Traffic Data Processing Research Articles

FaRNet

Extracting knowledge from big network traffic data is a matter of foremost importance for multiple purposes ranging from trend analysis or network troubleshooting to capacity planning or traffic classification. An extremely useful approach to profile traffic is to extract and display to a network administrator the multi-dimensional hierarchical heavy hitters (HHHs) of a dataset. However, existing schemes for computing HHHs have several limitations: 1) they require significant computational overhead; 2) they do not scale to high dimensional data; and 3) they are not easily extensible. In this paper, we introduce a fundamentally new approach for extracting HHHs based on generalized frequent item-set mining (FIM), which allows to process traffic data much more efficiently and scales to much higher dimensional data than present schemes. Based on generalized FIM, we build and evaluate a traffic profiling system we call FaRNet. Our comparison with AutoFocus, which is the most related tool of similar nature, shows that FaRNet is up to three orders of magnitude faster.

Expectation-maximization based algorithm for pattern recognition in traffic speed distribution

Understanding driving behavior is a complicated research topic. To describe accurate speed, flow and density of a multiclass users traffic flow, an adequate model is needed. Mostly, user-classes are categorized by vehicle type characteristics. However, driving behavior is also influenced by drivers and socio-economic characteristics. Categorizing user-class by vehicle type may not reflect multiclass users traffic flow properly. On the other hand, driving behavior is studied through tracking trace of individual vehicles, experimenting in a driving simulator or inquiring by questionnaire generally. It costs a lot and may produce bias because of the design of the questionnaire or experiment. Therefore, a new method, which is based on a pattern recognition technique, is proposed to classify driving behavior in multiclass users traffic flow. In this study, the speed is considered as the result of driving behavior and the speed distribution on a road is assumed to be a mixture of Gaussian distributions. According to the assumptions, the expectation-maximization algorithm is employed to train and classify different user-classes. With this method, an economical and automatic way for traffic data processing and parameter extraction is obtained.

ITS 사업평가를 위한 효과척도 대안

본 연구의 목적은 ITS 사업의 시행효과를 효율적으로 측정할 수 있는 지표를 선정하고 지표와 관련된 자료의 수집 및 가공방법을 제시함으로써, ITS 사업의 효과를 객관적으로 평가할 수 있는 방법론을 정립하는 것이다. 본 연구는 국내 외 ITS 사업평가 방법론 및 평가지표를 검토하였으며 검토 과정 중 ITS 효과 평가 시 평가의 범위, 표준화된 ITS 효과분석 방법론의 부재, 효과평가 부문별 지표의 중복성의 문제를 파악하였다. 따라서 이러한 문제점을 해결하고자, 평가지표로 도로의 통행상태를 직접적으로 반영할 수 있는 평균통행속도와 통행량을 선정하였다. 선정한 두 지표는 기존의 효과평가 방법론에서 적용한 지표들의 의미를 포함하면서 소통 상태를 잘 반영할 수 있는 것으로 기존의 연구에서 제시한 지표들에 비해 단순하면서 객관적 평가가 가능하다. 통행량의 측정방법은 Cordon line, All Point Average 방법이 있으며, 평균통행속도의 측정방법은 교통류의 특성(연속류/단속류) 및 소통상태(혼잡/비혼잡)에 따라 측정방법을 달리 적용한다. 본 연구에서 제안한 평가방법론은 누구나 쉽게 이해할 수 있고 어느 사업에나 적용이 가능한 평가체계를 수립할 수 있다. 이를 정책적으로 활용할 목적으로 도입한다면 표준화된 ITS 평가체계를 수립할 수 있을 것으로 기대한다. The objective of this study is to build a methodology for evaluating intelligent transportation systems (ITS) projects, by selecting measures of effectiveness (MOEs) and developing an approach to collect and process traffic data. While reviewing the existing MOEs and evaluation methodologies for ITS projects, several problems were found, such as the complication in delineating study areas, the absence of standardized evaluation methodologies, and the duplication in selecting MOEs. To tackle these problems, two MOEs capable of directly evaluating traffic conditions were chosen: i.e., average vehicle speed and traffic volume. Both MOEs can not only include all the functions of the existing MOEs, but also be simpler and more objective in evaluating real traffic conditions. The traffic volume can be measured by using either "cordon line" or "all point average" methods. On the other hand, measuring the average vehicle speed depends on site-specific characteristics such as traffic flow states (interrupted or uninterrupted) and traffic conditions (congested or uncongested). The present methodology is easily understandable for anyone and applicable for any ITS project, and is also expected to contribute to building a standardized evaluation system.

