August 7 | 2017
Spatial Transferability of a Microresidential Mobility Model in the Integrated Land Use, Transportation, and Environment Modeling System
Fatmi, M.R. and Habib, M.A. (2015)
Transportation Research Record: Journal of the Transportation Research Board No. 2496, pp. 29-36
https://doi.org/10.3141/2494-04
Abstract
This paper presents the spatial transferability analysis of a microbehavioral model from the residential mobility component of the integrated land use, transportation, and environment (ILUTE) modeling system developed and implemented in the Greater Toronto and Hamilton Area, Ontario, Canada. The study examined whether ILUTE could be spatially transferred with the current model components to a different geographic area: Halifax, Nova Scotia, Canada. The residential mobility component within ILUTE is a continuous-time, hazard-based duration model, developed with retrospective survey data from the Residential Mobility Survey 2 in the Greater Toronto and Hamilton Area. This study developed a similar continuous-time, hazard-based duration model for the residential mobility decisions of households in Halifax on the basis of retrospective survey data from a household mobility and travel survey. The model results suggested that households in Halifax and the Greater Toronto and Hamilton Area exhibited profound differences in residential mobility decisions. Sociodemographic, dwelling, and neighborhood characteristics significantly affected residential mobility decisions in the Greater Toronto and Hamilton Area. The effects of land use and accessibility measures were noteworthy for Halifax. For instance, home-to-work distances in Halifax affected the decision to move; however, such an effect could not be confirmed in the Greater Toronto and Hamilton Area. Households’ first periods of residence after household formation in a residential location were shorter in Halifax than in the Greater Toronto and Hamilton Area. It was concluded that the direct transfer of micromodels from one spatial context to another could be difficult.