study also shows that the further apart the directions are, the higher the resulting cross-wave. Travelling waves break when they reach a certain limit, this is when they reach their maximum steepness. The study shows that waves with a multidirectional spreading can get as much as 80% steeper than this limit before they start breaking, which means they can get almost twice as high as ‘normal waves’ before they start to break. Travelling wave (l) and a wave with high directional spreading (r) Breaking waves that grow Next, the researchers found another highly unusual phenomenon that defies existing theories, a phenomenon that is unprecedented according to Van den Bremer: “Once a conventional wave breaks, it forms a white cap, and there is no way back. But when a wave with a high directional spreading breaks, it can keep growing.” The study shows that these enormous waves can grow to twice their original steepness while breaking, which is already twice bigger than the conventional limit. Together, the waves can grow four times steeper than previously thought possible. Damage to offshore structures The knowledge that multidirectional waves can become as much as four times larger than was deemed possible can help design safer marine structures. "The three-dimensionality of waves is often overlooked in the design of offshore wind turbines and other structures in general; our findings suggest this leads to designs that are less reliable", says Mark McAllister of the University of Oxford, who led the experiments and is now a senior scientist at Wood Thilsted. Innovative vertical sensors made it possible to take accurate 3D measurements of waves. Innovative 3D measurement method A 3D measurement method developed in the FloWave lab paved the way for these new insights. “Conventional 2D wave measurement methods weren’t up to the task”, Van den Bremer explains, which is why the research group designed a new way to create 3D wave measurements. Ross Calvert of the University of Edinburgh: “This is the first time we've been able to measure wave heights at such high spatial resolution over such a big area, giving us a much more detailed understanding of complex wave breaking behaviour." FloWave Ocean Energy Research Facility in Edinburgh. The circular basin has a diameter of 25 metres and can be used to generate waves from multiple directions. Header image by: Fabien Duboc

A strong nose for trouble

Ammonia gas – it is both highly poisonous and corrosive, yet also often used in industry, or present around intensive livestock farming. To better mitigate the risks of ammonia leaks, Paddy French together with Lina Sarro and Sten Vollebregt will make smaller, more robust and eventually more affordable ammonia sensors. The NWO OTP programme has funded their research into a fully-integrated, single-chip sensor that incorporates two porous silicon carbide structures for detecting ammonia and humidity, all in a single layer and one etch step.

Stefan Witte joins Imaging Physics as professor

Delft University of Technology's Executive Board has appointed Stefan Witte as professor of Optics for nanoscale metrology in the Imaging Physics department of the Faculty of Applied Sciences as of September 2024. Thanks to the TU Delft Excellence Fund, Stefan Witte can get off to a flying start with his new research group and expects to obtain the first research results soon.

October 2024: Equity, Diversity & Inclusion Month

Theme: “United for Change, Together we Grow” The EDI Month programme focuses on bringing together people from different backgrounds, perspectives, and opinions to foster an environment where meaningful conversations and innovative ideas can flourish. From interactive workshops on diversity and teamwork to inclusive leadership training, all with an emphasis on “practical learning” and free of charge . Curious about what's on offer? Visit the online calendar for a full overview of the events and sign up.

SPTL contributions at European Transport Conference (ETC) (18-20 September, Antwerp)

The 2024 edition of the European Transport Conference (ETC) will be held from 18-20 September at University of Antwerp, Belgium. The following Smart PT Lab contributions will be presented: Passenger preferences for (emerging) public transport access/egress modes Nejc Gerzinic, Mark van Hagen, Hussein Al Tamimi, Dorine Duives, Niels van Oort [Abstract] - [Slides] An application of Latent Class Analysis to explore the impact of sociodemographics on travel behaviour profiles – the importance of sociodemographics in transport policy Nadine Martje Eichenauer, Niels van Oort, Bert van Wee, Jan Anne Annema [Abstract] - [Slides] Towards a Global Standard for Estimating Life Cycle Greenhouse Gas Emissions from Public Transport Services Marko Kapetanovic, Niels van Oort [Abstract] - [Slides] The full program can be found here: https://aetransport.org/etc/programme

Towards more efficient solar cells, with a lower carbon footprint

The SiLEAN project, funded by the European Union, has officially launched with the goal of setting a new benchmark in solar cell efficiency by over 25.5%, while reducing costs and lowering the carbon footprint by up to 75%.