Risk Analysis and Mitigation Strategy for System Design

The traffic management system to date relies on the available infrastructure for the collection, transmission and processing of traffic data. The technology in this field has not kept up with the state of the art due to lack of funding, lack of interest from the private sector and the lack of synergy between the various users of technology. On the other hand great strides have been made in the processing of traffic data to obtain dynamic solutions to relieve congestion, predict traffic flows and thwart threatening traffic jams. Academia and industry have been developing algorithms and models that run on traditional Central Processing Unit (CPU)-based machines to provide solutions.

The regulation of location-based services: challenges to the European Union data protection regime

To further the growth of location-based services (LBS), clear regulation is indispensable. The European Union (EU) legal framework includes a strict regime for the processing of personal data, location data and traffic data. After a short introduction of two core elements of European data protection legislation – the Data Protection Directive and the ePrivacy Directive – a number of technological and societal trends are described, which have made the applicability of the existing EU regulation on personal data processing in general and location data in particular more problematic. Recently, the Art. 29 Working Party released Opinion 13/2011 on geolocation services on smart mobile devices. An analysis of the Opinion reveals that it fails to clarify how the European legal framework should be applied in practice on two accounts. First, a clear separation between different actors and their roles has become unrealistic, and particularly the practical differences between telecom operators and information society service providers are quickly becoming merely theoretical. Second, location data used in LBS are increasingly the result of an amalgamation of other data, which sources cannot be practically traced anymore. Thus, the approach to apply different regimes to different types of data is increasingly less tenable. Next, two possible routes for improvement are suggested. The first opts to extend the current regime that only deals with a limited category of location data to all types of location data, thus strengthening the current ePrivacy Directive. The second chooses to abandon the current special regime for location data completely in favour of a more general protection of personal data, thus bestowing more weight to the Data Protection Directive. These suggestions provide valuable input for the current revision of the European legal framework on data protection.

How do we reduce the risk of deaths and injuries from incidents involving police cars? Understanding injury prevention in the Swedish context

Motor vehicles are an indispensible and daily tool of police work. The risk that police officers face when driving has attracted much academic attention, although in Sweden research in this field has been limited. In this paper, it is our intention to identify risks involving police cars, to review the causes of police car accidents in Sweden, and to give a better understanding of injury prevention in the Swedish police context. We note that, during the last few years, police officers' road traffic training has become more structured, covering emergency service driving and pursuits. Nevertheless, we call for more comprehensive driver education in police officer training, along with the establishment of strict routines for the gathering and processing of traffic incident data.

Bridge Structural Condition Assessment Based on Vibration and Traffic Monitoring

A stochastic model of traffic excitation on bridges is developed assuming that the arrival of vehicles traversing a bridge (modeled as an elastic beam) follows a Poisson process, and that the contact force of a vehicle on the bridge deck can be converted to equivalent dynamic loads at the nodes of the beam elements. The parameters in this model, such as the Poisson arrival rate and the stochastic distribution of vehicle speeds, are obtained by image processing of traffic video data. The model reveals that traffic excitations on bridges are spatially correlated. This important characteristic is usually incorrectly ignored in most output-only methods for the identification of bridge structural properties using traffic-induced vibration measurement data. In this study, the stochastic traffic excitation model with partial traffic information is incorporated in a Bayesian framework, to evaluate the structural properties and update their uncertainty for condition assessment of the bridge superstructure. The vehicle weights are also estimated simultaneously in this procedure. The proposed structural assessment methodology is validated on an instrumented highway bridge.

Processing Traffic Data Collected by Remote Sensing

Video data are being used more often to study traffic operations. However, extracting vehicle trajectories from video by current methods is a difficult process, typically resulting in many errors. The process requires extensive labor to correct the trajectories manually. This paper proposes a method to process video data from traffic operations. Instead of detecting a vehicle in each picture of the video separately, the video data are transformed so that the trajectories of the vehicles (their position over time) become visible in a single image. In this single image, the trajectories can be found by detecting lines. The difference from other methods is that trajectories rather than vehicles are detected. Trajectory (line) detection is more robust than vehicle (rectangle) detection; with this method, about 95% of the trajectories are detected correctly and, more important, the segments of each trajectory are much longer compared with results from other methods in the literature. Also, the detection is a quick process because only a single image is required to be analyzed. For a data set 5 min long, transforming costs several minutes, and automatically detecting and tracking costs 40 to 50 min per lane. Manual correction is then necessary, which costs about 10 min per lane. In contrast, with a different method the total processing time for analyzing traffic operations costs about 1 week for all lanes together.

