Research Projects

Stop the Video

Research Projects

STATUS: Complete YEAR: 2015 TOPIC AREA: Sustainability, energy, and health CENTER: MetroFreight

A new look at the environmental assessment of logistics sprawl Part 1

Project Summary

Urban sprawl does not only relate to residential patterns. Depending on their needs for land space or proximity to suppliers and customers, some economic activities may also decentralize within a given urban area. As a matter of fact, “logistic sprawl” (LS) - understood as the increase in mean distances between the city centre (or their center of gravity) and warehouses - has been observed by the literature. Because this trend generally generates an increase in total kilometers driven by freight vehicles, many researchers consider the environmental impact of LS to be negative.

This research project intends to question such common beliefs. We argue that a proper analysis of LS should not only look at the changing volume of kilometers driven within an urban agglomeration, but also at the evolution of their distribution according to the density of population in areas where freight movements are made.

To support this idea, we first document the U-shape relationship between traffic speed and the energy consumption of light- or heavy-vehicles, the latter implying the emission of CO2 and local air pollutants. In addition, we show that the engines' noise and risks of accidents also depend on freight vehicles' velocity. Since traffic speed decreases with respect to the density of vehicles in a given area - itself correlated with the density of population therein -, it is possible to express the total volume of environmental "externalities" generated by freight activities (CO2, local pollutants, noise, accidents) as a function of the various density levels within a given urban area. Put differently, our modeling approach aims at accounting for both the unitary nuisances implied by one kilometer driven by freight vehicles, as well as for the number of people exposed to these nuisances. Most of the required parameters defining the "externalities-density" relationship either can be drawn from technical studies or estimated with available (free-access) data.

In a second step, we rely on simulation methods to get information on the total volume of kilometers driven by freight vehicles before/after LS happened, but also on the distribution of these kilometers according to the zones where they are realized. Our case study will be the Paris region, where LS has been already documented and for which we have input data allowing us to implement the simulation methods. Crossing the distributions of kilometer with the theoretical "externalities-density" relationship, we will be able to assess the environmental impact of LS in this area. Because we study a dynamic phenomenon, we will pay attention to various counterfactual scenarios such as: no-change in the technology of vehicles (in terms of noise, energy consumption), no change in the total number of warehouses and/or in the total number of inhabitants (fixed supply/demand).



Martin Koning
Senior Researcher, Economics
14-20 boulevard Newton, Cite Descartes
Marne la Vallee cedex 2, 77447
[email protected]


Nicolas Coulombel
Senior Researcher


[email protected]