21.12.2025

How Synthetic Data Is Changing Urban Planning

Building design

A busy city street next to imposing skyscrapers, photographed by Bin White.

Urban planning with a touch of science fiction? Synthetic data is propelling urban planning out of the analog age—and turning assumptions into reliable forecasts. What’s really behind this digital gold? Who is already using synthetic data strategically? And why could it fundamentally change how urban planners see their role? A look behind the scenes of the new reality of planning.

Definition and Development of Synthetic Data for Urban Planning
How synthetic data is surpassing traditional urban models and enabling simulations
Practical examples from Germany, Austria, and international cities
Relevance for climate resilience, mobility management, and social urban development
Technical and legal challenges in the use of synthetic data
Risks: algorithmic bias, data protection, technocratic effects
Potential: Faster scenario development, resource-efficient planning, and improved citizen participation
Governance issues and the role of open-source ecosystems
How synthetic data is redefining the professional role of planners

Synthetic Data: What’s Behind the Hype?

Anyone who asks around in city administrations, planning firms, or development agencies today will find that hardly any other topic excites urban professionals as much as synthetic data. But what exactly does this term—which sounds like a mix of science fiction and mathematics—actually mean? Put simply, synthetic data consists of artificially generated datasets that model real-world phenomena without being derived directly from observations of reality. It is generated through computer-aided simulations, statistical models, or machine learning and enables a new level of data utilization that goes far beyond traditional measurements.

In the context of urban planning, synthetic data represents a quantum leap. It allows us to simulate scenarios that do not yet exist in reality or whose collection would be practically impossible, too expensive, or legally prohibited. For example: How would traffic in a city center change if an entire neighborhood were made car-free? Or: What microclimatic effects would 10,000 square meters of green roofs have in a densely built-up neighborhood? Synthetic data makes it possible to answer such questions without invasive measurement campaigns or years-long field studies.

However, generating synthetic data is anything but trivial. It requires a deep understanding of the systems being modeled, access to high-quality baseline data, powerful computing power, and sophisticated algorithms. This is where methods such as agent-based modeling, generative models (such as GANs: Generative Adversarial Networks), or rule-based simulations. These methods make it possible to create artificial populations, mobility patterns, or even social interactions—always with the goal of generating plausible and robust datasets.

In practice, synthetic data is frequently used to fill gaps in existing datasets or to meet data protection requirements. For example, if personal mobility data cannot be collected, synthetic movement profiles can be created that are realistic enough for simulations but completely anonymous. This opens up new avenues for evidence-based planning without legal pitfalls or ethical concerns.

But the hype surrounding synthetic data also carries risks. After all, when reality is recreated, there is a strong temptation to confuse the digital world with the real one. Planners must learn to deal with uncertainties, model assumptions, and algorithmic biases. The trick is to view synthetic data as a supplement—not a substitute—for sound expertise and local experience. Only then will it become a true asset for urban planning.

From Theory to Practice: How Synthetic Data Is Transforming Urban Planning

The use of synthetic data in urban planning is no longer limited to pilot projects or research experiments. Rather, it has made the leap into operational practice, particularly in major cities that rely on data-driven decision-making. But what does this transformation look like in concrete terms? And which disciplines benefit the most?

A key area of application is mobility planning. Cities such as Zurich and Copenhagen use synthetic data to simulate, in real time, the impact of new bike lanes, altered traffic routing, or urban logistics concepts. Here, millions of movement profiles are generated based on statistical models that replicate the behavior of different user groups. The result: reliable forecasts of how traffic flows, emissions, or local mobility will change—even before the first shovel hits the ground.

Synthetic data generation also plays a key role in the field of climate resilience. In Vienna, for example, a synthetic urban climate model was developed that simulates microclimatic effects based on geodata, building typologies, and weather data. This allows for the virtual testing of heat islands, cold air currents, and the impact of greening measures. For planners, this means they can develop targeted measures for heat protection or stormwater management without having to wait for lengthy measurement series.

