Citation

Marquis R, Wang XC. Transp. Res. Rec. 2015; 2517: 10-17.

Copyright

(Copyright © 2015, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

10.3141/2517-02

PMID

unavailable

Abstract

Truck-related crash occurrences in Manhattan, New York City, are analyzed during four time blocks: morning peak (AMP, 6:00 to 10:00 a.m.), midday (MD, 10:00 a.m. to 3:00 p.m.), afternoon peak (PMP, 3:00 to 7:00 p.m.), and nighttime (NT, 7:00 p.m. to 6:00 a.m.). The results of zero-inflated negative binomial models indicate that both the built environment and traffic flows contribute to temporal variation in truck-related crash occurrences. More specifically, tracts with larger populations and higher employment in the finance, insurance, and health care sectors tend to have fewer crashes at night. In contrast, larger household sizes and the retail, professional services, education, and accommodation industry sectors are associated with increased NT crash occurrences. In addition, if 1,000 trucks were shifted from AMP to the NT, the average tract would experience a net increase of truck crashes of 0.2160; the net increase would be 0.1948 if the trucks were shifted from the PMP to the NT. Shifting trucks from the MD to the NT reduces the count by 0.0267; this result suggests that this strategy might provide the best safety benefits. When possible induced nontruck demand is accounted for, even the largest impact on safety (during the PMP) increased crashes by only 3.56%. These findings fill the void of studies that focused on the influence of temporal effects on truck crash occurrences in congested urban settings.