Modeling Vehicle Fuel Consumption and CO2 Emissions: Assessing Alternative Methods, Lag Effects, and Internal–External Factors

Given the challenges associated with the transferability of specific emission modeling tools between different regions, developing accurate local emission models utilizing field measurements has become increasingly relevant for effectively reflecting local conditions. In this study, we employed a comprehensive benchmarking approach, drawing on an extensive set of on-road experiments encompassing various vehicle types. More specifically, this study aims to (1) conduct a thorough review of alternative modeling techniques used for modeling second-by-second fuel consumption and emission measures across different vehicle categories and (2) assess and compare the performance of identified modeling methods, employing either internal (OBD) or external (GPS) variables, and evaluate the impact of lag effects. Moreover, (3) we make available the collected data, preprocessing codes, and an implementation example as open-source resources for the research community to facilitate reproducibility. The outcomes of this research are expected to offer guidelines for both practical modeling applications and for future work.