MAGnUM, a study that plays with traffic jams in order to prevent them

Ludovic Leclercq, Research Director in Traffic Modelling, at Université Gustave Eiffel

Researchers from the MAGnUM project have developed an innovative strategy based on large-scale simulations, to create better strategies for regulating traffic flows. Now, they are making the data from their six years of research available to the public.

According to Insee, nearly nine out of ten French people live in a city or in a city's catchment area. As a consequence, traffic jams have become a daily reality for many of us, with the abundantly documented array of related problems. To unclog the blockages of our urban areas, public authorities are concentrating their efforts on developing alternative modes of transport.

Led by researcher Ludovic Leclercq, director of the laboratory LICIT-ECO7, a joint unit of Université Gustave Eiffel and ENTPE, and supported by the European Research Council (ERC), the MAGnUM¹ project used a complementary approach for its experiment. It aims to better regulate traffic flows, by optimizing transport systems as well as individual decisions in our journeys.

The objective was to develop models to represent the behaviour of users of multimodal networks (cars and public transport).

To better understand the choices we make, researchers used “serious games”. These are computer simulations in which multiple participants must get virtually from point A to point B.

The other unique thing about the approach is its scale: the real-life network of the Lyon greater urban area was used as the main setting for the game, which allowed researchers to test a large number of configurations.

The simulation was based on the three main choices that we make when taking a trip: the route, the mode of transport, and the departure time. Participants were free to choose these variables. The sessions made it possible to gather their reactions to the options being offered to them in real-time, in changing traffic conditions.

In practice, the decisions that seem optimal to us do not necessarily correspond to the collective interest. For example, by giving a lot of drivers the same alternative routes, GPS systems can contribute to network congestion.

Researchers estimate that if individual decisions were better organised, a transport system could be improved by 15 to 20%. They therefore tested the reaction to recommending routes likely to reduce traffic jams or cause less CO₂ emissions, even if this meant trips were longer.

The data sets made available may be of interest to researchers, design offices and transport authorities. 

One of the major advances from project MAGnUM and its spin-off, MAGnUM-plus (also funded by ERC), is a prototype of a navigation application focused on seeking to optimal solution for the collective.

The team has made available the observations from nearly 700 trips during the project’s development (1,300 planned by the end). The behavioural data collected from uses of the application could be used for research into user acceptance of different recommendations.

For reproducible research, all the raw observations made during the simulation sessions could be used for other applications needing to evaluate how acceptable users consider different navigation advice.

¹For “Multiscale and multimodal traffic modelling approach for sustainable management of urban mobility”.

Interview performed by the Kogito agency

Serious game: A serious game refers to a game which has a main aim other than entertainment.  It is an activity that combines a “serious” intention (information, education, training, research, etc.) with a fun approach from the world of games. It may be a board game, role play or video game.

Reproducible research: A piece of research is said to be reproducible if all the information concerning the work, including, but not limited to, the text, data and programming code, are made available so that any independent researcher can reproduce the results (Vandewalle et al., 2009).