Time

1:40 p.m. – 3:00 p.m.

Location

1605 Tilia, Room 1103, West Village

Speaker(s)

Naveen Eluru, Associate Professor, Department of Civil, Environmental and Construction Engineering, University of Central Florida

Abstract

In this presentation we introduce an innovative econometric framework that allows us to better analyze discrete outcome processes. The framework employs the latent segmentation model approach to examine alternative causal structures in multi-dimensional decision process. In the first part of our presentation we describe the development of the framework for analyzing commuter train users’ access mode and station choice. Typically, access mode and station choice for commuter train users is modeled as a hierarchical choice with mode being considered as the first choice in the sequence. In our approach we simultaneously consider two segments of station and access mode choice behavior: Segment 1 – station first and mode second and Segment 2 – mode first and station second. The allocation to the two segments is achieved through a latent segmentation approach that determines the probability of assigning the individual to either of these segments as a function of socio-demographic variables, level of service (LOS) parameters, trip characteristics, land-use and built environment factors, and station characteristics. The approach offers many advantages compared to the traditional alternatives. First, we gain a better understanding of the decision processes by examining who are the individuals who choose the station (or mode) first. Second, the approach proposed is free from simulation and easy to implement. Third, the results from our analysis will provide insights to transit agencies on how to improve transit service to reduce the automobile travel to commuter train stations. The proposed latent segment model is estimated using data from an on-board survey conducted by the Agence Métropolitaine de Transport (AMT) for commuter train users in Montreal region. The model is employed to investigate the role of socio-demographic variables, LOS parameters, trip characteristics, land-use and built environment factors, and station characteristics on commuter train user behavior.

In the second part of the presentation we apply a variant of the latent segmentation approach for analysing short term vehicle utilization process. Specifically, we explore the causal relationship between vehicle type and usage choices in the short term. The study employs data form the National Household Travel Survey (NHTS) collected in 2009 for New York, Washington DC, and Los Angeles. The three cities represent the spectrum ranging from transit friendly cities through auto-dependent cities. The two segments of vehicle type and distance include: Segment 1 – vehicle type first and distance second and Segment 2 – distance first and vehicle type second. The latent segmentation model is adapted to account for heterogeneity in the choices across regions by considering scale parameters in the vehicle utilization choices. The model estimation results were behaviorally plausible and provide interesting insights for policy analysis. Specifically, the study found differences in the choices across the regions and significant scale parameters in the vehicle type choice and distance models.

Biographical Sketch

Prof. Naveen Eluru is an Associate Professor in the Department of Civil, Environmental and Construction Engineering at the University of Central Florida. He is primarily involved in the formulation and development of discrete choice models that allow us to better understand the behavioral patterns involved in various decision processes. He is actively involved in the development of integrated modeling frameworks for travel demand modeling and vehicular emissions for urban metropolitan regions. He has published journal articles in wide ranging topics including transportation planning, land-use modeling, integrated demand supply models, activity time-use analysis and transportation safety. Prof. Eluru is currently a member of Transportation Research Board (TRB) committee on Statistical Methods (ABJ80). He is a member of the Editorial Advisory Board of Analytic Methods in Accident Research and Sustainable Cities and Society journals.