Transportation Research Part D-transport and Environment最新文献

Combating carbon dioxide (CO₂) emissions across sectors becomes inevitable due to negative impacts. The transport sector takes place among the most important sectors. Accordingly, the study examines transport-related CO₂ (TCO₂) emissions in the top four emitting countries (namely, the United States, Canada, Saudi Arabia, & Australia) by considering six explanatory variables, using data from 1990/Q1 to 2020/Q4, and performing an artificial intelligence approach. The outcomes show fresh insights that (i) super learner (SL) algorithm overwhelms other machine-learning algorithms in terms of model performance; (ii) energy intensity has an increasing impact on TCO₂ emissions, whereas others (e.g., financial development, income, globalization, oil use, & urbanization) have a mixed impact across countries; (iii) the influential variables have some critical thresholds, where the power of impacts differentiate across these limits. Hence, the SL algorithm presents robust outcomes for TCO₂ emissions. Accordingly, a set of policy endeavors for the countries examined are also discussed.

There is growing interest in understanding the interaction between weather and transportation and the ability of communities and the nation’s infrastructure to withstand extreme conditions and events. This study aims to provide detailed insights on how people adjust and change their activity-travel and time use behaviors in the face of extreme heat conditions. By leveraging time use records integrated with weather data, the study compares activity-mobility patterns between extreme heat days and non-extreme days. A series of models are estimated to understand the impact of extreme heat even after controlling for other variables. The findings reveal that heat significantly impacts time use and activity-mobility patterns, with some groups exhibiting potentially greater vulnerability arising from the inability to adapt sufficiently to extreme heat. Designing dense, shaded urban environments, declaring heat days to facilitate indoor stays, and providing transportation vouchers for vulnerable populations can help mitigate the ill-effects of extreme heat.

This study is motivated by the need to understand what the most appropriate policies are to promote battery electric vehicles (BEVs) in various countries around the world. A systematic review of reviews on policies promoting BEVs was conducted to synthesise knowledge on the relative effectiveness of various policies. This study addresses three limitations in existing research on policies to promote electric vehicles (EVs). Firstly, it disentangles findings for BEVs from that of other EVs. Secondly, it examines the relative effectiveness of these policies to find optimal policy mixes. Finally, it compares policy effectiveness across nations at various stages of economic development and EV adoption. Purchase subsidies and tax incentives were found to be highly effective policies to kickstart EV adoption but may not be as effective at later stages of the EV adoption and may not be an affordable policy for all countries. Some countries that led the way with such subsidies and tax incentives have now begun curtailing or ceasing them, and contrary to earlier reviews, without serious adverse effect on adoption rates. Policies supporting public charging infrastructure are crucial enablers that complement EV purchase subsidies but are important even without purchase subsidies in place. High-occupancy vehicle (HOV) lane access and toll waivers are the next most effective demand side policies, but only in cities where such incentives make sense and cents for consumers. Supply side policies, such as zero-emission vehicle (ZEV) mandates and vehicle emissions standards, must match or precede demand side policies to avoid bottlenecks and effectively drive uptake of BEVs in the earliest stage of adoption.

Zero-emission powertrains, connectivity, and automation are the future of automotive mobility, though their collective impacts on fuel economy is difficult to study. This paper develops a novel methodology to simulate the impacts of cooperative driving automation on battery electric (BEVs) and fuel-cell electric (FCEVs) vehicles. Enabling V2I connectivity for city driving resulted in fuel economy improvement of 6 % to 13 % for BEVs and 9 % to 15 % for FCEVs. Enabling aerodynamic drag reduction in V2V highway driving resulted in fuel economy improvement of 5 % to 32 % for BEVs and 5 % to 26 % for FCEVs. Sensitivity analysis on battery and fuel cell efficiency was conducted to determine how technological improvements could impact connected mobility. Improving powertrain component efficiencies decreased performance gains for V2I city driving while increasing performance gains for V2V highway driving. Fuel-cell efficiency improvements had greater impacts on connectivity gains than battery efficiency improvements. Vehicle testing should verify these results.

Asphalt releases volatile organic compounds (VOCs) during paving processes, posing risks to workers and the environment. The complex composition of asphalt and the evolving of VOCs present challenges in accurately assessing their potential environmental and health impacts using traditional experimental approaches. This study aimed to develop a robust computational framework integrating machine learning and network pharmacology to predict the risks from the asphalt VOCs. The results show that the MACCS+XGBoost model achieved the highest predictive performance, with an accuracy of 0.85, balanced accuracy of 0.84, sensitivity of 0.83, specificity of 0.84, and F1-score of 0.84 in the external validation. The network pharmacology analysis revealed that the identified VOCs with reproductive toxicity potential may disrupt key processes such as spermatogenesis, ovarian function, and hormonal regulation, providing mechanistic insights into their potential impacts. This advancement supports a proactive approach to environmental protection and fosters the transition towards a more sustainable, low-carbon transportation.

This research uses three machine learning algorithms to predict transport-related CO₂ emissions, considering transport-related factors and socioeconomic aspects. We analyze the top 30 countries that produce the highest transport-related global CO₂ emissions, split evenly between Tier 1 and 2. Tier 1 comprises the five leading nations that produce 61% of the world’s CO₂ emissions, while Tier 2 comprises the subsequent twenty-five nations that produce 35% of the global CO₂ emissions. We assess the efficacy of our model by using four statistical measures (R2, MAE, rRMSE, and MAPE) in a four-fold cross-validation procedure. The Gradient-Boosted Regression (GBR) machine learning model, which incorporates a combination of economic and transportation factors, outperforms the other two machine learning approaches (Support Vector Machine and Ordinary Less Square). Our findings indicate that among Tier 1 and Tier 2 countries, socioeconomic factors like population and GDP are more influential on the models than transportation-related factors.

The lack of a globally recognized measurement technique combined with a limited ability to comprehend the actual level of GHG emissions in intricate logistics operations causes significant obstacles for firms in assessing the magnitude of their environmental footprint. Nevertheless, linking, upkeeping, and managing gas detectors on mobile vehicles under varying road and weather circumstances present an expensive solution for predicting GHG emissions. This article presents the development and evaluation of a reliable and accurate real-time technique for capturing GHG emissions using the Internet of Things (IoT) and Artificial Intelligence (AI). The findings indicate that the integration of gradient-boosting models (LightGBM, xGBoost, and gradient-boosting decision trees) via ensemble learning enhances the precision of CO2 emission predictions. The weighted ensemble method attains an RMSE of 1.8625, surpassing the performance of individual models. Visualizations validated a robust correlation between anticipated and actual CO2 concentrations, illustrating the model’s precision and negligible prediction errors.