An Empirical Investigation of Factors Affecting Parking Violation Frequencies of Commercial Vehicles: A Case Study in New York City
Dabiri Zanjani, Saman
MetadataShow full item record
Parking violations by commercial vehicles such as delivery trucks have been a major issue in urban areas for many years. They reduce traffic capacity and cause delays, especially during peak hours when both commercial vehicles and passenger cars compete for parking spaces. In this thesis we evaluated factors associated with parking violations made by commercial vehicles during peak hours in New York City. Count-data-based regression models were developed to quantify the effects of land use, census tract demographic, and road characteristics on parking violation frequencies of commercial vehicles. Given that we only consider parking violations occurred during peak hours and that the number of roadway segments with zero violation frequency is not excessive, both standard negative binomial regression models and zero-inflated regression models were fitted. The former method has been proved to be good at handling particularly non-zero-frequency data while the latter is advanced at screening zero frequencies from non-zero ones. The comparison of the modeling results indicates that the conventional negative binomial regression provides a better fit to the data. Among many factors explored, roadway characteristics (such as roadway functional classes and number of parking signs on a roadway segment) and surrounding environment indicators such as land use and census based demographic attributes have significant impact on the parking violation frequencies of commercial vehicles during peak hours. The modeling methods and the affecting factors identified can be used to predict parking violation occurrences and to revise parking policies in urban areas that suffer from commercial vehicle parking violations.