Developing Machine learning models for hyperlocal traffic related particulate matter concentration mapping. (December 2022)

• Record Type: Journal Article
• Authors: Desai, Salil; Tayarani, Mohammad; Oliver Gao, H.
• Is Part Of: Transportation research. Volume 113(2022)
• Publication Date: 2022-12
• Subjects: Machine Learning -- Traffic -- Air Quality -- Convolutional Neural Network -- Convolutional Long Short-Term Memory; Transportation -- Research -- Periodicals; Transportation -- Environmental aspects -- Periodicals; 354.76
• DOI: 10.1016/j.trd.2022.103505
• Languages: English
• ISSNs: 1361-9209

Abstract

Abstract: Recently, estimating air pollution concentrations and contributions from various sources has become a major research focus. Our research adds to the body of knowledge by developing machine learning (ML) models to avoid intermediate modeling steps using traffic and meteorological data. The ML model's overall performance in predicting air pollution concentrations at receptors outperforms previous methods. Our best model, Convolutional Long Short-Term Memory (ConvLSTM), has a Mean Relative Error (MRE) of 38.9%, which is lower than the 47.5% MRE for the single hidden layer model, the 63.2% MRE for the Convolutional Neural Network model, and the 41.5% MRE for ConvLSTM with time-series data. Memory cells help the ConvLSTM model predict a large number of spatially correlated observations. The novel ML modeling approach has low data requirements and is computationally efficient, which makes it promising for future transportation planning, epidemiology, and environmental justice assessments.