Using Impact-Aware AI to Make Smart Cities Socially Equitable
News from the project
Florian Rupp defended his doctoral thesis.
On December 11th, 2025, Florian Rupp successfully defended his doctoral thesis “Combining Procedural Content Generation and Heuristic …
Presentation at In_equality Conference 2026
Frederic Gerdon will present our latest research on algorithmic refugee placement systems at the In_equality Conference 2026 in Konstanz (15-17 …
Full Grant Project (2021 - 2025)
AI systems help to efficiently allocate scarce public resources and are at the core of many smart city activities. Yet, the same systems may also result in unintended societal consequences, particularly by reinforcing social inequalities. CAIUS will identify and analyze such consequences. To this end, we develop an innovative methodology combining expertise from computer science and social science. Using agent-based models (ABM), we analyze the effects of AI-based decisions on societal macro variables of social inequality such as income disparity. The data input for these ABMs consists of both Open Government Data, digital traces, and own surveys. The goal is to train AI systems to account for their social consequences within specific fairness constraints; this synthesis of ABM and fair reinforcement learning lays the groundworks for what we call “Impact-aware AI” in urban contexts. With CAIUS, we accompany two smart city applications planned by partners in the Rhine-Neckar Metropolitan Region: dynamic pricing of parking space and traffic law enforcement via Internet-of-Things sensors. Our results contribute to research of human-AI interaction and will be condensed into general guidelines for decision-makers regarding the ethical implementation of AI-based decision-making systems in urban contexts.
Planning Phase (2019-2020)
“Smart Cities” is the buzzword for the development of data-driven processes in the public and governmental space to improve life quality, enable new applications or simply to optimise processes for more efficiency. Artificial intelligence plays a key role in the implementation of smart city services, for example in detecting patterns in the usage of public services to optimize the service quality, or in the identification of unusual behaviour or measurements to trigger an alarm. The basis for such smart services is always data, which can be generated specifically for these services (such as the installation of air pollution sensors) or which can be made available from existing government data or other data sources which are relevant for the general public (e.g. traffic information, public transportation). This project focuses on a potential negative effect of smart cities with the question: Where do smart city applications lead to potential erosion of solidarity of the urban society?
Erosion of solidarity (or desolidarisation) in the context of big data / data mining applications is a well-known, yet not a well researched problem, where profiling and high personalization leads to disadvantages for people with a bad profile, or with no profile at all. Similar processes also exist in smart city applications. This notion of desolidarisation processes in the context of artificial intelligence applications, with a main focus on the urban society, is the central theme for this project.
Simulation
Using agent-based models (ABM), we analyze the effects of AI-based decisions on societal macro variables of social inequality such as income disparity. The data input for these ABMs consists of both Open Government Data, digital traces, and own surveys. The goal is to train AI systems to account for their social consequences within specific fairness constraints; this synthesis of ABM and fair reinforcement learning lays the groundworks for what we call “Impact-aware AI” in urban contexts.
We use agent-based models and fair machine learning to model the impact of smart city policies on society
People
Professors
Researchers
Research
Publications
2025
- Kappenberger, J., Strasser Ceballos, C., Gerdon, F., Szafran, D., Rupp, F., Eckert, K., Stuckenschmidt, H., Bach, R., Kreuter, F., and Kern, C. (2025). Unintended Impacts of Automation for Integration? Simulating Integration Outcomes of Algorithm-Based Refugee Allocation in Germany. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society 8 (2), 1375-1387. https://doi.org/10.1609/aies.v8i2.36638
- Szafran, D., & Bach, R. L. (2025). Unfamiliar but desired: citizens’ attitudes toward smart city applications. AI & SOCIETY, 1-18. https://doi.org/10.1007/s00146-025-02815-8
- Novotny, M., Weber, W., Kern, C., and Kreuter, F. (2025). Measuring Public Opinion Towards Artificial Intelligence: Development and Validation of a General AI Attitude Short Scale. AI & Society. https://doi.org/10.1007/s00146-025-02478-5
- Strasser Ceballos, C., Novotny, M., and Kern, C. (2025). Beyond Proxy Variables: Extending Refugee Allocation Algorithms for Equitable Predictions. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society 8 (3), 2442-2455. https://doi.org/10.1609/aies.v8i3.36729
- Strasser Ceballos, C. and Kern, C. (2025). Location Matching on Shaky Grounds: Re-Evaluating Algorithms for Refugee Allocation. In Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, New York, NY, USA, 2180–2199. https://doi.org/10.1145/3715275.3732149
- Strasser Ceballos, C., Novotny, M., and Kern, C. (2025). Re-evaluating the role of refugee integration factors for building more equitable allocation algorithms. Proceedings of Fourth European Workshop on Algorithmic Fairness, PMLR 294:428–433. https://proceedings.mlr.press/v294/ceballos25a.html.
