Data Quality Management for Traffic Monitoring Systems based on Wireless Location Technology
|Presentation on||15. Oct 2009 14:30|
|Presentation room||Seminarraum I5|
|Add presentation to calendar||
Departments of transportation (DOTs) in many countries in the world are trying to manage the roadway network better to improve surface transportation. As a result, there is an increasing demand for traffic condition data across the roadway network. Consequently, DOTs place more emphasizes on better traffic condition monitoring. Traffic condition monitoring systems are widely adopted by DOTs to measure traffic conditions such as vehicle speeds and traffic flow on the transportation network. Traffic monitoring systems have been gradually developed and are classified into two categories: networks of point detectors and probe-based systems. The latter one is further divided into two general types: to sample probe vehicles at fixed locations and to sample vehicles anywhere on the roadway network randomly. Traffic condition data collected by traffic condition monitoring systems are now being used in a wide range of transportation applications, so that transportation professionals are able to manage traffic, provide information to motorists in real-time, develop a database of the historic performance of a road and plan infrastructure.
The most common examples of the second type of probe-based systems are wireless location technology (WLT)-based traffic condition monitoring systems. It has been developed to allow mobile wireless devices (such as cellular phones) to be geolocated. WLT-based monitoring systems anonymously sample the wireless devices equipped in vehicles as probes in the traffic stream and monitor the locations of probe vehicles. By using a series of these probe vehicle positions, true point-to-point speed estimates can be generated on links with a minimal infrastructure investment. It offers the opportunity to directly measure link travel times on a significantly larger portion of the roadway network, which is not provided by currently used point detector systems.
While the idea of WLT-based monitoring is conceptually appealing, over the past decade deployments of the technology have generally not been entirely successful in generating speed or travel time estimates of the quality and quantity that would be needed for traffic management applications. Previous simplified simulation studies and field deployments of early-generation systems have advanced the understanding of how WLT-based traffic monitoring systems are likely to operate, but there still remain a number of unanswered questions about how they should be designed and operated. When DOTs are approached by vendors of WLT-based monitoring systems, they pay much attention to capabilities of WLT-based systems, however, they have little criteria to determine whether the systems are likely to be effective. As a result, further research on the use of WLT for traffic monitoring is needed.
In this work, a simulation based approach is used to treat the concept of WLT-based monitoring in a technology independent manner. A data quality model is developed for the data quality requirements of traffic condition data generated in simulation. The objectives of this work are as follows:
- to explore a measured method, making it possible to show that the type of map matching used has a significant impact on generating accurate and complete vehicle position estimates as well as traffic speed estimates in a WLT-based monitoring system;
- to identify roadway network characteristics and system design parameters that influence the effectiveness of WLT-based monitoring systems and to investigate the Data Quality Management for Traffic Monitoring Systems based on Wireless Location Technology 3 impact of these factors and their interactions on the performance of WLT-based monitoring system (i.e. the number and accuracy of speed estimates generated by the system) using a rigorous experimental design approach;
- to evaluate the efficacy of WLT-based monitoring systems on simulated roadway networks, to present some recommendations for future system design and to provide general guidance that DOTs can use to determine whether a potential WLTbased system is workable.