Traffic Probe Data Processing for Full-Scale Deployment of Vehicle-Infrastructure Integration

Vehicle-infrastructure integration (VII) is expected to provide greatly improved transportation planning and operations data by enabling many–and eventually all–road vehicles to function as traffic data probes. This means that VII probe data can be processed to produce useful information about the operation of an arterial roadway based on studies using a traffic microsimulation. Simulation provides a “truth model” representation of all vehicle trajectories, which serves as the basis for comparison with probe data sampling strategies under development in the VII program. Modifications to the probe data sampling strategies are suggested to enhance the quality of the data, especially its timeliness. Applications to weather condition detection, incident detection, and realtime adaptive traffic signal control, each with significantly different data requirements, were considered, and the ability of the probe sampling system to support these applications was evaluated. The least-demanding applications, such as weather condition detection, can be served well even at low market penetrations of VII-capable vehicles, but the most demanding applications (real-time adaptive signal control) appear to require the majority of vehicles to be equipped. Recommendations were defined for additional research needed to support the development and deployment of a viable VII traffic probe data system.

VGITS: ITS based on intervehicle communication networks and grid technology

The intelligent traffic system based on intervehicle communication networks and grid technology is proposed. This system adopts hybrid architecture, and diverse real-time traffic services are provided in a centralized or decentralized way. Grid technology is introduced to provide the high performance computing platform for massive traffic data processing and real-time traffic service presenting. Moreover, the computation and communication capacity of the IVC network is developed and exploited to provide decentralized traffic services to drivers. Therefore, the system load is balanced among grid nodes and in-vehicle terminal, and better performance is achieved. At last, a prototype is implemented to demonstrate the efficiency and effectivity.

Effective arterial road incident detection: A Bayesian network based algorithm

Timely and accurate incident detection is an essential part of any successful advanced traffic management system. The complex nature of arterial road traffic makes automated incident detection a real challenge. Stable performance and strong transferability remain major issues concerning the existing incident detection algorithms. A new arterial road incident detection algorithm TSC_ar is presented in this paper. In this algorithm, Bayesian networks are used to quantitatively model the causal dependencies between traffic events (e.g. incident) and traffic parameters. Using real time traffic data as evidence, the Bayesian networks update the incident probability at each detection interval through two-way inference. An incident alarm is issued when the estimated incident probability exceeds the predefined decision threshold. The Bayesian networks allow us to subjectively build existing traffic knowledge into their conditional probability tables, which makes the knowledge base for incident detection robust and dynamic. Meanwhile, we incorporate intersection traffic signals into traffic data processing. A total of 40 different types of arterial road incidents are simulated to test the performance of the algorithm. The high detection rate of 88% is obtained while the false alarm rate of the algorithm is maintained as low as 0.62%. Most importantly, it is found that both the detection rate and false alarm rate are not sensitive to the incident decision thresholds. This is the unique feature of the TSC_ar algorithm, which suggests that the Bayesian network approach is advanced in enabling effective arterial road incident detection.

Assessing Issues, Technologies, and Data Needs to Meet Traffic Input Requirements by Mechanistic–Empirical Pavement Design Guide

As part of the implementation initiatives undertaken by the Indiana Department of Transportation Research Division, this paper presents the effort made to identify potential issues arising from traffic data processing and to assess technologies and data needs to meet the requirements of traffic design inputs in the Mechanistic–Empirical Pavement Design Guide. Global Positioning Systems (GPSs) and geographical information system (GIS) technologies were proposed to manage weigh-in-motion (WIM) and automatic vehicle classification site information and manipulate the traffic design input database. Computer programs were developed to process the raw data ASCII files generated from a WIM vendor's software. A platform was developed to combine GPS coordinates, GIS base maps, data processing programs, and the traffic database into an integral unit. Three WIM sites were selected for trial study. It was demonstrated that, with this platform, the WIM sites and database can be accessed visually and more efficiently. In addition, the computer programs can save significant data processing time. Other issues, such as the possible effect of unclassified vehicle count, were identified. On the basis of findings from the implementation initiatives, necessary efforts and future implementation activities are outlined.

Assessing Issues, Technologies, and Data Needs to Meet Traffic Input Requirements by Mechanistic-Empirical Pavement Design Guide: Implementation Initiatives

As part of the implementation initiatives undertaken by the Indiana Department of Transportation Research Division, this paper presents the effort made to identify potential issues arising from traffic data processing and to assess technologies and data needs to meet the requirements of traffic design inputs in the Mechanistic-Empirical Pavement Design Guide. Global Positioning Systems (GPSs) and geographical information system (GIS) technologies were proposed to manage weigh-in-motion (WIM) and automatic vehicle classification site information and manipulate the traffic design input database. Computer programs were developed to process the raw data ASCII files generated from a WIM vendor's software. A platform was developed to combine GPS coordinates, GIS base maps, data processing programs, and the traffic database into an integral unit. Three WIM sites were selected for trial study. It was demonstrated that, with this platform, the WIM sites and database can be accessed visually and more efficiently. I...

