Development of a Methodology for Measuring Traffic Congestions Using Discrete Time Markov Chain

This paper reports on a study that developed a method for estimating the probabilities of traffic congestion occurring and its subsequent impacts to upstream and downstream segments using Discrete Time Markov Chains (DTMC). The authors note that it is difficult to predict both when traffic congestion will occur and how far the impacts of congestion will propagate. They define breakdown as a transition state between non-congested and congested conditions and use that state to signify the beginning of congestion. DTMC is one of the stochastic processes that can express the changes of traffic state at every given time unit over the time and space. The study included speed and volume data that were collected by Remote Traffic Microwave Sensor (RTMS) at each detection location for a 4-month period, from May to August 2004 along US-202 southbound, Philadelphia, Pennsylvania, an urban freeway which has a 55 mph (about 88km/h) speed limit. Breakdown is defined to occur when the speed drops below 50 mph for at least 5 minutes, and it is assumed that the congestion ends when the speed goes up to 50 mph again. The authors conclude that the model developed successfully provides amount of traffic congestion for a given facility. Traffic Management Centers (TMCs) can easily use the model developed and calibrate it to their needs to obtain amount of congestion for various time periods.

Language

  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; References; Tables;
  • Pagination: 12p
  • Monograph Title: ITS Connections: Saving Time. Saving Lives

Subject/Index Terms

Filing Info

  • Accession Number: 01140656
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Sep 16 2009 3:50PM