/Local Digital Twins in daily urban policy support

Local Digital Twins in daily urban policy support

PhD - Antwerpen | More than two weeks ago

Expanding the possibilities of cross-domain data-driven decision making in cities.

Local Digital Twins in daily urban policy support

Expanding the possibilities of cross-domain data-driven decision making in cities

Digital Twins

Since its inception around the turn of the century, the definition of a Digital Twin has been debated and challenged by various authors (e.g. Grieves and Vickers 2017, Alam and El Saddik 2017, Zheng et al. 2019). One of these definitions, which we will adopt in this paper, describes a Digital Twin as a “virtual representation of a physical entity with a bi-directional communication link” (adapted from Tao et al 2018). One of the key terms in this definition is the bi-directional communication link, which can be split-up into a communication link from the physical entity to the Digital Twin, and a communication link from the Digital Twin back to the physical entity.

Communication link from the physical to the Digital Twin

The communication link from the physical to the Digital Twin refers to the constant use of several data collection techniques and devices (like sensors connected through Internet of Things - IoT technology) to convene and combine data about the physical entity. This data allows the virtual Digital Twin to constantly learn from and evolve along with its physical counterpart, mirroring its lifecycle. As a result, a Digital Twin can be used to get insights into the current state of the physical entity as well as predict its future states through causal data models and simulation algorithms.

Communication link from the Digital to the physical twin

The Digital Twin may transform the physical entity by either automatically driving actuators connected through IoT technology or by informing human decision-makers who can make adaptations to the physical entity. The latter can be made through daily operational management or through long-term policy planning. In either case, the Digital Twin should be able to capture the changes made to the physical entity so that it can use those changes and their impact to update its data models. Thus a Digital Twin is not merely a detached virtual representation of a physical twin, but rather an evolving representation that constantly interacts with - and tests the impact of - novel configurations in the physical entity that it represents.

Local Digital Twins

With our improved understanding of Digital Twins in general, we can now better understand what is intended by a "Local" Digital Twin: it is a version where the physical entity is a city, a rural or a local entity and which is used to support decisions that pertain to this location. We will define them as follows: Local Digital twins are dynamic, data-driven, multi-dimensional digital replicas of a physical rural, local or urban entity. They encompass potential or actual physical assets, processes, people, places, systems, devices, and the natural environment. Their aim is to capturing reality for supporting reactive and predictive policies at relevant decision-making levels in order to ensure minimum environmental impact, improved quality of life, and enhanced performance of the local entity. Complexity in Local Digital Twins

The dynamics of cities have been widely studied as a hierarchical and rigid system since the formulation of the general systems theory (GST) in the 70’s. Most of the urban planning of the last

few decades was based on the perspective that cities have stable structures and are limited to their geographic position. Consequently, urban planning was based on a model where the elements interact with order, regularity, predictability, and control. The growth of the physical space of cities and migratory movements from rural areas to urban centers has added a level of difficulty in urban planning to encompass efficient distribution of e.g. food, energy, and water. Policymakers have needed new approaches to better understand the evolution of common urban problems of modern cities like traffic congestion, inefficient public transports, social segregation, and poor access to leisure spaces. Analyzed separately, these problems cannot be fully understood, since they are the result of the interaction of different systems within the cities. A better approach to have a holistic view of such challenges is the modeling of cities as complex systems.

In this text, we understand complexity as the collective behaviors that emerge from individual interactions. In complex systems, each part of the system is governed by a given number of local rules. Every subsystem is aware of its local environment and its resources but is it not aware of the rules governing other subsystems. These often-nonlinear interactions usually present spontaneous and efficient self-organization. They are adaptive and robust to changes. As a result, they present different emerging behaviors.

In the last decades, particularly the field of complex networks stimulated the creation of models for e.g. epidemics, transport, energy, communications, and social interactions. These models have been used to study subsystems within cities. However, cities are complex systems themselves. They are composed of an extensive number of agents, they evolve over time, they present collective intelligence, and are in constant adaptation. The complexity of systems within cities results in a constant process of spatial reorganization. The city changes with the incorporation of new areas, the need for more resources, the renewal of physical spaces, and other changes to the social and economic context. The intricate relation of these subsystems in the physical space results in urban complexity that should be part of LDT implementations.