- Kern, C., Kappenberger, J., Gerdon, F., Strasser Ceballos, C., Szafran, D., Rupp,F., and Bach, R. (2025). A Simulation Framework for Studying the Social Impacts of Algorithm-Based Refugee Matching. Proceedings of Fourth European Workshop on Algorithmic Fairness, PMLR 294:487-491. https://proceedings.mlr.press/v294/kern25a.html.
- Rupp, F., Eckert, K. (2025). Level the Level: Balancing Game Levels for Asymmetric Player Archetypes With Reinforcement Learning. In Proceedings of the 20th International Conference on the Foundations of Digital Games. https://doi.org/10.1145/3723498.3723747
2024
- Schenk, P. O. and Kern, C. (2024). Connecting Algorithmic Fairness to Quality Dimensions in Machine Learning in Official Statistics and Survey Production. arXiv. https://arxiv.org/abs/2402.09328
- Jaime, S. and Kern, C. (2024). Ethnic Classifications in Algorithmic Fairness: Concepts, Measures and Implications in Practice. In the 2024 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’24). Association for Computing Machinery, New York, NY, USA, 237–253. https://doi.org/10.1145/3630106.3658902.
- Simson, J., Fabris, A. and Kern, C. (2024). Lazy Data Practices Harm Fairness Research. In the 2024 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’24). Association for Computing Machinery, New York, NY, USA, 642–659. https://doi.org/10.1145/3630106.3658931.
- Simson, J., Pfisterer, F. and Kern, C. (2024). One Model Many Scores: Using Multiverse Analysis to Prevent Fairness Hacking and Evaluate the Influence of Model Design Decisions. In the 2024 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’24). Association for Computing Machinery, New York, NY, USA, 1305–1320. https://doi.org/10.1145/3630106.3658974
- Fischer-Abaigar, U., Kern, C. and Kreuter, F. (2024). The Missing Link: Allocation Performance in Causal Machine Learning. Humans, Algorithmic Decision-Making and Society: Modeling Interactions and Impact, co-located with ICML 2024. https://arxiv.org/abs/2407.10779.
- Szafran, D. and Bach, R. L. (2024). The Human Must Remain the Central Focus: Subjective Fairness Perceptions in Automated Decision-Making. Minds and Machines, 34(3), 24. https://doi.org/10.1007/s11023-024-09684-y
- F. Rupp, A. Puddu, C. Becker-Asano and K. Eckert. (2024). It might be balanced, but is it actually good? An Empirical Evaluation of Game Level Balancing. 2024 IEEE Conference on Games (CoG), Milan, Italy, 2024, pp. 1-4, https://doi.org/10.1109/CoG60054.2024.10645642.
- F. Rupp and K. Eckert. (2024). G-PCGRL: Procedural Graph Data Generation via Reinforcement Learning. 2024 IEEE Conference on Games (CoG), Milan, Italy, 2024, pp. 1-8, https://doi.org/10.1109/CoG60054.2024.10645633.
- F. Rupp and K. Eckert. (2024) GEEvo: Game Economy Generation and Balancing with Evolutionary Algorithms. 2024 IEEE Congress on Evolutionary Computation (CEC), Yokohama, Japan, 2024, pp. 1-8, https://doi.org/10.1109/CEC60901.2024.10612054
- Rupp F., Eberhardinger M., and Eckert K. (2024). Simulation-Driven Balancing of Competitive Game Levels with Reinforcement Learning, IEEE Transactions on Games, pp. 1–11, 2024. https://doi.org/10.1109/TG.2024.3399536
- Rupp F., Schnabel B., and Eckert K. (2024). Implementing Data Workflows and Data Model Extensions with RDF-star. The Electronic Library (TEL), Emerald Publishing Limited. https://doi.org/10.1108/EL-04-2023-0102
2023
- Bach, R. L., Kern, C., Mautner, H., and Kreuter, F. (2023). The Impact of Modeling Decisions in Statistical Profiling. Data & Policy 5, E32. https://doi.org/10.1017/dap.2023.29
- Rupp F., Eberhardinger M., and Eckert K. (2023). Balancing of competitive two-player Game-Levels with Reinforcement Learning. IEEE Conference on Games (CoG). https://doi.org/10.1109/CoG57401.2023.10333248
- Fischer Abaigar, U., Kern, C., Barda, N., Kreuter, F. (2023). Bridging the Gap: Towards an Expanded Toolkit for ML-Supported Decision-Making in the Public Sector. arXiv. https://arxiv.org/abs/2310.19091
- Szafran D. and Bach R. L. (2023): The human must remain the central focus: Subjective fairness perceptions in algorithmic decision-making. ESRA 2023 Conference.
