Travel Behavior Modeling: Exploring Household Characteristics as Predictors

Effective transportation infrastructure planning requires road agencies to develop a sequential Travel Demand Model (TDM) process, beginning with Trip Generation. This involves using the total number of person-trips generated in specific areas as the dependent variable, with Household (HH) and socioeconomic factors as independent variables influencing travel behavior. In India, accurately collecting household information and trip characteristics is challenging due to dynamic travel patterns influenced by socioeconomic and urban development factors, making secondary data unreliable over time. Trip rates, indicating the average number of trips per unit of a specific factor, are essential for understanding travel demand trends. This study examines travel patterns in Surat, Gujarat, India, using data from one thousand households collected through home interviews. It investigates the relationship between Per Capita Trip Rates (PCTR) and household variables such as size, vehicle ownership, income, students, and earners, based on previous studies and mobility plans for Indian cities. The hypothesis testing showed no significant difference between the sample mean and population mean at a 10% significance level, with Surat’s mean trip rate estimated at 1.57. Specific trip rates for work, education, shopping, and other activities were 0.82, 0.46, 0.13, and 0.16, respectively. Analysis revealed a negative correlation between household size and trip rates, and a positive correlation between earning members, income groups, and vehicle ownership. Regression analysis using Linear Regression (LR) and Multiple Linear Regression (MLR) highlighted the unpredictability of infrequent trips like shopping and recreation, while regular work and educational trips showed significant relationships with household variables. The study provides valuable insights into Surat’s travel behavior, aiding future trip production estimations as the urban population grows.