An estimation of heavy-duty vehicle fleet CO2 emissions based on sampled data

[Display omitted] •Three sampling approaches were investigated for estimating HDV fleet CO2 emissions.•Mean CO2 emissions deviate less than 2% in the best cases and less than 5% overall.•The JDE sampling preserves well the distribution characteristics.•MEDP and MSD are suitable to investigate fleet...

Full description

Saved in:
Bibliographic Details
Published inTransportation research. Part D, Transport and environment Vol. 94; p. 102784
Main Authors Zacharof, Nikiforos, Fontaras, Georgios, Ciuffo, Biagio, Tansini, Alessandro, Prado-Rujas, Iker
Format Journal Article
LanguageEnglish
Published Elsevier Ltd 01.05.2021
Subjects
CO2 emissions
Fleet
Heavy-duty
Simulation
Truck
VECTO
Heavy-duty
Truck
Fleet
VECTO
Simulation
CO2 emissions
Online AccessGet full text

Cover

More Information
Summary:[Display omitted] •Three sampling approaches were investigated for estimating HDV fleet CO2 emissions.•Mean CO2 emissions deviate less than 2% in the best cases and less than 5% overall.•The JDE sampling preserves well the distribution characteristics.•MEDP and MSD are suitable to investigate fleet emissions when limited data are available.•The alternative methodologies can be used to predict CO2 emissions in future fleet scenarios. Certification and monitoring of heavy vehicle CO2 emissions in several countries are based on individual vehicle simulation. Smaller fleet subsets can be used for accurate fleet-level results while preserving the characteristics of the underlying fleet-emissions distributions. The paper focuses on three approaches to capture fleet CO2 emissions: a) sampling directly from the fleet-data, b) sampling from data of individual vehicle components and c) using key statistics regarding the fleet composition that are available. The first and second approach deliver marginal divergences of the mean, between 1.1 and 2.1% and below 2.7 respectively, preserving the characteristics of the distribution. The third deviated by up to 5%, but lacked the detailed characteristics of the underlying statistical distribution. All three are useful when setting up fleet-wide monitoring schemes where detailed data are not available and to investigate the potential CO2 savings of various future fleet compositions, and scenarios regarding the diffusion of different types of technologies.
ISSN:1361-9209
1879-2340
DOI:10.1016/j.trd.2021.102784