Root Cause Analysis of Traffic Anomalies Using Uneven Diffusion Model

Abstract

Detection and analysis of traffic anomalies are important for the development of intelligent transportation systems. In particular, the root causes of traffic anomalies in road networks as well as their propagation and influence to the surrounding areas are highly meaningful. The root cause analysis of traffic anomalies aims to identify those road segments, where the traffic anomalies are detected by the traffic statuses significantly deviating from the usual condition and are originated due to incidents occurring in those roads such as traffic accidents or social events. The existing methods for traffic anomaly root cause analysis detect all traffic anomalies first and then apply, implicitly or explicitly, specified causal propagation rules to infer the root cause. However, these methods require reliable detection techniques to accurately identify all traffic anomalies and extensive domain knowledge of city traffic to specify plausible causal propagation rules in road networks. In contrast, this paper proposes an innovative and integrated root cause analysis method. The proposed method is featured by 1) defining a visible outlier index as the probabilistic indicator of traffic anomalies/disturbances and 2) automatically learning spatiotemporal causal relationship from historical data to build an uneven diffusion model for root cause analysis. The accuracy and effectiveness of the proposed method have been demonstrated by experiments conducted on a trajectory dataset with 2.5 billion location records of 27 266 taxies in Shenzhen city.INDEX TERMS Root cause analysis, traffic anomalies, spatiotemporal causal relationship, visible outlier index, uneven diffusion model.

Intelligent Graph Convolutional Neural Network for Road Crack Detection

“…This is done with the help of CNN so that anomalies can be identified. Huang et al [15] studied the factors that contribute to traffic volume in an extensive metropolitan network. The obvious characteristics of an outlier are evaluated to determine whether or not they can serve as a signal of abnormal traffic behavior.…”

Cognitive Traffic Anomaly Prediction from GPS Trajectories Using Visible Outlier Indexes and Meshed Spatiotemporal Neighborhoods

“…In [27], an uneven diffusion model is proposed using a deep learning architecture (i.e., Stacked AutoEncoder (SAE) plus Back Propagation (BP)) to learn the relationship of traffic disturbances between the target region and the surrounding parts in road networks. Then, the historical traffic data at the previous time bin t − 1 is used to predict the traffic at t. Our approach is to use both spatial and temporal dependency for the traffic prediction of abnormal levels (scored by VOIs).…”

“…Among the known mathematical function, the logarithmic function is just suitable for this need. According to the analysis and comparison in our previous work [27], we find the best is logarithm based on 10, based on which we define a new metric, VOI, in Definition 1. As shown in Table 1, VOI can be more intuitively understood by people compared with P value.…”