Using the Multilevel Random Effect Model to Analyze the Behavior of Carpool Users in Different Cities

Tzu-Ying Chen,Yi-Chang Chiu,Rong-Chang Jou

doi:10.3390/su13020937

Tzu-Ying Chen, Yi-Chang Chiu + Show 1 more

Open Access

https://doi.org/10.3390/su13020937

Copy DOI

Abstract

The purpose of this study was to offer a comprehensive econometrical framework based on a multilevel random effect logistic model that could highlight important contributors to carpool users among different cities with various attributes. The data was collected from the three cities of Tucson, AZ, USA; El Paso, TX, USA; and Austin, TX, USA and was based on register-based travel trip data from the Metropia platform and American Community Survey information from 2016 to 2017. The empirical results indicated there were statistically significant differences among carpool users in different cities due to the transportation mode, number of vehicles available, total number of males driving alone, and number of single-parent households. The individual level result showed that incentives had a significant effect on the promotion of carpool passenger and driver behavior. In addition, the time of finding the parking space at work, living situation of the household, flexibility to change departure times, gender, and age could effectively increase the possibility of carpool usage. The results of this study give a better understanding of the events in the initial factors of carpooling behavior and can be used by the government or commercial company to design an effective solution for traffic congestion.

Highlights

Car ownership has rapidly increased in recent years; road capacity is restricted due to the limitations of road space
Ln (The total worker population of government workers’ principal mode to get from home to work is public transportation in each city) ln (The total worker population that the number of available vehicles at home and household members’ use is less than one vehicle in each city) ln (The total worker population of the commuter worker is male, and the primary travel type of working is driving alone in each city) ln (The total worker population that single-parent household that has children between the age of three and 17 who are enrolled in school in each city) lnTotalTrip ln (The number of total trips per month)
Level 2 included natural logarithmic values such as “the total worker population of government workers’ principal mode to get from home to work is public transportation in each city”, “the total worker population in which the number of available vehicles at home and household members’ use is less than one vehicle in each city”, and “the total worker population that are single-parent households that have children between the age of three and 17 who are enrolled in school in each city”, we assume a positive effect on the carpool user

Summary

Introduction

Car ownership has rapidly increased in recent years; road capacity is restricted due to the limitations of road space. One of the solutions for reducing congestion is to decrease the number of vehicle miles traveled (VMT). Public transit provides a service with fixed routes and schedules, making it hard to migrate travelers from private vehicles to public transit. As an alternative, carpooling has become a popular approach for traffic congestion management, as it directly reduces the number of cars participating in the transportation system. Less ownership of vehicles could become important for the environmental challenges associated with car use. A previous study indicated that the greater the number of carsharing users, the fewer vehicles owned per household [1]. Carsharing and carpooling are believed to be “a potential new pathway to sustainability” [2]

Objectives

Methods

Results

Conclusion