LDTs are presented as a framework capable of modeling cities and their complexity. Given the LDT’s layered architecture, they can embed different networks in the simulation of physical spaces. This has been possible in the last years due to the advances in big data analysis, new specialized machine learning algorithms, and the possibility of high processing speed and memory-consuming computations available in cloud-based environments. LDTs have the potential to give a broad vision of the cities' activities. They can be applied to understand underlying causes of urban wicked problems. Another advantage of modeling cities with LDT is the large number of scenarios that can be simulated by changing parameters in different systems. For instance, LDT can be applied to evaluate the impact of increasing human activity in post-pandemic cities. Another application in the same context is the analyses of different post-COVID-19 socio-economic strategies that will take place in the upcoming years.

It is important to emphasize that systems related to the social domain should be integrated in LDTs. Cities are made of multiple qualitative relations (economic, political, social, and cultural). This aspect should not be undervalued in computational modeling. Simulations that are only considering the physical aspect of the cities could give naive interpretations of urban system, consequently giving insights into the wrong problems. A digital version of a city should be a tool where the citizens are not only agents but also part of public policy changes. The implementation of LDT might assist cities to evolve into a more participative and inclusive technological society.

Challenges of Local Digital Twins in daily local decision support

We propose that the PhD candidate selects at least three from this list, details these with subquestions and if necessary, expands the list of RQ’s with additional ones of interest in mutual agreement with the PhD promotor(s). We have categorised the RQ’s according to three themes.

Theme 1: complexity

Local Digital Twins can be understood as the interplay between GIS, IoT and Machine learning (or other types of) algorithms. Arcaute et al (2021) explain how a purely data-driven, ML based approach will not cover all aspects of the complex nature of society. They call for the addition of approaches that allow the emergent and self-organising nature of societal phenomena to be included in the way LDT’s operate. One way to do this is to incorporate concepts from complexity theory in LDT systems. This brings to a first research question.

RQ1: how to address the complex nature of local decisions using LDTs?

Currently, LDTs are very much seen as a tool for elected policy makers or public servants. However, there is also a call to make local decision making more inclusive, as e.g. evidenced by the New European Bauhaus movement (European Commission 2021). Still, there is a fundamental tension between the decisions that are taken by the politically or administratively elected bodies in a top- down way and the decision that are desired by the citizens, bottom-up. Democratic and administrative processes are imperfect and at risk of resulting in changes to local living conditions that are opposed by the people living in them. Therefore, we see the following question to be relevant:

RQ2: How to support citizen-based, bottom-up and inclusive decision making in LDTs ?

Theme 2: modelling

To be able to grow with the decisions that a local government needs to take, an LDT needs to be open and expansible. This means it needs to be able to incorporate new datasets and new models (i.e. algorithms) from a variety of data owners and model providers. Such models and data sources need to be made available in a way that they can interact with the LDT architectures of a variety of local governments. In addition, in order to support the cross-domain decision support features of LDTs, models need to be able to interact, making sure that the outputs of one model (e.g. traffic flow prediction) can become the input of another model (e.g. air quality prediction). Data gathering for this RQ could start from the most impactful and sought-after use cases for LDTs and work from the functional requirements of these use cases to establish efficient ways for models to interact. Such model interactions can be formalised by designing model interaction graphs that together with model signatures (describing model meta-data) can implement the LDT use cases. These interaction graphs and model signatures can be leveraged to offer information to prospective LDT developers through a model marketplace. Thus, we ask the following question:

RQ4: What requirements should be asked of models in order to allow them to interact, how can model interaction graphs and model signatures support this, how can this be the input for an LDT marketplace and how can models be made pluggable in an LDT framework?

The simulations that are produced by LDTs very often are based on historical data. This approach ensures that the validity dimension of the simulation is assured, as the data was at some point generated in the same geographical area as where the simulation is applied. However, this also holds some challenges. For one, it is very hard to simulate rare situations based on historical data, as these may not have occurred yet and are therefore not reflected in historical data. An example would be the identification of the impact of a pandemic on the critical infrastructure of a city. The most recent large-scale pandemic to have occurred in a region like Flanders was the Spanish Flue in 1918, of which the evolution was not captured in the same detailed dataset as would be necessary for modern pandemic modelling. Another problem of historical data is data some of it is prone to privacy concerns and is therefore hard to use in decision support tools like LDTs. Finally, some domains of interest to simulation simply are not measured, making the dataset locally inexistent. To address these issues,

Papyshev & Yarime (2021) call for the inclusion of synthetic datasets in LDTs. Such datasets are modelled on top of existing datasets that have traits which are perceived to be relevant to the simulation topic at hand: “Fully synthetic data can be understood as artificial data that are statistically similar to the original, but the new data have no identifiable information about their origin(Papyshev & Yarime 2021, p8). This brings us to the following RQ:

RQ5: How can synthetic data be used in LDTs?