Another area is social urban development. Here, synthetic population data makes it possible to simulate the composition of neighborhoods under various assumptions. How does a changing age structure affect the demand for daycare centers or local amenities? What happens when certain migrant groups move in or new forms of housing are promoted? Synthetic data provides scenarios that help plan resources efficiently and in line with actual needs.

But it’s not just large cities that benefit. Smaller municipalities are also increasingly relying on synthetic data, for example, when developing residential areas or planning bus routes. Here, open-source tools and freely available data sources are particularly in demand to create robust models on limited budgets. The motto is: simulation instead of speculation—and with manageable effort.

The greatest strength of synthetic data lies in its flexibility. It allows for systematically answering “what-if” questions, comparing alternatives, and identifying risks early on. This makes it an indispensable tool for anyone who wants to maintain an overview in an increasingly complex urban environment. But to fully harness this potential, more than just technology is needed: a new mindset in planning culture is required.

Technical Foundations and Challenges: From Data Quality to Data Protection

Before synthetic data can work its magic, technological and methodological hurdles must be overcome. The quality of synthetic datasets stands or falls with the quality of the underlying models. These models must be based on reliable, up-to-date, and as comprehensive as possible baseline data—whether it be geodata, traffic counts, climate data, or sociodemographic information. Without a solid foundation, simulation results risk being implausible or even misleading.

Another key issue is the modeling itself. It is not enough to simply “generate” data. Rather, the models must be designed to realistically capture the complexity of urban systems. This is where methods such as stochastic simulation, agent-based modeling, and machine learning come into play. Each method has its strengths and weaknesses—and its own specific requirements for data, computing power, and expertise.

Data protection is another minefield. Even though synthetic data, by definition, does not represent real individuals, it can still allow inferences to be drawn about real-world behavioral patterns or groups. Caution is particularly warranted in small cities or when dealing with high-resolution mobility data. It is essential to maintain data minimization and prevent misuse. Legal expertise, transparency, and close collaboration with data protection officers are indispensable here.

From a technical standpoint, integrating synthetic data into existing urban models also poses a challenge. Many municipalities work with fragmented GIS systems, proprietary software solutions, or inadequately documented datasets. To make effective use of synthetic data, interfaces must be created, standards established, and interoperability ensured. Open urban platforms and open data ecosystems are key components here.

Finally, the issue of governance must not be underestimated. Who controls the models and decides on assumptions and scenarios? If synthetic data is delivered as a “black box” by service providers, there is a risk of losing transparency and planning autonomy. The public sector must build expertise in this area, set standards, and ensure traceability. Only in this way can we prevent the urban future from being driven by algorithms rather than by societal goals.

Opportunities and Risks: How Synthetic Data Challenges Our Understanding of Planning

The potential of synthetic data for urban planning is enormous—but it comes with fundamental changes to the profession’s self-image. On the one hand, it opens up the possibility of shaping urban development in an experimental, evidence-based, and resource-efficient manner. New developments, infrastructure projects, or climate adaptation measures can be simulated, optimized, and tailored to needs in advance. Citizen participation gains in quality when complex interrelationships are clearly visualized and various scenarios are made understandable.

At the same time, however, new risks are emerging. A central problem is algorithmic bias: models are only as good as their assumptions—and these often reflect dominant perspectives or political goals. Without critical reflection, there is a risk of technocratic urban development in which social, cultural, or ecological aspects are overlooked. Planners must therefore learn to deal with uncertainties and not be blinded by the apparent precision of synthetic data.

Another risk is the commercialization of urban models. When synthetic data is generated by external providers, dependencies can arise that jeopardize the sovereignty of the public sector. This requires clear contractual provisions, open standards, and a conscious commitment to open source wherever possible. Only in this way can control over urban data remain in public hands—and planning remain democratically legitimate.

Ethical questions are also coming to the forefront. Who decides which scenarios are simulated? Are minorities or specific groups given sufficient consideration? How can transparency be ensured when models are becoming increasingly complex and opaque? Synthetic data requires new forms of communication, participation, and accountability. Here, public administrations, policymakers, and civil society share equal responsibility.

