article · Scientific Reports
Accurately estimating carbon dioxide emissions from road vehicles is vital for environmental regulators and automotive engineers designing cleaner powertrains. A newly developed machine learning framework combines a long short-term memory neural network with nature-inspired optimisation techniques: specifically, binary firefly feature selection alongside a hybrid firefly and ant colony algorithm for hyperparameter tuning. Tested on an open government registry of over 22,000 vehicles, this optimised model achieved an R-squared accuracy of 98.71 percent in predicting tailpipe emissions, outperforming alternative architectures including convolutional networks and multilayer perceptrons. The selected features highlight that certified fuel consumption metrics strongly drive predictive power, although general vehicle characteristics still provide meaningful signal when fuel data is unavailable. Because the analysis relied on cross-sectional data rather than time series, the network functions primarily as a tool for identifying complex interactions among vehicle specifications.
Precise estimation of greenhouse gas emissions helps authorities enforce environmental regulations and allows automotive developers to benchmark powertrain performance without relying solely on costly physical testing. Demonstrating that vehicle descriptors can yield highly accurate emission estimates through optimised algorithmic pipelines provides a reliable, data-driven foundation for assessing vehicle compliance and supporting policy decisions.
The abstract positions this framework for regulatory planning and cleaner powertrain development. Relevant users include environmental transport agencies assessing fleet compliance and automotive engineers benchmarking early-stage powertrain concepts. The work represents applied computational research tested against retrospective public registry data. Moving towards commercial or operational use would require integrating the pipeline into established regulatory software or proprietary vehicle engineering design suites.
AI-generated from the published abstract. Always read the original work before citing.
Accurate vehicle-level CO₂ emission prediction is essential for regulatory planning and greener powertrain design. This study presents an integrated FA-ACO-LSTM framework for estimating vehicle-level tailpipe CO₂ emissions from standard vehicle descriptors. The contribution lies in combining leakage-aware Binary Firefly feature selection, mixed-variable FA-ACO hyperparameter tuning, repeated-run evaluation, and post hoc interpretability into a unified prediction pipeline. Using a public fuel-consumption dataset sourced from the official Canadian vehicle fuel-consumption and emissions registry (open government portal) (22,556 vehicles), we adopt a 70/20/10 train/validation/test split and perform wrapper-based feature selection with Binary Firefly Algorithm (BFA). The FA-ACO scheduler jointly tunes hyperparameters for five candidate learners (LSTM, CNN, GRU, MLP, TabNet) and selects the best. FA-ACO-LSTM attains MSE = 0.0099, MAE = 0.0791, MedAE = 0.0664, MAPE = 0.84%, and R² = 98.71% on the test set, surpassing FA-ACO-CNN (R² = 93.55%), FA-ACO-GRU (90.03%), FA-ACO-MLP (87.60%), and FA-ACO-TabNet (84.88%). Compared with untuned baselines, FA-ACO improves LSTM from R² = 96.53% to 98.71% and reduces MSE by 45%. BFA yields the lowest average error among feature selectors versus BMWO, BFO, and BGWO, and an ANOVA across models confirms significance (F = 30.2, p < 0.0001). These results represent a fuel-consumption-informed prediction scenario because the selected feature set includes COMB (L/100 km), HWY (L/100 km), and COMB (mpg), which are strongly associated with certified CO₂ emissions. When all fuel-consumption indicators were excluded before feature selection and training, model performance decreased from [Formula: see text]to [Formula: see text]. The ablation confirms that fuel-consumption proxies account for a substantial part of the main predictive performance, while non-consumption vehicle attributes retain meaningful predictive value, and indicates that, under the adopted fixed feature-ordering representation and FA-ACO tuning protocol, LSTM achieved the best performance among the evaluated neural and tabular baselines. However, because the dataset is cross-sectional rather than temporal, the LSTM should be interpreted as a feature-interaction learner rather than a temporal forecasting model.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1038/s41598-026-68293-0
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