Spatial–Temporal Traffic Data Analysis Based on Global Data Management Using MAS

The spatial-temporal traffic data analysis based on global data management is a newly developed and crucial approach to help traffic managers having the global view of urban traffic status in the level of road network, which is very clearly useful in traffic control and route guidance. The multiagent systems are used in traffic data management with full consideration of the characteristics of traffic data and the cooperation and workflow among them. In software implementation of data management, the agent-based common object request broker architecture is adopted taking the distributed urban traffic data in the large area under network environments into account. Based on the global traffic data, the approach of visualized spatial-temporal analysis is then induced. The similarity of traffic data is analyzed first for each link and its profile is achieved to undertake the primary processing of urban traffic data. Furthermore, analysis results are shown on the basis of the geographic information systems for transportation. The two types of visualization, pseudocolor and contour maps, are adopted in the demonstration to display the traffic status graphically and its changing frames. Among the applications in some big cities in China, the case of urban traffic analysis for Beijing is studied to demonstrate the implementation of the approach.

Congested Freeway Microsimulation Model Using VISSIM

A procedure for constructing and calibrating a detailed model of a freeway by using VISSIM is presented and applied to a 15-mi stretch of I-210 West in Pasadena, California. This test site provides several challenges for microscopic modeling: a high-occupancy vehicle (HOV) lane with an intermittent barrier, a heavy freeway connector, 20 metered on-ramps with and without HOV bypass lanes, and three interacting bottlenecks. Field data used as input to the model were compiled from two separate sources: loop detectors on the on-ramps and main line (PeMS) and a manual survey of on-ramps and off-ramps. Gaps in both sources made it necessary to use a composite data set, constructed from several typical days. FREQ was used as an intermediate tool to generate a set of origin-destination matrices from the assembled boundary flows. The model construction procedure consists of (1) identification of important geometric features, (2) collection and processing of traffic data, (3) analysis of the main-line data to identify recurring bottlenecks, (4) VISSIM coding, and (5) calibration based on observations from Step 3. A qualitative set of goals was established for the calibration. These were met with relatively few modifications to VISSIM's driver behavior parameters.

Real-Time Indicators of Vehicle Kilometers of Travel and Congestion: One Year of Experience

Automated traffic data posted on the Internet by four cities have been continuously downloaded, processed, and archived for more than 1 year by an automated system developed by Oak Ridge National Laboratory and funded by the U.S. Bureau of Transportation Statistics. Although the experimental system is far from national in scale and scope, it has shown that automated collection and processing of local traffic data via the Internet for national purposes is feasible and practical. Strong seasonal patterns make it too early to estimate statistical models of traffic growth, but comparisons of the same months in 1998 and 1999 indicate changes ranging from 1 percent to 3 percent for the monitored systems. Direct measurements of delay on the monitored systems are lower than published estimates for previous years. Although some progress in the input of missing data has been made, missing data are still a major problem, and better methods are needed.

Intelligent Traffic Data Processing for ITS Applications

Real-time traffic data are the lifeline sustaining the operation of Advanced Traffic Management and Advanced Traveler Information Systems (ATMS/ATIS). Data are essential to drive algorithms related to congestion/incident detection, travel time forecasting, and real-time route guidance. A common problem in many ATMS/ATIS applications is the sparsity of real-time traffic data, reflecting the financial constraints of acquiring and maintaining large-scale traffic monitoring systems. This paper proposes the use of intelligent processing, or data integration tools, to overcome the data sparsity problem and make the best use of existing data resources. This approach recognizes the elements of uncertainty and vagueness in defining and solving the problem. An example application of the proposed data integration method is presented in the context of a congestion detection algorithm. The method uses an imprecise knowledge representation within the framework of fuzzy operator logic (FOL) and the modified Dempster-Shafer rule of combination. Results indicate that knowledge of a link congestion status (i.e., congested or uncongested) increased several folds after the data integration algorithm was applied. Further work is needed to calibrate the algorithm in the field (in this study simulation was used) and to apply the procedures on large-scale networks.

System engineering for reliability and ease of maintenance

The paper discusses the lessons learned from several years of investigation into the performance of a large air traffic control data processing system. To achieve the twin requirements implicit in the title, the time between faults must be made as long as possible, whilst the time required to clear a fault must be reduced to the minimum. The practice of building up some degree of system redundancy with stand-by equipment is also one that can be employed to ensure that the occurrence of a fault does not cause an appreciable loss of system facilities.The experimental air traffic control system, on which the original findings were based, illustrates the feasibility of designing a high standard of reliability into data processing equipment. Extensive data on the performance of the system have since been analysed and used to improve further the reliability factor.