A Modified Hidden Semi-Markov Model for Traffic Related PM10 Pollution Levels Estimation

Traffic related PM10 (particulate matter with less than 10 microns) pollution exposure leads to different types of diseases e.g., lung function changes, heart rate variability, immune cell responses and asthma attacks. New investigations confirmed a massive increase in particulate matter of central congested urban areas in Tehran metropolitan which exceeds the standards of Environmental Protection Agency (EPA, 150 µg/m3) and World Health Organization (WHO, 20 µg/m3 annual mean and 50 µg/m3 24-hour mean). Long term continuous real-time monitoring of air quality at these areas is essential but is not possible due to financial and operational constraints. Hence, using an alternative tool is important to ensure compliance with the standards and also provides a choice for commuters to reduce their unnecessary trips in contaminated areas across the city. In this study, a stochastic framework is developed based on a Hidden Semi-Markov model (HSMM) to predict the state of PM10 particulates. Our proposed HSMM model predicts PM10 concentration for the next day, based on PM10 levels in previous days. The result of simulation shows the proposed technique achieves good accuracy in estimation of PM10. It also indicates that the model can be used for one-day ahead forecast to alert individuals in the study area which is particularly useful in situations where the information on external variables such as traffic volume is not available.

Language

  • English

Media Info

  • Media Type: Web
  • Pagination: 1 PDF file, 410 KB, 22p.
  • Monograph Title: Canadian Transportation Research Forum 50th Annual Conference - Another 50 Years: Where to From Here?//Un autre 50 ans : qu'en est-il à partir de maintenant? Montreal, Quebec, May 24-26, 2015

Subject/Index Terms

Filing Info

  • Accession Number: 01605022
  • Record Type: Publication
  • Source Agency: Transportation Association of Canada (TAC)
  • Files: ITRD, TAC
  • Created Date: Jul 26 2016 5:04PM