Outlier Mining Based Traffic Incident Detection Using Big Data Analytics

1 Early detection of incidents is one of the key step to reduce incident-related congestion. With the 2 increasing usage of GPS based navigation, promising data-scalable crowdsourced probe data is 3 now available which can provide near-real time traffic speed information. This study utilizes such 4 extensive historical datasets (approximately 500 GB) to gain useful insights on the normal traffic 5 pattern of each segment. The insights come in the form of speed threshold for different time of the 6 day and days of week for each segment. Thereafter, the anomalous traffic behaviour are classified 7 as incidents. The dynamic thresholds developed for each segment simplifies the calibration steps 8 that is often required when applying a model to a different dataset. Also, in this study, two alter9 natives of the traditional Standard Normal Deviate (SND) based incident detection algorithm are 10 tested. The proposed algorithms can handle the masking effect of SND method where the outliers 11 inflate the mean and standard deviation values and result in lower threshold values and in turn, 12 lower detection rate. The high detection rate (94-97%) obtained by these algorithms compared to 13 the SND method (83%) shows the efficacy of the models. Although higher false alarm rate (FAR) 14 are observed for these models, but their values (4 false alarms/day) are quite lower than the accept15 able FAR (10 false alarms/day) reported in previous literature (1). 16 17

[1]  Weifeng Lv,et al.  Outlier mining based Automatic Incident Detection on urban arterial road , 2009, Mobility Conference.

[2]  Bruce Hellinga,et al.  Automatic Vehicle Identification Technology-Based Freeway Incident Detection , 2000 .

[3]  William H. K. Lam,et al.  New Automatic Incident Detection Algorithm Based on Traffic Data Collected for Journey Time Estimation , 2013 .

[4]  Neal Hawkins,et al.  Reliability of Probe Speed Data for Detecting Congestion Trends , 2015, 2015 IEEE 18th International Conference on Intelligent Transportation Systems.

[5]  Yixuan Li,et al.  Motorway incident detection using probe vehicles , 2005 .

[6]  Conrad L Dudek,et al.  USING PROBE-MEASURED TRAVEL TIMES TO DETECT MAJOR FREEWAY INCIDENTS IN HOUSTON, TEXAS , 1996 .

[7]  Dimitrios Tzovaras,et al.  Mining traffic data for road incidents detection , 2014, 17th International IEEE Conference on Intelligent Transportation Systems (ITSC).

[8]  Lee D. Han,et al.  Toward Training-Free Automatic Detection of Freeway Incidents , 2012 .

[9]  D. Schrank,et al.  2015 Urban Mobility Scorecard , 2015 .

[10]  Darcy M. Bullock,et al.  Shock Wave Boundary Identification Using Cloud-Based Probe Data , 2015 .

[11]  Ronald K. Pearson,et al.  Mining imperfect data - dealing with contamination and incomplete records , 2005 .

[12]  Emily Parkany,et al.  A COMPLETE REVIEW OF INCIDENT DETECTION ALGORITHMS & THEIR DEPLOYMENT: WHAT WORKS AND WHAT DOESN'T , 2005 .

[13]  Joseph L. Schofer,et al.  Arterial incident detection using fixed detector and probe vehicle data , 1995 .

[14]  Frank S. Koppelman,et al.  USE OF VEHICLE POSITIONING DATA FOR ARTERIAL INCIDENT DETECTION , 1996 .

[15]  D. Schrank,et al.  THE 2004 URBAN MOBILITY REPORT , 2002 .

[16]  Billy M. Williams,et al.  Traffic Management Center Use of Incident Detection Algorithms: Findings of a Nationwide Survey , 2007, IEEE Transactions on Intelligent Transportation Systems.