Finally, it must be emphasized that synthetic data is changing the role of planners. Traditional designers and administrators are increasingly becoming facilitators, data managers, and scenario architects. The ability to deal with uncertainties, critically examine results, and integrate various sources of knowledge is becoming a key competency. Those who embrace this challenge can actively shape the city of tomorrow—rather than being shaped by data algorithms.

Perspectives: Synthetic Data as the Driving Force Behind a New Urban Planning Culture

Where do we stand today—and where are we headed? Synthetic data is more than just a passing trend. It is the driving force behind a new, process-oriented approach to urban planning, in which experimentation, reflection, and adaptation are central principles. Cities like Vienna, Zurich, and Helsinki demonstrate how synthetic data can be used to tackle complex challenges such as climate change, the mobility transition, and demographic shifts. Here, urban decision-making processes are not replaced but rather enriched by data-driven simulations.

In Germany, Austria, and Switzerland, progress in this area remains uneven. While some municipalities are boldly forging ahead, others are hesitant in the face of technical, legal, or cultural hurdles. Yet the trend is unmistakable: Those who take the leap into synthetic data today gain an innovative edge—and can make urban spaces more resilient, livable, and sustainable.

To realize its full potential, however, more than just technology is needed. It requires education, dialogue, and cooperation. Universities, government agencies, the private sector, and civil society must work together to build expertise, set standards, and establish an open, learning-oriented planning culture. Only then will synthetic data become a catalyst for democratic, sustainable urban development.

Citizen participation is also undergoing a transformation. Synthetic data can help make complex issues understandable and bring different perspectives to light. It makes participation more inclusive, transparent, and comprehensible—provided that models and assumptions are communicated openly. The city of the future is not created in a data center, but through the dialogue between data, knowledge, and lived experience.

The question remains: What does this mean in practice? It is time to view planning processes as open, iterative systems in which experimentation is permitted and mistakes are opportunities to learn. Synthetic data is not a panacea—but it is the tool that propels planning into the digital age. Those who use it wisely shape change—and stay in tune with a city that is changing at an ever-faster pace.

In summary, synthetic data is far more than a technical gimmick. It is a powerful tool for making urban planning more flexible, forward-looking, and transparent. Its use requires technical expertise, critical reflection, and an open approach to uncertainty. When applied with sound judgment, professional competence, and democratic oversight, it opens new horizons for sustainable, resilient, and livable cities. The future of planning is synthetic—and it begins now.

15.12.2024

Fairs

Full house at Strassacker

Building design

The industry meets at the Info Days 2019 Advertorial Article Parallax Article

The industry meets at the 2019 Info Days

(more…)

15.01.2025

Profession

Really wrong!

Building design

About legal canvas copies and the limits of copyright A conversation with Dr. Anke Schierholz, legal advisor to VG Bild-Kunst, about copyright and the difference between a forgery and a legal copy of a work of art. VG Bild-Kunst represents the copyrights of over 54,000 artists who create works of art. In addition to concluding license agreements, VG […]

About legal canvas copies and the limits of copyright law

A conversation with Dr. Anke Schierholz, legal advisor to VG Bild-Kunst, about copyright and the difference between a forgery and a legal copy of a work of art. VG Bild-Kunst represents the copyrights of over 54,000 artists who create works of art. In addition to concluding license agreements, VG Bild-Kunst monitors the receipt and distribution of the agreed remuneration and is also committed to strengthening copyrights at national and international level.

The Beltracchi case caused quite a stir a few years ago. To what extent were copyright issues involved here?
Dr. Schierholz: Unfortunately, the Beltracchi case has little to do with copyright law. Copyright law is only affected in the area of forgeries that appear in the art trade if it involves the reproduction of an existing and protected work. From a copyright point of view, action can only be taken against forgeries if there is a reference to an existing work and it is more than a free adaptation. A forgery goes beyond a mere suggestion if, for example, details are taken from the original.

The forgeries brought onto the market by Beltracchi are based on historical black and white images of lost paintings. So no protected work is involved here?
Dr. Schierholz: Basically, yes, but VG Bild-Kunst was not involved in the Beltracchi case. As a rule, we are involved by auction houses when there are doubts about a picture up for auction. Normally, the auctioneers know their field of art and the range of works by the artists they represent very well. If, for example, a “lost Beckmann” is to be foisted, they usually recognize this. In cases of doubt, the police are then called in to confiscate the paintings in question. We are contacted by the criminal investigation department and file a criminal complaint not only for fraud, but also for copyright infringement.

