Commercial vehicles time of day choice behavior in urban areas

Abstract

This study presents a mixed Multiple Discrete-Continuous Probit (MDCP) modeling framework to describe time of day choice behavior of commercial vehicles and the vehicle-miles-traveled (VMT) at the chosen time(s) of day. The mixed MDCP model recognizes the unobserved heterogeneity across commercial vehicles in response to customer-to-customer travel speeds, and service time/dwell time at customer stops due to customer preferences for predefined business hours. Possible common unobserved factors affecting different times of day are also captured estimating a general covariance error structure. The estimated models show that a host of factors affect commercial vehicle time of day choice behavior. These include commercial vehicle class, type, commodity type, total unloading weight, as well as location features of base and intermediate stops, frequency of stops to office locations and construction sites; and time of day attributes including travel speed and service time. The results indicate that travel needs, are for the most part, influenced by activity needs. The estimated model can be used to predict the likelihood of commercial vehicle time of day choice for performing daily activities and the VMT generated at the chosen time(s) of day.