Non-recurrent traffic congestion detection on heterogeneous urban road networks

Anbaroğlu, Berk; Cheng, Tao; Heydecker, Benjamin

doi:10.1080/23249935.2015.1087229

“…The significance level is determined to be 0.05, so that whenever the p-value of an STR is less than 0.05 it would be considered to be significant. On the other hand, the only parameter of Percentile based NRC detection, π, is varied between 75 and 95, as this range would correspond to the unusually high LJTs (Anbaroğlu et al, 2015).…”

Section: Resultsmentioning

confidence: 99%

“…The likelihood ratio function of an STR s whose individual LJTs are lognormally distributed is stated in equation (2) (Anbaroğlu et al, 2015).…”

Section: Stss Based Nrc Detectionmentioning

confidence: 99%

See 1 more Smart Citation

How Travel Demand Affects Detection of Non-Recurrent Traffic Congestion on Urban Road Networks

Anbaroglu¹,

Heydecker²,

Cheng³

2016

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Occurrence of non-recurrent traffic congestion hinders the economic activity of a city, as travellers could miss appointments or be late for work or important meetings. Similarly, for shippers, unexpected delays may disrupt just-in-time delivery and manufacturing processes, which could lose them payment. Consequently, research on non-recurrent congestion detection on urban road networks has recently gained attention. By analysing large amounts of traffic data collected on a daily basis, traffic operation centres can improve their methods to detect non-recurrent congestion rapidly and then revise their existing plans to mitigate its effects. Space-time clusters of high link journey time estimates correspond to non-recurrent congestion events. Existing research, however, has not considered the effect of travel demand on the effectiveness of non-recurrent congestion detection methods. Therefore, this paper investigates how travel demand affects detection of non-recurrent traffic congestion detection on urban road networks. Travel demand has been classified into three categories as low, normal and high. The experiments are carried out on London's urban road network, and the results demonstrate the necessity to adjust the relative importance of the component evaluation criteria depending on the travel demand level.

show abstract

“…making the distribution bimodal) on the distribution of travel times (Susilawati et al, 2011). Furthermore, only few studies mention the data cleaning procedure, which might have a substantial impact on the distribution of LJTs (Anbaroğlu et al, 2015).…”

Section: Literature Reviewmentioning

confidence: 99%

“…There is a growing research interest to detect NRCs on an urban road networks; yet, the variation of travel demand on the effectiveness of such approaches has not been investigated so far (Anbaroğlu et al, 2015). Therefore, this paper aims to investigate how different travel demand levels affect the performance of NRC detection methods.…”

Section: Introductionmentioning

confidence: 99%

How Travel Demand Affects Detection of Non-Recurrent Traffic Congestion on Urban Road Networks

Anbaroğlu

¹

,

Heydecker

²

,

Cheng

³

2016

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

0

View full text Add to dashboard Cite

ABSTRACT:Occurrence of non-recurrent traffic congestion hinders the economic activity of a city, as travellers could miss appointments or be late for work or important meetings. Similarly, for shippers, unexpected delays may disrupt just-in-time delivery and manufacturing processes, which could lose them payment. Consequently, research on non-recurrent congestion detection on urban road networks has recently gained attention. By analysing large amounts of traffic data collected on a daily basis, traffic operation centres can improve their methods to detect non-recurrent congestion rapidly and then revise their existing plans to mitigate its effects. Space-time clusters of high link journey time estimates correspond to non-recurrent congestion events. Existing research, however, has not considered the effect of travel demand on the effectiveness of non-recurrent congestion detection methods. Therefore, this paper investigates how travel demand affects detection of non-recurrent traffic congestion detection on urban road networks. Travel demand has been classified into three categories as low, normal and high. The experiments are carried out on London's urban road network, and the results demonstrate the necessity to adjust the relative importance of the component evaluation criteria depending on the travel demand level.

show abstract

Spatially‐correlated time series clustering using location‐dependent Dirichlet process mixture model

Jung,

Kim,

Kim

2023

Statistical Analysis

0

View full text Add to dashboard Cite

The Dirichlet process mixture (DPM) model has been widely used as a Bayesian nonparametric model for clustering. However, the exchangeability assumption of the Dirichlet process is not valid for clustering spatially correlated time series as these data are indexed spatially and temporally. While analyzing spatially correlated time series, correlations between observations at proximal times and locations must be appropriately considered. In this study, we propose a location‐dependent DPM model by extending the traditional DPM model for clustering spatially correlated time series. We model the temporal pattern as an infinite mixture of Gaussian processes while considering spatial dependency using a location‐dependent Dirichlet process prior over mixture components. This encourages the assignment of observations from proximal locations to the same cluster. By contrast, because mixture atoms for modeling temporal patterns are shared across space, observations with similar temporal patterns can be still grouped together even if they are located far apart. The proposed model also allows the number of clusters to be automatically determined in the clustering procedure. We validate the proposed model using simulated examples. Moreover, in a real case study, we cluster adjacent roads based on their traffic speed patterns that have changed as a result of a traffic accident occurred in Seoul, South Korea.

show abstract

Non-recurrent traffic congestion detection on heterogeneous urban road networks

Cited by 42 publications

References 33 publications

How Travel Demand Affects Detection of Non-Recurrent Traffic Congestion on Urban Road Networks

How Travel Demand Affects Detection of Non-Recurrent Traffic Congestion on Urban Road Networks

How Travel Demand Affects Detection of Non-Recurrent Traffic Congestion on Urban Road Networks

Spatially‐correlated time series clustering using location‐dependent Dirichlet process mixture model

Contact Info

Product

Resources

About