Data-driven models for microscopic vehicle emissions
DOI: 10.1016/j.trd.2019.09.013
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the need for accurate microscopic vehicle emission models to support traffic management policies that account for air pollution. Current models often rely on parametric assumptions or complex classifications that may not fully capture the nonlinear relationships between vehicle movement and tailpipe emissions. The authors propose a data-driven approach using Generalized Additive Models for Location, Scale and Shape (GAMLSS) to estimate second-by-second emissions of CO2, CO, and NOx. This method utilizes spline functions to model the intricate effects of instantaneous speed and acceleration without preconceived parametric forms, aiming to improve estimation accuracy, particularly for pollutants affected by catalyst efficiency like CO and NOx. The study utilizes second-by-second emission data collected from laboratory tests at Millbrook Laboratory. Two vehicles—a compact petrol car (Vehicle A) and a supermini diesel car (Vehicle B)—were tested on a chassis dynamometer simulating real driving cycles recorded in urban, suburban, and motorway areas of London. These cycles covered various traffic conditions, including AM peak, inter-peak, and free flow. The GAMLSS framework was implemented using the GJMR R package. The model building process involved selecting explanatory variables (speed, acceleration, and their product) via backward selection based on the Bayesian Information Criterion (BIC). The authors tested twelve potential distributions for the error structure, ultimately selecting Fisk and Dagum distributions based on BIC values and quantile-quantile plots. The model estimates the location, scale, and shape parameters of these distributions using smooth spline functions of the explanatory variables. The results demonstrate that the GAMLSS model provides a substantially better goodness of fit compared to traditional parametric models, such as classified log-polynomial regression and generalized linear models. The improvement is most notable for CO emissions in petrol vehicles and NOx emissions in diesel vehicles. The analysis of smooth functions reveals distinct emission patterns for each vehicle type and pollutant. For instance, CO2 emissions in the petrol vehicle showed a constant positive slope with speed above 60 km/h, while the diesel vehicle exhibited fluctuating trends across different speed ranges. Interaction effects between speed and acceleration were significant, particularly during harsh deceleration at medium speeds for CO2 and CO, and at low speeds for NOx. Cross-validation and residual analysis confirmed the model's robustness and ability to accurately predict emissions across the tested driving cycles. The significance of this work lies in its demonstration that data-driven GAMLSS models can outperform existing emission estimation frameworks by capturing complex, nonlinear relationships without requiring complex engine-out calculations or rigid classification schemes. By providing more accurate microscopic emission estimates, this approach enables better evaluation of traffic management policies regarding their impact on air quality. The findings suggest that incorporating such flexible modeling techniques can lead to more effective strategies for reducing vehicular pollution, especially for regulated pollutants like CO and NOx.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | OpenAlex-citations | — | — | 1 | 2026-06-20 |
| archive | success | unpaywall | — | — | 2 | 2026-06-26 |
| extract | success | pdftotext | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-26 |
| chunk | success | chunk | — | — | 1 | 2026-06-26 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-26 |
| enrich | success | semantic_scholar | — | — | 4 | 2026-06-26 |
| promote | success | — | — | — | 1 | 2026-06-20 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-26 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Theoretical Contribution: computational model