Ensuring safe and quick food bike-delivery services in cities

Cyclists rushing to deliver food on time to hungry customers is an increasing problem in cities with the growing demand for fast home delivery services. This surge has resulted in more traffic in bike lanes, contributing to accidents and raising concerns about rider safety. In the framework of SINERGI project researchers like Yousef Maknoon are looking into solutions that will help boost logistic operations that are both efficient and safe. ‘Bike delivery services are a great way to transport goods from A to B in a green manner, replacing delivery cars or vans. These micro-delivery systems can really help make city logistics more sustainable,’ says logistics and modelling expert Yousef Maknoon, who is involved in developing real-time decision models for the SINERGI project. Besides looking into sustainability, the project aims to empower delivery providers to offer cost-efficient services, while prioritising the well-being of riders. ‘We want to help bring down the number of accidents,’ stresses Maknoon. Private and public stakeholders have joined to conduct pilot initiatives in the cities of Amsterdam, Shanghai, Singapore and Copenhagen. Yousef Maknoon Safe routing ‘To improve rider safety, the app automatically provides recommended routes when food delivery riders receive their task, helping them avoid dangerous intersections and busy streets. In Amsterdam, the goal is to develop a map that will show the safety of cycling infrastructure. The map is planned to be enriched with real-time information such weather conditions, busy days, and times of the week to not only improve delivery time prediction but also assess route safety in real time,’ explains Maknoon. ‘But everything depends on the cyclist's willingness to take the safer route. When a rider is paid per delivery, they might not be willing to take a longer route. Especially when they know the area well, it can be tempting to take a shorter route that may not always be the safest option.’ A survey held amongst 300 delivery riders found that they can be nudged to follow the recommended routes when they are compensated financially. Profit versus safety Maknoon: ‘What we also found is that paying an hourly fee, instead of paying per on-time delivery, promotes safer riding behaviour, like Just Eat Take​away does in the pilot in Amsterdam. However, other issues arise because riders might not be so motivated to deliver on time. Then you need other incentives to motivate efficiency among riders. Otherwise, customers might switch to other service providers. This puts the business model which is based on hourly payment and prioritize safety, under pressure.’ It seems now that delivery services need to make a trade-off between profit and safety. ‘With the SINERGI project we want to empower delivery providers to offer cost-efficient services while prioritising the well-being of riders.’ Government regulation Public authorities can, for example, improve road safety by enhancing infrastructure, such as building separate bicycle lanes.’ Delivery providers can introduce incentives for riders that compensate the extra time a longer, but safer, route might take. But this puts pressure on profitability and could cause micro-delivery logistics platforms to struggle or even go out of business. It is likely that government regulation will need to play a role in profit-driven societies, as safety risks being overshadowed. Fair delivery system This is also the case for ensuring a fair delivery system where all customers are serviced, whether they live in the busy city centre or the less densely populated countryside. Otherwise, customers who live farther away may not be serviced, as longer cycling times mean less money for riders when they are paid by commission. Riders can simply reject these orders and choose to deliver only to places where they think they can earn more. Human factor In this digitally connected world, we need to consider the human factor in real-time decision-making. The wellbeing of riders, their working conditions, and their quality of life are also important when we talk about making operational decisions. It is interesting to compare the four cities and how they manage the tension between cost-efficiency and safety, and to see what they can learn from each other. About SINERGI SINERGI is a three-year project that runs from 2023 to 2026. It involves thirteen research institutions, governmental authorities, and companies collaborating on a comprehensive framework to improve sustainable city logistics. The project aims to enable real-time management and strategic planning of efficient, safe and user-friendly delivery services. Besides Yousef Maknoon other researchers from TU-Delft involved are: Shadi Sharif Azadeh (project leader), Gonçalo Correia, Kuldeep Kavta and Dongyang Xia. Read more about the SINERGI project.