Abstract
This paper discusses use of the widely used transport simulator, MATSim, as one component in a large complex agent based microsimulation where dynamic changes in the environment require the agents to be reactive as well as goal directed. We describe a number of refinements to MATSim that have been made to facilitate its use within our deployed wildfire evacuation applications, as well as some tools that have been developed which complement MATSim. All code is freely available under open source licenses. As applications increasingly require complex microsimulations, with many aspects, it is important to use existing software where possible. However most simulation systems, like MATSim, have been developed as standalone systems. We identify ways that MATSim has needed to be extended or modified in order for it to be used as a component in a larger whole. The paper provides details that will be useful for anyone wanting to use MATSim within their specific application.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
The code for these models is accessible from https://github.com/agentsoz/ees.
- 2.
A plan encompasses activities, trips which contain the (possibly multi-modal) movement between activities, and routes which are the detailed road/path segments to be traversed by a vehicle/person.
- 3.
http://matsim.org/javadoc \(\rightarrow \) matsim main \(\rightarrow \) EditPlans, EditRoutes, EditTrips.
- 4.
The custom AgentInCongestionEvent is triggered on LinkLeaveEvent if \( (t_{k,i} - t^{*}_{k,i})/t^{*}_{k,i} > w \ , \) where k is some traversed link, i is the current link, time \(t_{k,i}\) is the recorded travel time for the route taken from k to i, and \(t^{*}_{k,i}\) is the expected travel time if travelling at freespeed on that route. The constant w is the congestion tolerance threshold. Practically, we set a time period T for congestion evaluation and take the maximum permissible \(t_{k,i}\) such that \(t_{k,i} \le T\). For instance, \(T = 300, w = 0.4\) means that an agent will consider itself to be stuck in congestion if over the last 5 min, the time delay in travelling the route from k to i was greater than 40% of the expected travel time for that route. – The nextLinkBlocked event is generated when the following link has freespeed close to zero as the intent is to prevent the agent from entering a blocked link where it might get stuck forever.
- 5.
- 6.
Software is available at https://github.com/agentsoz/synthetic-population.
- 7.
This software can be accessed at https://github.com/agentsoz/ees-synthetic-population/tree/master/plan-algorithm.
- 8.
The aim is not to build calibrated populations, but instead build representative populations that capture sufficient richness of activities while being relatively easy for domain experts to specify. However this tool is potentially suitable only for relatively small geographical areas in its current state, as with large scale scenarios, a random destination choice with a uniform distribution leads to distances that are too large (in the average half the scenario diameter). Nevertheless it has been useful for the current applications and is an area of ongoing work.
- 9.
An intention is simply the code stack resulting from a top-level instantiated goal.
- 10.
- 11.
Macbook Pro 15,2 with 4 i7 cores (2.7 GHz) and 16 GB RAM.
- 12.
The reported metric is the simulation to real time ratio (s/r), i.e. how much faster than reality the simulation is.
References
Anda, C., Erath, A., Fourie, P.J.: Transport modelling in the age of big data. Int. J. Urban Sci. 21(sup1), 19–42 (2017). https://doi.org/10.1080/12265934.2017.1281150
Balmer, M., Axhausen, K., Nagel, K.: A demand generation framework for large scale micro simulations. Transp. Res. Rec. 1985, 125–134 (2006). https://doi.org/10.3141/1985-14
Balmer, M., Rieser, M., Vogel, A., Axhausen, K., Nagel, K.: Generating day plans using hourly origin-destination matrices. In: Bieger, T., Laesser, C., Maggi, R. (eds.) Jahrbuch 2004/05 Schweizerische Verkehrswirtschaft, pp. 5–36, Schweizer Verkehrswissenschaftliche Gesellschaft (2005)
Batty, M.: Fifty years of urban modeling: macro-statics to micro-dynamics. In: The Dynamics of Complex Urban Systems: An Interdisciplinary Approach, Ascona/Ticino, Switzerland, 4–6 November 2004, p. 27 (2004)
Bazzan, A., Amarante, M.D.B.D., Sommer, T., Benavides, A.J.: ITSUMO: an agent-based simulator for ITS applications. In: Rossetti, R., Liu, H., Tang, S. (eds.) Proceedings of the 4th Workshop on Artificial Transportation Systems and Simulation. IEEE, September 2010
Bouman, P., Lovric, M.: Rotterdam: revenue management in public transportation with smart-card data enabled agent-based simulations. In: Horni, A., Nagel, K., Axhausen, K.W. (eds.) The Multi-Agent Transport Simulation MATSim, chap. 81. Ubiquity, London (2016). https://doi.org/10.5334/baw
Bruch, E., Atwell, J.: Agent-based models in empirical social research. Sociol. Methods Res. 44(2), 186–221 (2015)
