tailieunhanh - Modeling Incident-Related Traffic and Estimating Travel Time with a 2 Cellular Automaton Model

The purpose of this study was to estimate travel time under incident conditions on a freeway. This paper presents a cellular automaton model that mimics traffic conditions and estimates travel times from upstream locations through the incident bottleneck, given an estimate of the incident duration. | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Modeling Incident-Related Traffic and Estimating Travel Time with a Cellular Automaton Model Zhuojin Wang Department of Civil and Environmental Engineering Virginia Tech 7054 Haycock Road Falls Church Va 22043 uSa jeanjeanelle@ and Pamela Murray-Tuite corresponding author Assistant Professor Department of Civil and Environmental Engineering Virginia Tech 7054 Haycock Road Falls Church VA 22043 USA Tel 703 538-3764 Fax 703 538-8540 Email murraytu@ July 2009 Revised November 2009 5324 words 7 figures 1750 2 tables 500 7574 words Submitted for presentation at the 89th Annual Meeting of the Transportation Research Board 1 TRB 2010 Annual Meeting CD-ROM Paper revised from original submittal. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Modeling Incident-Related Traffic and Estimating Travel Time with a Cellular Automaton Model Zhuojin Wang and Pamela Murray-Tuite Abstract Incident related travel time helps drivers decide whether to take an alternate route or delay a trip. The purpose of this study was to estimate travel time under incident conditions on a freeway. This paper presents a cellular automaton model that mimics traffic conditions and estimates travel times from upstream locations through the incident bottleneck given an estimate of the incident duration. This model is among the first applications of cellular automata to incidents and incorporates driving behavior on a freeway with on- and off-ramps shoulder lanes bottlenecks and incidents. Innovative features of the model include driving behavior on shoulder lanes and speed oscillation in ramp influence zones. The model was applied to a 16 mile section of eastbound Interstate 66 in northern Virginia and successfully reproduced daily bottlenecks. Four incidents were tested and based on the selected MAPE and GEH thresholds the model accurately reproduced queue lengths and .