The framework of eco-driving evaluation relying on vehicle trajectory data is to quantify the disparities of the fuel consumption for individual driving behavior and to develop a baseline under various traffic conditions. The baseline represents the typical driving behavior in a city, and it is a pivotal parameter for eco-driving evaluation. The applicability of the evaluation method in different cities is overlooked, encompassing the suitability of parameters and the minimum data required. This study aims to investigate whether the evaluation baseline developed with sufficient data can be applied to a new city. The results reveal that the baseline developed in Beijing cannot be directly transferred to the eco-driving evaluation in Toronto due to the significantly more aggressive and competitive driving behavior exhibited by Toronto drivers. This study further examines the minimum data sample size necessary to develop a robust evaluation baseline and proposes a localized method to construct the evaluation system for eco-driving evaluation in different cities.

Accurate heavy-duty vehicle load estimation is crucial for transportation and environmental regulation, yet current methods lack precision in data accuracy and practicality for field implementation. We propose a Self-Supervised Reconstruction Heterogeneous Graph Convolutional Network (SSR-HGCN) for load estimation using On-Board Diagnostics (OBD) data. The method integrates physics-constrained heterogeneous graph construction based on vehicle speed, acceleration, and engine parameters, leveraging graph neural networks’ information propagation mechanisms and self-supervised learning’s adaptability to low-quality data. The method comprises three modules: (1) a physics-constrained heterogeneous graph structure that, guided by the symmetry (invariance) of physical laws, introduces a structural asymmetry by treating kinematic and dynamic features as distinct node types to enhance model interpretability; (2) a self-supervised reconstruction module that learns robust representations from noisy OBD streams without extensive labeling, improving adaptability to data quality variations; and (3) a multi-layer feature extraction architecture combining graph convolutional networks (GCNs) and graph attention networks (GATs) for hierarchical feature aggregation. On a test set of 800 heavy-duty vehicle trips, SSR-HGCN demonstrated superior performance over key baseline models. Compared with the classical time-series model LSTM, it achieved average improvements of 20.76% in RMSE and 41.23% in MAPE. It also outperformed the standard graph model GraphSAGE, reducing RMSE by 21.98% and MAPE by 7.15%, ultimately achieving < 15% error for over 90% of test samples. This method provides an effective technical solution for heavy-duty vehicle load monitoring, with immediate applications in fleet supervision, overloading detection, and regulatory enforcement for environmental compliance.