2022
- Kern C., Gerdon F., Bach R. L., Keusch F., and Kreuter F. (2022). Humans versus Machines: Who is Perceived to Decide Fairer? Experimental Evidence on Attitudes Toward Automated Decision-Making. Patterns. https://doi.org/10.1016/j.patter.2022.100591
- Gerdon F., Bach R. L., Kern C., and Kreuter, F. (2022). Social Impacts of Algorithmic Decision-Making: A Research Agenda for the Social Sciences. Big Data & Society. https://doi.org/10.1177/20539517221089305
- Kappenberger J., Theil K. and Stuckenschmidt, H. (2022). Evaluating the Impact of AI-Based Priced Parking with Social Simulation. In: Hopfgartner, F., Jaidka, K., Mayr, P., Jose, J., Breitsohl, J. (eds) Social Informatics. SocInfo 2022. Lecture Notes in Computer Science, vol 13618. Springer, Cham. https://doi.org/10.1007/978-3-031-19097-1_4
- Rupp F., Schnabel B., and Eckert K. (2022). Easy and Complex: New Perspectives for Metadata Modeling Using RDF-Star and Named Graphs. In: Villazón-Terrazas, B., Ortiz-Rodriguez, F., Tiwari, S., Sicilia, MA., Martín-Moncunill, D. (eds) Knowledge Graphs and Semantic Web . KGSWC 2022. Communications in Computer and Information Science, vol 1686. Springer, Cham. https://doi.org/10.1007/978-3-031-21422-6_18
Presentations
- Gerdon, Frederic, Kilian Theil, Christoph Kern, Ruben Bach, Frauke Kreuter, Heiner Stuckenschmidt, and Kai Eckert (2020). Exploring impacts of artificial intelligence on urban societies with social simulations. 40th Conference of the German Sociological Association.
- Kappenberger, Theil, Stuckenschmidt (2022). Evaluating the Impact of AI-Based Priced Parking with Social Simulation. 13th International Conference on Social Informatics, Glasgow, Scotland
- Rupp, Schnabel, Eckert (2022). Easy and Complex: New Perspectives for Metadata Modeling Using RDF-Star and Named Graphs. 4th Knowledge Graph and Semantic Web Conference, Madrid, Spain.
- Gerdon, Szafran, Kappenberger, Bach, and Kern (2023). Using survey experiments and agent-based modeling to simulate mobility behavior in smart cities. ESRA 2023 Conference, Milan, Italy.
- Kern, Gerdon, Bach, Keusch, and Kreuter (2023): Humans vs. Machines: Who is Perceived to Decide Fairer? An Experiment about Citizens’ Attitudes. The 2nd International Conference on Hybrid Human-Artificial Intelligence, Munich, Germany.
- Rupp, Eberhardinger, Eckert (2023). Balancing of competitive two-player Game-Levels with Reinforcement Learning. IEEE Conference on Games (CoG), Boston, MA, USA.
- Rupp, Eckert (2024). GEEvo: Game Economy Generation and Balancing with Evolutionary Algorithms. IEEE Congress on Evolutionary Computation (CEC), Yokohama, Japan.
- Rupp, Eckert (2024). G-PCGRL: Procedural Graph Data Generation via Reinforcement Learning. 2024 IEEE Conference on Games, Milan, Italy, 2024.
- Rupp, Eckert (2024). It might be balanced, but is it actually good? An Empirical Evaluation of Game Level Balancing. 2024 IEEE Conference on Games, Milan, Italy, 2024.