So VG Bild-Kunst is responsible for prosecuting these offenses, but not for investigating them on a restoration or scientific level?
Dr. Schierholz: Exactly. We can sometimes recognize that a detail on a supposed Beckmann painting is by another artist, but that is the exception. We look after the rights, the material investigations are carried out by the galleries and auction houses.

Copyright must be distinguished from the possibility of creating legal copies of a work. For example, when can I legally “copy” and exhibit a painting?
Dr. Schierholz: This is possible if the author, i.e. the artist, has been dead for more than 70 years. “Simply dead” is not enough, at least 70 years must have passed since the death. This circumstance often leads to misunderstandings in the sense of “I’m allowed to paint Picasso, he’s dead”. Although this is true, it is wrong in the context of copyright law. On the other hand, anyone may copy works of art within the scope of private copying, whether by hand, photographically or otherwise. This is possible at any time, but these copies may not be published. Neither in a doctor’s surgery nor in the stairwell of an apartment building. There is also the right to adapt, which allows existing and protected works to be adapted. This applies primarily to translations, which require the consent of the original author before publication. However, the right of adaptation also applies to works of art.

There are painting studios that specialize in copying pictures. Are paintings by artists who have been dead for over 70 years allowed to be copied 1:1 or is there a need for a different format, for example?
Dr. Schierholz: These works are free, i.e. no longer protected by copyright. They may be reproduced in any format of your choice. A problem arises when such a legal counterfeit is later placed on the market with the claim that it is the original. This constitutes fraud. This claim can be remedied by choosing a different format, affixing a signature on the reverse or making other obvious changes. Forgery trials such as the Beltracchi case are primarily fraud trials if someone claims to be in possession of the original – which is not true. Fraud is committed when a picture is given an aura that does not exist in the public eye, as it is simply a painted picture.

Inother words, applying a signature to a legal copy is not a copyright problem?
Dr. Schierholz: Exactly, because as long as I don’t claim it’s the original, it’s not a problem. The doubts about putting a signature on a copy stem from the old Kujau trials. Konrad Kujau had claimed to own paintings by Kandinsky and Picasso. In the fraud trial, he was convicted of forgery, as the artist’s signature authenticates the work, so to speak, and supposedly makes it an original, which is not the case. This reasoning sounds somewhat absurd by today’s standards, but it was the only point of attack.

Apparently, restorers have often become art forgers in the past. Sometimes there was a trial over forged Nussbaum works, sometimes over Cranach paintings. Does our profession pose an above-average danger?
Dr. Schierholz: Restorers are naturally very good at the painter’s technique. If someone knows exactly how to do it, they are of course a good restorer. And there are certainly the odd black sheep, just as there is the occasional policeman who deals drugs. But the job description does not indicate any particular tendency towards danger.

Where do you think the desire to decorate oneself or one’s home with a copy as a substitute for the original comes from?
Dr. Schierholz: The motivation is not clear to me either. I have to assume that the bourgeois impetus is a major motivation for the purchase of canvas copies. A canvas copy is seen as an attribute for a sense of art – just like the Brockhaus encyclopaedia with gold edging – as a key signal for the educated middle classes and an expression of solidity. It seems to me a rather petty-bourgeois idea that a painting on canvas is valuable per se. Yet a copy has no more value than a poster of the same work from a museum store.

In your experience, where is the preferred market for painted copies of famous paintings?
Dr. Schierholz: A few years ago, the European market was massively flooded with Chinese copies, which were certainly of good quality. Initially, many of them were protected works and we regularly had to clear the stands at trade fairs. During one such event, I learned that the market for these copies was limited to Germany, Belgium and northern Italy. There are a few buyers in England, but there is zero interest in France and the Netherlands.

The interview was conducted by Heike Schlasse.

You can read more articles on the subject of art forgery in RESTAURO 7/2015.