Applying machine learning methods in managing urban concentrations of traffic-related particulate matter (PM10 and PM2.5)

Abstract This study presents a new method for evaluating the effectiveness of roadside PM10 and PM2.5 reduction scenarios using Machine Learning (ML) based models. The ML methods include Artificial Neural Networks (ANN), Boosted Regression Trees (BRT) and Support Vector Machines (SVM). Traffic, meteorological and pollutant data collected at nineteen Air Quality Monitoring (AQM) sites in London for a period between 2007 and 2012 was used. The ML models performed very well in predicting the concentrations of PM10 and PM2.5 with around 95% of their predictions falling within the factor of two of the observed concentrations at the roadsides. The prediction errors observed were very small as indicated by the average normalised mean gross errors of 0.2. Also, the predictions of the models correlated well with the observed concentrations as shown by the average values of R (0.8) and index of agreement (0.74). Additionally, when some PM10 and PM2.5 reduction scenarios were modelled, the ML models predicted various degree of reductions in the roadside concentrations. In conclusion, well trained ANN and BRT models can be successfully applied in predictions of roadside PM10 and PM2.5 concentrations. Moreover, they can be applied in measuring the effectiveness of roadside particle reduction scenarios.

[1]  Pedro G. Lind,et al.  Air quality prediction using optimal neural networks with stochastic variables , 2013, 1307.3134.

[2]  V. Prybutok,et al.  A neural network model forecasting for prediction of daily maximum ozone concentration in an industrialized urban area. , 1996, Environmental pollution.

[3]  Ruwim Berkowicz,et al.  OSPM - A Parameterised Street Pollution Model , 2000 .

[4]  Yu Xue,et al.  Prediction of particulate matter at street level using artificial neural networks coupling with chaotic particle swarm optimization algorithm , 2014 .

[5]  C. A. McHugh,et al.  ADMS-Urban: an air quality management system for traffic, domestic and industrial pollution , 2014 .

[6]  N. Pérez,et al.  Neural network model for the prediction of PM10 daily concentrations in two sites in the Western Mediterranean. , 2013, The Science of the total environment.

[7]  Aranildo R. Lima,et al.  Nonlinear regression in environmental sciences by support vector machines combined with evolutionary strategy , 2013, Comput. Geosci..

[8]  Kim N. Dirks,et al.  Development of an ANN-based air pollution forecasting system with explicit knowledge through sensitivity analysis , 2014 .

[9]  Hui Li,et al.  Evolutionary artificial neural networks: a review , 2011, Artificial Intelligence Review.

[10]  Tiziano Tirabassi,et al.  Air dispersion model and neural network: A new perspective for integrated models in the simulation of complex situations , 2006, Environ. Model. Softw..

[11]  Alison S. Tomlin,et al.  A field study of factors influencing the concentrations of a traffic-related pollutant in the vicinity of a complex urban junction , 2009 .

[12]  Shikha Gupta,et al.  Identifying pollution sources and predicting urban air quality using ensemble learning methods , 2013 .

[13]  Jérôme Chave,et al.  Scale and Scaling in Ecological and Economic Systems , 2003 .

[14]  Lili Liu,et al.  Research of neural network algorithm based on factor analysis and cluster analysis , 2011, Neural Computing and Applications.

[15]  Vivien Mallet,et al.  Ensemble forecasting with machine learning algorithms for ozone, nitrogen dioxide and PM10 on the Prev'Air platform , 2014 .

[16]  Yunqian Ma,et al.  Practical selection of SVM parameters and noise estimation for SVM regression , 2004, Neural Networks.

[17]  Gordon J. Esplin,et al.  Approximate explicit solution to the general line source problem , 1995 .

[18]  Max Kuhn,et al.  Building Predictive Models in R Using the caret Package , 2008 .

[19]  M. Ragosta,et al.  Input strategy analysis for an air quality data modelling procedure at a local scale based on neural network , 2015, Environmental Monitoring and Assessment.

[20]  Greet Janssens-Maenhout,et al.  Near-term projection of anthropogenic emission trends using neural networks , 2014 .

[21]  Andrew Quinn,et al.  Assessment and prediction of the impact of road transport on ambient concentrations of particulate matter PM10 , 2016 .

[22]  Miles Tight,et al.  Hybrid Neural Networks and Boosted Regression Tree Models for Predicting Roadside Particulate Matter , 2016, Environmental Modeling & Assessment.

[23]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[24]  Stephen Dorling,et al.  Statistical surface ozone models: an improved methodology to account for non-linear behaviour , 2000 .

[25]  Stef van Buuren,et al.  MICE: Multivariate Imputation by Chained Equations in R , 2011 .

[26]  Fatih Taşpınar,et al.  Improving artificial neural network model predictions of daily average PM10 concentrations by applying principle component analysis and implementing seasonal models , 2015, Journal of the Air & Waste Management Association.

[27]  Stef van Buuren,et al.  Multiple imputation of discrete and continuous data by fully conditional specification , 2007 .

[28]  Aminu Suleiman,et al.  Multivariate study of vehicle exhaust particles using machine learning and statistical techniques , 2016 .

[29]  S. Samarasinghe,et al.  Complex time series analysis of PM10 and PM2.5 for a coastal site using artificial neural network modelling and k-means clustering , 2014 .

[30]  Ernő Mészáros.  Atmospheric Chemistry , 1981 .

[31]  Chunming Wu,et al.  Application of support vector regression to predict metallogenic favourability degree , 2010 .

[32]  M. W Gardner,et al.  Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences , 1998 .

[33]  D. R. Middleton,et al.  Field Study of Wind and Traffic to Test a Street Canyon Pollution Model , 2000 .