Analysis and prediction of household location choice in

advertisement
Topic area F1: Integrated Land Use and Transport Planning/Implementation
Topic area F2: Urban and Regional Modelling
Simulating residential location choice in the Lyon Urban Area
with transportation-land use modelling framework UrbanSim
Marko Kryvobokov, Dominique Bouf, Nicolas Ovtracht, Pierre-Yves Peguy
Laboratory of Transport Economics (LET), University of Lyon, CNRS – Lyon, France
Abstract
The increasing popularity of transportation-land use modelling frameworks is explained by their
capacity to analyse and simulate urban development with a reasonably high level of representation
of a complex reality of an urban area with interactions between land use, transportation, population
and employment. The possibility to predict future development trends make these tools the
important instruments for decision-makers and researchers in transportation and urban planning.
In the current study, a transportation-land use modelling framework UrbanSim is applied. It
was developed at the University of Washington (Waddell et al., 2003) as an open-source project. Its
popularity among researchers and practitioners in different countries of the world is increasing (e.g.
de Palma et al., 2005). UrbanSim includes a system of models, which are used for simulation of
distribution of population, jobs, real estate development, real estate prices and other parameters in
future years on a yearly basis. Our focus is in the group of UrbanSim models dealing with
household location choice. We simulate future geographical distribution of households with the
household location choice model itself using data from the external transportation model MOSART
(Modelling and Simulation of Accessibility to Networks and Territories) and updating the following
necessary attributes across the area via the corresponding models: future residential development,
residential land share, and real estate prices. The methodology applied in UrbanSim includes the
multinomial logit and the Ordinary Least Squares regression. We analyse the Lyon Urban Area,
which is the second biggest in France by population, with 1.7 million inhabitants in 2005. The aim
of the paper is to evaluate various measures of transport policy and to predict changes in population
distribution.
An important feature of UrbanSim is a possibility to apply different scenarios of urban
development. Thus, current transportation network can be incorporated as well as new links can be
added in order to observe the influence of this development on distribution of population, in
particular on further increasing of urban sprawl. One more important feature to simulate is a
dynamics of residential development projects. Describing a social component of household location
choice with variables for income level and car-ownership helps to explain some patterns of
segregation. The predictability of simulation of residential location choice is evaluated using actual
data for past years.
As a tool for urban simulation for the Lyon Urban Area, this UrbanSim application can be
used in a decision-making process providing illustrations of possible consequences of different
scenarios. The study is a part of the on-going project PLAINSUDD (Innovative Numerical
Platforms of Urban Simulation for Sustainable Development) sponsored through French ANR.
Keywords: transportation-land use modelling, UrbanSim, household location choice, the Lyon Urban Area.
References
Waddell, P., Borning, A., Noth, M., Freier, N., Becke, M., and Ulfarsson, G. (2003). Microsimulation of
Urban Development and Location Choices: Design and Implementation of UrbanSim. Networks and Spatial
Economics, 3 (1): 43-67.
de Palma A., Motamedi, K., Picard, N., and Waddell, P. (2005). A model of residential location choice with
endogenous housing prices and traffic for Paris region. European Transport, 31: 67-82.
Download