Dahmann, J.S., Kuhl, F., Weatherly, R.: Standards for simulation: as simple as possible but not simpler: the high level architecture for simulation. Simulation 71(6), 378–387 (1998)
Dobler, C., Nagel, K.: Within-day replanning. In: Horni, A., Nagel, K., Axhausen, K.W. (eds.) The Multi-Agent Transport Simulation MATSim, chap. 30, pp. 187–200. Ubiquity Press, London (2016)
Epstein, J., Axtell, R.: Growing Artificial Societies. MIT Press, Cambridge (1996)
Gloor, C., Cavens, D., Nagel, K.: A message-based framework for real-world mobility simulations. In: Klügl, F., Bazzan, A., Ossowski, S. (eds.) Applications of Agent Technology in Traffic and Transportation, Whitestein Series in Software Agent Technologies and Autonomic Computing, pp. 193–209. Birkhäuser, Basel (2005). https://doi.org/10.1007/3-7643-7363-6_13
Hamacher, H., Tjandra, S.: Mathematical modelling of evacuation problems: a state of art. Technical report, Fraunhofer ITWM (2001)
Horni, A., Nagel, K., Axhausen, K.W. (eds.): The Multi-Agent Transport Simulation MATSim. Ubiquity, London (2016). https://doi.org/10.5334/baw
Lämmel, G., Klüpfel, H., Nagel, K.: Risk minimizing evacuation strategies under uncertainty. In: Peacock, R., Kuligowski, E., Averill, J. (eds.) Pedestrian and Evacuation Dynamics, pp. 287–296. Springer, Berlin (2011). https://doi.org/10.1007/978-1-4419-9725-8_26
Message Passing Interface (MPI) web page. https://www.mpi-forum.org. Accessed 2019
Nicolai, T.W., Nagel, K.: Coupling MATSim and UrbanSim: software design issues. SustainCity Working Paper 6.1, also VSP WP 10–13 (2010). http://www.vsp.tu-berlin.de/publications
Padgham, L., Nagel, K., Singh, D., Chen, Q.: Integrating BDI agents into a MATSim simulation. In: ECAI 2014, pp. 681–686. IOS Press, Prague (2014)
Padgham, L., Singh, D.: Making MATSim agents smarter with the Belief-Desire-Intention framework. In: Horni, A., Nagel, K., Axhausen, K.W. (eds.) The Multi-Agent Transport Simulation MATSim, chap. 31, pp. 201–210. Ubiquity Press, London (2016)
Padgham, L., Winikoff, M.: Developing Intelligent Agent Systems: A practical guide. Wiley Series in Agent Technology. Wiley, Hoboken (2004)
Protocol Buffers Web. https://developers.google.com/protocol-buffers/. Accessed 2015
Rieser, M., Nagel, K., Beuck, U., Balmer, M., Rümenapp, J.: Truly agent-oriented coupling of an activity-based demand generation with a multi-agent traffic simulation. Transp. Res. Rec. 2021, 10–17 (2007). https://doi.org/10.3141/2021-02
Russel, S.J., Norvig, P.: Artificial Intelligence - A Modern Approach, 3rd edn. Pearson Education, Upper Saddle River (2010)
Singh, D., Padgham, L.: OpenSim: a framework for integrating agent-based models and simulation components. In: ECAI 2014, pp. 837–842. IOS Press, Prague (2014)
Singh, D., Padgham, L.: Emergency Evacuation Simulator (EES) - a tool for planning community evacuations in Australia. In: Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence Organization, pp. 5249–5251, Melbourne, August 2017. https://doi.org/10.24963/ijcai.2017/780
Singh, D., Padgham, L., Logan, B.: Integrating BDI agents with agent-based simulation platforms. Auton. Agent. Multi-Agent Syst. 30(6), 1050–1071 (2016). https://doi.org/10.1007/s10458-016-9332-x
Strauch, D., et al.: Linking Transport and Land Use Planning: The Microscopic Dynamic Simulation Model ILUMASS, chap. 20, pp. 295–311. CRC Press, Boca Raton (2005)
W3C: eXtensible Markup Language (XML). World Wide Web Consortium (W3C) (2008). www.w3.org/XML
Waddell, P., Ševčíková, H., Socha, D., Miller, E., Nagel, K.: OPUS: an open platform for urban simulation. In: 9th Conference on Computers in Urban Planning and Urban Management (CUPUM), vol. 428, University College London, UK (2005). http://128.40.111.250/cupum/searchpapers/detail.asp?pID=428
Waddell, P., Borning, A., Noth, M., Freier, N., Becke, M., Ulfarsson, G.: Microsimulation of urban development and location choices: design and implementation of urbansim. Netw. Spat. Econ. 3(1), 43–67 (2003)
Weiss, G. (ed.): Multiagent Systems. A Modern Approach to Distributed Artificial Intelligence. The MIT Press (1999)
Zilske, M., Nagel, K.: A simulation-based approach for constructing all-day travel chains from mobile phone data. Procedia Comput. Sci. 52, 468–475 (2015). https://doi.org/10.1016/j.procs.2015.05.017
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Singh, D., Padgham, L., Nagel, K. (2020). Using MATSim as a Component in Dynamic Agent-Based Micro-Simulations. In: Dennis, L., Bordini, R., Lespérance, Y. (eds) Engineering Multi-Agent Systems. EMAS 2019. Lecture Notes in Computer Science(), vol 12058. Springer, Cham. https://doi.org/10.1007/978-3-030-51417-4_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-51417-4_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-51416-7
Online ISBN: 978-3-030-51417-4
eBook Packages: Computer ScienceComputer Science (R0)