Exploratory and confirmatory spatial data analysis tools in transport demand modeling

Abstract

The main goal of this paper is to discuss a method for the definition of spatial dependence indicators and their inclusion as variables into transportation demand models. The method is based on ESDA (Exploratory Spatial Data Analyses) and CSDA (Confirmatory Spatial Data Analyses) tools, which have been used in two ways, both in a GIS (Geographic Information Systems) environment: i) to produce indicators of spatial dependence; ii) to evaluate the models estimations. The proposed method is applied in a case study in the city of Porto Alegre, State of Rio Grande do Sul, Brazil, based on origin-destination data obtained through household surveys. The results of this work show that ESDA and CSDA tools are very important for the definition of spatial dependency indicators, identification and selection of the most significant spatial variables, specification and evaluation of demand forecast models and evaluation of the results.