MATSim User Meeting 2024: The use of MATSim for transport simulations and mobility

DARe Hub team member David Alvarez Castro shares his thoughts on innovations and projects showcased at the MATSim User Meeting 2024 in Espoo, Finland.

This June, I attended the MATSim User Meeting 2024 (MUM24), hosted at Aalto University in Espoo, Finland. This is the annual conference where MATSim (Multi-Agent Transport SIMulation) developers and users meet to discuss and showcase innovations and projects developed using the agent-based modelling (ABM) tool for transport simulation and mobility purposes.

Aalto University main campus. A multi-modal space where the use of sustainable modes (metro, tram, bicycle, scooter and walking) are combined.

A wide variety of topics were presented at this year’s conference, ranging from code optimisation for more reliable and faster simulations, to demand generation (e.g., activity plans), decarbonisation policies (e.g., stick and carrot), autonomous vehicles simulation, and socio-demographic analysis and visualisation tools. From the DARe Hub’s perspective, the most interesting presentations were on transport decarbonisation, the analysis of socio-demographic components, and the possibility of generating and visualising inputs (i.e., demand and supply) and analysing outputs in space and time.

The first group showed scenarios on the implementation of parking policies, mobility pricing, low-emissions zones and electric mobility. They all try to minimise the use of private and polluting cars in favour of more sustainable modes in order to reduce greenhouse gas emissions, which is in line with one of DARe’s ambitions. The second, presented by one of our Hub partners, Arup, exhibited the impact of mobility policies on different groups in society depending on their socio-demographic characteristics (age, gender, income), taking into account geographical components like where people live, work, and shop, a fundamental aspect when using ABMs to understand how people behave and act during their daily routines.

Last but not least, several tools to visualise the outcomes in space and time were shown (e.g., 3D and multi-scenario visualisations), besides the possibility of updating/modifying transport networks and public transport routes from a graphical user interface taking into account their geospatial geometries. This is a great improvement compared to previous alternatives, where the user has to deal with enormous, heavy and plain-coded xml files. Unfortunately, none of the presentations focused on transport resilience or the impact of weather events like rainfall, heat, and wind on the transport system. The combination of transport simulations and climate change seems, for now, to be mostly focused on transport decarbonisation.

Presentations were evenly distributed between academia and industry (55% and 45%, respectively), showing the current impact and acceptance of the tool in both environments. However, almost half of them came from Germany (47%), where the tool is used at all institutional levels (government, academia and industry), followed by France (16%), England (11%), and Sweden (11%) where the use of MATSim and other ABM tools is gaining momentum – since May 2024, the UK Department for Transport has included the use of agent-based methods in their TAG Unit M5.4. The vast majority of the audience and presenters were males, with only two presentations delivered by females, exhibiting the current gap of women in STEM and the work still needed to bring greater gender balance into these fields.

Main square at Aalto University

From the DARe Hub, we expect to contribute to next year’s meeting with some transport resilience scenarios. We are exploring how extreme weather events can be simulated with MATSim to quantify economic (e.g., loss), social (e.g., vulnerable groups) and environmental (e.g., increase of GHG emissions) impacts to transport systems and populations. In addition, several adaptation and mitigation measures can also be simulated to estimate their efficiency in reducing the exposure not only of the transport systems, but also of the population. The simulation of a flood-reinforced road network and the possibility of warning the population to avoid travelling in the event of an extreme weather condition are just a couple of examples that could be considered.

In terms of travelling, I was keen to follow Alistair’s example of a sustainable, low-carbon transport journey. As the meeting was just for one day, it was decided to reduce the time spent on a plane as much as possible in favour of more sustainable modes (i.e., public and active modes). Therefore, the York – London Heathrow leg was made by train, underground and on foot. Then, a direct flight from London to Helsinki, followed by train, bus and on foot to arrive at Espoo. Although the trip took longer than a two-flight connection, I achieved a substantial reduction in carbon emissions by travelling this way.


P.S. Unrelated with the conference but out of curiosity, I found a ‘napping pod’ at the university which is available to all students and staff. It seems that the Nordics are adopting the customs of southern Europe. 

The ‘napping pod’ to enjoy a ‘siesta’.