Abstract
Across the world, increasing numbers of cars in urban centers lead to congestion and adverse effects on public health as well as municipal climate goals. Reflecting cities’ ambitions to mitigate these issues, a growing body of research evaluates the use of innovative pricing strategies for parking, such as Dynamic Pricing (DP), to efficiently manage parking supply and demand. We contribute to this research by exploring the effects of Reinforcement Learning (RL)-based DP on urban parking. In particular, we introduce a theoretical framework for AI-based priced parking under traffic and social constraints. Furthermore, we present a portable and generalizable Agent-Based Model (ABM) for priced parking to evaluate our approach and run extensive experiments comparing the effect of several learners on different urban policy goals. We find that (1) outcomes are highly sensitive to the employed reward function; (2) trade-offs between different goals are challenging to balance; (3) single-reward functions may have unintended consequences on other policy areas (e.g., optimizing occupancy punishes low-income individuals disproportionately). In summary, our observations stress that fair DP schemes need to account for social policy measures beyond traffic parameters such as occupancy or traffic flow.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
All code required for reproducing this study is available at https://github.com/JakobKappenberger/ai-priced-parking.
- 2.
Due to a lack of empirical evidence, the parameters of the distribution were manually calibrated to preserve the correlation between income and WTP and to ensure the functioning of the underlying parking routines in the model, i.e. to avoid excluding low-income drivers completely from parking.
- 3.
To achieve a distribution in line with empirical evidence (approximately 0.28 vehicles per 100m road [23]), we calibrated the parameter to render a quarter of the drivers potential parking offenders.
- 4.
We chose this library based on a detailed review of RL frameworks by Nguyen et al. [24].
- 5.
A limiting factor is a relatively high volatility of prices and a lack of a priori knowledge of them by the agents. Therefore, in a real-world use case, policy measures would have to guarantee drivers a reserved parking space with a price fixed prior to them entering the city.
- 6.
The choice to use the occupancy levels of the individual CPZs instead of the global one was made as it otherwise would have been conceivable that the agent balances under- and full utilization of different CPZs to achieve a favorable score nonetheless.
- 7.
Running 35 simulations in parallel on a Intel Xeon Gold 6230 processor, training took approximately 39 h for \(\mathrm {r_{occupancy}}\).
- 8.
This tariff was chosen following the pricing policy in Mannheim.
- 9.
Analyzing a reward function’s median run should yield robust results concerning its reward score. However, individual features (e.g., the vanishing speed of cars from certain income classes) are likely still influenced by randomness.
- 10.
All experiments were conducted with a batch size of 36. Following Kuhnle et al. [16], entropy regularizations \(< 0.00001\) were set to 0.
References
Amato, G., Carrara, F., Falchi, F., Gennaro, C., Meghini, C., Vairo, C.: Deep learning for decentralized parking lot occupancy detection. Expert Syst. Appl. 72, 327–334 (2017). https://doi.org/10.1016/j.eswa.2016.10.055
Arnott, R., de Palma, A., Lindsey, R.: A temporal and spatial equilibrium analysis of commuter parking. J. Public Econ. 45(3), 301–335 (1991). https://doi.org/10.1016/0047-2727(91)90030-6. https://linkinghub.elsevier.com/retrieve/pii/0047272791900306. ISSN 0047-2727
Benenson, I., Martens, K., Birfir, S.: PARKAGENT: an agent-based model of parking in the city. Comput. Environ. Urban Syst. 32(6), 431–439 (2008). https://doi.org/10.1016/j.compenvurbsys.2008.09.011. https://linkinghub.elsevier.com/retrieve/pii/S0198971508000689. ISSN 0198-9715
Chatman, D., Manville, M.: Theory versus implementation in congestion-priced parking: an evaluation of SFpark, 2011–2012. Res. Transp. Econ. 44, 52–60 (2014). https://doi.org/10.1016/j.retrec.2014.04.005. https://linkinghub.elsevier.com/retrieve/pii/S0739885914000067. ISSN 0739-8859
Clementi, F., Gallegati, M.: Pareto’s law of income distribution: evidence for germany, the United Kingdom, and the United States. In: Salzano, M., et al. (eds.) Econophysics of Wealth Distributions, pp. 3–14. Springer, Milano (2005). https://doi.org/10.1007/88-470-0389-X_1. ISBN 978-88-470-0329-3 978-88-470-0389-7
