An Application of Exploratory Multivariate Data Analysis Techniques in a Peer Study of Land Use Influence on Individual Destination Choices
Cira Souza Pitombo, Bruno Vieira Bertocini, Eiji Kawamoto · 2008
This study is an initial attempt to improve the understanding of relationships between land use patterns and travel behavior. By using Exploratory Multivariate Data Analysis (EMDA), this work investigates the factors that influence the destination choices (Travel distances sequences) during the trip-chaining of the residents in the São Paulo Metropolitan Area (SPMA). It is an exploratory analysis that aims to answer some research questions. In this work, two techniques of EMDA had been used: (1) Cluster Analysis, to group and to characterize the Traffic Zones (TZ) of the SPMA and, (2) Decision Tree, to find a priori unknown relationships between socioeconomic characteristics, land use attributes and destination choices. The results indicate that the degree of activity of origin TZ influences the travel patterns.