Theme 3: impact

There are currently several investment programs being drafted to deploy and scale LDTs in the EU. The Netherlands, Luxemburg, Denmark, Flanders and the European Commission (through the Digital Europe Program) are getting ready to invest tens of millions of Euros to this end. Although few doubt the validity of the concept, the impact of LDTs is unclear, as their application domain is quite broad (e.g. air quality, traffic, pandemics, logistics, energy, emergency response,...). There is a need to describe and if possible, quantify the impact of LDTs, so investments can be weighed against their likely returns:

RQ3: How can the impact of LDTs be described, either qualitatively or quantitatively?

Digital Twins at imec

Digital Twins have gained a lot of attention at imec in the past three years. First there is the focus on Urban Digital Twins, which has become one of the main focal points in the City of Things smart city program. There, Digital Twins are seen as a main platform for supporting decisions that pertain to the Urban context. Different cities (e.g. Bruges, Antwerp, Ghent) in Flanders are working with imec on Digital Twin projects. In addition, imec is involved in initiatives at the EU level to deploy Urban Digital Twins in various cities, like Pilsen, Athens, Helsinki and Santander.

Imec’s interest in the Digital Twin concept stems from its ability to integrate various sensors in a way that allows end-users to make better decisions in diverse domains. It can integrate competences from many different imec teams, like sensor development, IoT communications, algorithm development, semantic modelling, GUI design, business modelling and governance support.

As it is a natural fit with its DNA, imec has decided to become member of the Digital Twin Consortium, which coordinates standardisation and valorisation of Digital Twin activities across domains (infrastructure, aerospace, health, ...). Also, imec is driving the effort in various EU projects and initiatives that are related to LDTs, like Precinct, Pioneers, Urbanage, Duet, the Digital Europe Program and Living-in.eu.

References

Arcaute, E., Barthelemy, M., Batty, M., & Caldarelli, G., Gershenson, C., Helbing, D., Moreno, Y., Ramasco, J. J., Rozenblat, C., Sánchez, A. (2021). Future Cities: Why Digital Twins Need to Take Complexity Science on Board., https://www.researchgate.net/publication/354446988_Future_Cities_Why_Digital_Twins_N eed_to_Take_Complexity_Science_on_Board

  • Alam, K. M., & El Saddik, A. (2017). C2PS: A digital twin architecture reference model for the cloud-based cyber-physical systems. IEEE access 5, 20502062

  • European Commission, (2021). New European Bauhaus, https://europa.eu/new-european- bauhaus/index_en , Retrieved 28/09/2021

  • Grieves, M., & Vickers, J. (2017). Digital twin: Mitigating unpredictable, undesirable emergent behavior in complex systems. Transdisciplinary perspectives on complex systems, 85113. Springer.

  • Ortman SG, Lobo J, Smith ME (2020) Cities: Complexity, theory and history. PLoS ONE 15(12): e0243621. https://doi.org/10.1371/journal.pone.0243621

  • Papyshev, G., & Yarime, M. (2021). Exploring city digital twins as policy tools: A task-based approach to generating synthetic data on urban mobility. Data & Policy, 3.

  • Santos, Samuel Steiner dos. (2013). Complex Cities: What urban planning in the perspective of complexity theory?

  • Zheng, Y., S. Yang, and H. Cheng. 2019. An application framework of digital twin and its case study. Journal of Ambient Intelligence and Humanized Computing, 10(3), 1141-1153.

 

Required background: computer science, software engineering, business studies

 

Type of work: 70% modeling/simulation, 20% experimental, 10% literature

Supervisor: Pieter Ballon

Co-supervisor: Tanguy Coenen

Daily advisor: Stefan Lefever

The reference code for this position is 2022-110. Mention this reference code on your application form.