Dieussaert, K., Aerts, K., Steenberghen, T., Maerivoet, S., Spitaels, K.: SUSTAPARK: an agent-based model for simulating parking search. In: 12th AGILE International Conference on Geographic Information Science Hannover, pp. 1–11. Association for Geographic Information Laboratories in Europe, Hannover (2009). https://agile-online.org/conference_paper/cds/agile_2009/agile_cd/pdfs/133.pdf
Eschmann, J.: Reward function design in reinforcement learning. In: Belousov, B., Abdulsamad, H., Klink, P., Parisi, S., Peters, J. (eds.) Reinforcement Learning Algorithms: Analysis and Applications. SCI, vol. 883, pp. 25–33. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-41188-6_3
Friesen, M., Mingardo, G.: Is parking in Europe ready for dynamic pricing? A reality check for the private sector. Sustainability 12(7), 2732 (2020). https://doi.org/10.3390/su12072732. https://www.mdpi.com/2071-1050/12/7/2732. ISSN 2071-1050
Gilbert, N.: Agent-Based Models. SAGE Publications Inc, Thousand Oaks (2020). https://doi.org/10.4135/9781506355580. https://methods.sagepub.com/book/agent-based-models-2e. ISBN 978-1-5063-5560-3 978-1-5063-5558-0
Glazer, A., Niskanen, E.: Parking fees and congestion. Reg. Sci. Urban Econ. 22(1), 123–132 (1992). https://doi.org/10.1016/0166-0462(92)90028-Y. https://linkinghub.elsevier.com/retrieve/pii/016604629290028Y. ISSN 0166-0462
Hampshire, R., Shoup, D.: How much traffic is cruising for parking? Transfers Magazine, vol. 4 (2019). https://transfersmagazine.org/magazine-article/issue-4/how-much-traffic-is-cruising-for-parking/
Inci, E.: A review of the economics of parking. Econ. Transp. 4(1–2), 50–63 (2015). https://doi.org/10.1016/j.ecotra.2014.11.001. https://linkinghub.elsevier.com/retrieve/pii/S2212012214000562. ISSN 2212-0122
Jakob, M., Menendez, M.: Parking pricing vs. congestion pricing: a macroscopic analysis of their impact on traffic. Transportmetrica A: Transp. Sci. 17(4), 462–491 (2021). https://doi.org/10.1080/23249935.2020.1797924
Jaxa-Rozen, M., Kwakkel, J.H.: PyNetLogo: linking NetLogo with python. J. Artif. Soc. Soc. Simul. 21(2), 4 (2018). https://doi.org/10.18564/jasss.3668. http://jasss.soc.surrey.ac.uk/21/2/4.html. ISSN 1460-7425
Jeihani, M., Ardeshiri, A., Du, J., Rakha, H.: Drivers willingness to pay progressive rate for street parking. Technical report, U.S. Department of Transportation, January 2015. https://www.mautc.psu.edu/docs/MAUTC-2013-04.pdf
Kuhnle, A., Schaarschmidt, M., Fricke, K.: Tensorforce: A Tensorflow Library for Applied Reinforcement Learning (2017). https://github.com/tensorforce/tensorforce
Litman, T.: Parking Management Best Practices. Routledge, New York (2019). ISBN 978-0-367-33012-5, oCLC: 1110487321
Mannion, P., Devlin, S., Mason, K., Duggan, J., Howley, E.: Policy invariance under reward transformations for multi-objective reinforcement learning. Neurocomputing 263, 60–73 (2017). https://doi.org/10.1016/j.neucom.2017.05.090. https://linkinghub.elsevier.com/retrieve/pii/S0925231217311037. ISSN 0925-2312
Manville, M.: Parking pricing. In: Ison, S., Mulley, C. (eds.) Transport and Sustainability, vol. 5, pp. 137–155. Emerald Group Publishing Limited, Bingley, UK (2014). https://doi.org/10.1108/S2044-994120140000005019. https://www.emerald.com/insight/content/doi/10.1108/S2044-994120140000005019/full/html. ISBN 978-1-78350-919-5 978-1-78350-920-1
Marsden, G.: Parking policy. In: Ison, S., Mulley, C. (eds.) Transport and Sustainability, vol. 5, pp. 11–32. Emerald Group Publishing Limited, Bingley, UK (2014). https://doi.org/10.1108/S2044-994120140000005016. https://www.emerald.com/insight/content/doi/10.1108/S2044-994120140000005016/full/html. ISBN 978-1-78350-919-5 978-1-78350-920-1
Meyer, M.D.: Demand management as an element of transportation policy: using carrots and sticks to influence travel behavior. Transp. Res. Part A: Policy Pract. 33(7–8), 575–599 (1999). https://doi.org/10.1016/S0965-8564(99)00008-7. https://linkinghub.elsevier.com/retrieve/pii/S0965856499000087. ISSN 0965-8564
Mingardo, G., van Wee, B., Rye, T.: Urban parking policy in Europe: a conceptualization of past and possible future trends. Transp. Res. Part A: Policy Pract. 74, 268–281 (2015). https://doi.org/10.1016/j.tra.2015.02.005. https://linkinghub.elsevier.com/retrieve/pii/S0965856415000221. ISSN 0965-8564
Morillo, C., Campos, J.M.: On-street illegal parking costs in urban areas. Procedia - Soc. Behav. Sci. 160, 342–351 (2014). https://doi.org/10.1016/j.sbspro.2014.12.146. https://www.sciencedirect.com/science/article/pii/S1877042814062478. xI Congreso de Ingenieria del Transporte (CIT 2014). ISSN 1877-0428
Nguyen, N.D., Nguyen, T.T., Nguyen, H., Creighton, D., Nahavandi, S.: Review, Analysis and Design of a Comprehensive Deep Reinforcement Learning Framework. arXiv e-print 2002.11883 (2021). https://arxiv.org/abs/2002.11883
Nourinejad, M., Roorda, M.J.: Impact of hourly parking pricing on travel demand. Transp. Res. Part A: Policy Pract. 98, 28–45 (2017). https://doi.org/10.1016/j.tra.2017.01.023. https://linkinghub.elsevier.com/retrieve/pii/S0965856415303396. ISSN 0965-8564
OECD: Under Pressure: The Squeezed Middle Class. OECD, May 2019. https://doi.org/10.1787/689afed1-en. https://www.oecd-ilibrary.org/social-issues-migration-health/under-pressure-the-squeezed-middle-class_689afed1-en. ISBN 978-92-64-54283-9 978-92-64-15034-8 978-92-64-98251-2 978-92-64-69927-4
Pierce, G., Shoup, D.: Getting the prices right: an evaluation of pricing parking by demand in San Francisco. J. Am. Plann. Assoc. 79(1), 67–81 (2013). https://doi.org/10.1080/01944363.2013.787307. http://www.tandfonline.com/doi/abs/10.1080/01944363.2013.787307. ISSN 0194-4363, 1939-0130
Proost, S., Van Dender, K.: Optimal urban transport pricing in the presence of congestion, economies of density and costly public funds. Transp. Res. Part A: Policy Pract. 42(9), 1220–1230 (2008). https://doi.org/10.1016/j.tra.2008.03.009. https://linkinghub.elsevier.com/retrieve/pii/S0965856408000864. ISSN 0965-8564
Saharan, S., Kumar, N., Bawa, S.: An efficient smart parking pricing system for smart city environment: a machine-learning based approach. Future Gener. Comput. Syst. 106, 622–640 (2020). https://doi.org/10.1016/j.future.2020.01.031. https://linkinghub.elsevier.com/retrieve/pii/S0167739X19322496. ISSN 0167-739X
Shoup, D.: The High Cost of Free Parking. Planners Press, American Planning Association, Chicago, updated edn. (2011). ISBN 978-1-932364-96-5, oCLC: ocn742069668
Shoup, D.: Free parking or free markets. In: Shoup, D. (ed.) Parking and the City, 1st edn, pp. 270–275. Routledge, New York (2018). https://doi.org/10.4324/9781351019668. https://www.taylorfrancis.com/books/9781351019651. ISBN 978-1-351-01966-8
Simhon, E., Liao, C., Starobinski, D.: Smart parking pricing: a machine learning approach. In: 2017 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), Atlanta, GA, pp. 641–646. IEEE, May 2017. https://doi.org/10.1109/INFCOMW.2017.8116452. http://ieeexplore.ieee.org/document/8116452/. ISBN 978-1-5386-2784-6
Vickrey, W.: Statement to the joint committee on Washington, DC, metropolitan problems (with a foreword by Richard Arnott and Marvin Kraus). J. Urban Econ. 36(1), 42–65 (1994). https://doi.org/10.1006/juec.1994.1025. https://linkinghub.elsevier.com/retrieve/pii/S0094119084710254. ISSN 0094-1190
Wilensky, U.: Netlogo Traffic Grid Model (2003). http://ccl.northwestern.edu/netlogo/models/TrafficGrid
Acknowledgements
This research was conducted as part of the grant “Consequences of Artificial Intelligence for Urban Societies (CAIUS),” funded by Volkswagen Foundation. We would like to thank the students of our spring 2020 research seminar “Social Simulation,” who developed an early prototype of our ABM in NetLogo: Madeleine Aziz, Jens Daube, Paul Exner, Jakob Kappenberger (née Gutmann), Jonas Klenk, and Aamod Vyas. We furthermore would like to thank Frederic Gerdon and the other members of project CAIUS for their helpful feedback to an earlier version of this article.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendices
Methodology
Table 1 lists the information contained in \(S_t\), and Fig. 4 shows four out of the five deployed reward functions. Figures 5 and 6 visualize the hyperparameter tuning processes that resulted in the utilized model configurations for the ABM.
Parameters for Reproduction
Table 2 contains the parameters of our Netlogo ABM, and Table 3 lists the final hyperparameters across all reward functions.Footnote 10
Detailed Results
Figs. 7, 8, 9, 10, 11, 12 and 13 show in-depth results with CPZ fees, occupancy, number of cars, social composition, and car speed per reward function against the time of day.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Kappenberger, J., Theil, K., 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
Download citation
DOI: https://doi.org/10.1007/978-3-031-19097-1_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-19096-4
Online ISBN: 978-3-031-19097-1
eBook Packages: Computer ScienceComputer Science (R0)