Showing papers by "Xiaolei Ma published in 2010"

Developing a GPS-Based Truck Freight Performance Measure Platform

Abstract: Although trucks move the largest volume and value of goods in urban areas, relatively little is known about their travel patterns and how the roadway network performs for trucks. Global positioning systems (GPS) used by trucking companies to manage their equipment and staff and meet shippers’ needs capture truck data that are now available to the public sector for analysis. The Washington State Department of Transportation (WSDOT), Transportation Northwest (TransNow) at the University of Washington (UW), and the Washington Trucking Associations (WTA) have partnered on a research effort to collect and analyze GPS truck data from commercial, in-vehicle, truck fleet management systems used in the central Puget Sound region. The research project is collecting commercially available GPS data and evaluating their feasibility to support a state truck freight network performance monitoring program. WSDOT is interested in using this program to monitor truck travel times and system reliability, and to guide freight investment decisions. This report discusses the steps taken to build, clean, and test the data collection and analytic foundation from which the UW and WSDOT will extract network-based truck performance statistics. One of the most important steps of the project has been to obtain fleet management GPS data from the trucking industry. Trucking companies approached by WSDOT and the UW at the beginning of the study readily agreed to share their GPS data, but a lack of technical support from the firms made data collection difficult. The researchers overcame that obstacle by successfully negotiating contracts with GPS and telecom vendors to obtain GPS truck reads in the study region. The next challenge was to gather and format the large quantities of data (millions of points) from different vendors’ systems so that they could be manipulated and evaluated by the project team. Handling the large quantity of data meant that data processing steps had to be automated, which required the development and validation of rule-based logic that could be used to develop algorithms.

Using Truck Fleet Management GPS Data to Develop the Foundation for a Performance Measures Program

Abstract: Global positioning systems (GPS) used for fleet management by trucking companies provide probe data that can support a truck performance-monitoring program. This paper discusses the steps taken to acquire fleet management data and then process those data so they can eventually be used for a network-based truck performance measures program. While other studies have evaluated truck travel by using GPS, they have used a limited number of project-specific and temporary devices that have collected frequent location reads, permitting a fine-grained performance analysis of specific roadway segments. In contrast, this fleet management GPS data project involved infrequent reads but a relatively large number of different trucks with ongoing data collection. The most effective approach to obtaining the fleet management data was to purchase the data directly from GPS vendors. Because a performance measures program ultimately monitors trips generated by trucks as they travel between origins and destinations, an algorithm was developed to extract trip end information from the data. The large volume of data required automated processing without manual intervention. Because performance measures require travel times and speeds, it was also necessary to evaluate whether speed data from a large number of trucks could compensate for infrequent location reads. Spot speeds recorded by the trucks’ GPS devices were compared to speed data from roadway loops. The researchers concluded that spot speed data can indicate free flow conditions, but sufficient quantities of data are probably necessary to measure congested travel.

Real-Time Travel Time Prediction on Urban Traffic Network

Abstract: Travel time is one of the most desired operational and system Measures of Effectiveness (MOEs) for evaluating the performance of freeways and urban arterials. With accurate travel time information, decision makers, road users, and traffic engineers can make informed decisions. However, retrieving network-level travel time information has several challenges, such as travel time estimation, prediction, and data processing. This research addresses these challenges by developing innovative methodologies and computer applications. First, the authors developed a two-step empirical approach to effectively estimating link journey speeds using merely advance loop detector outputs. Second, an α–β filter is adopted to dynamically predict and smooth real-time loop measured spot speeds. In addition to travel time estimation and prediction, a time dependent shortest path algorithm is also developed, to determine the shortest travel time route based on real-time traffic. Lastly, the developed algorithms are implemented in a web-based Real-time Analysis and Decision-making for ARterial Network (RADAR Net) system. In order to achieve real-time performance, sensor and signal control databases are carefully designed to ensure fast query over a huge amount of network-level traffic data. Furthermore, the data visualization and statistical analysis modules are also added to RADAR Net to facilitate user applications. Currently, the RADAR Net system is part of the Digital Roadway Interactive Visualization and Evaluation Network (DRIVE Net) (www.uwdrive.net), developed by the STAR Lab of the University of Washington. RADAR Net is capable of performing all required tasks efficiently in real-time

A Pitfall to Avoid When Issuing Transit Signal Priority Treatments under Coordinated Control Strategies

Abstract: Transit Signal Priority (TSP) is an advanced control mechanism to facilitate transit vehicle operations along signalized arterials. In practical application, TSP systems are integrated with coordinated signal control strategies. People take for granted the reduced transit delays through TSP treatments. However, due to conflicts between TSP control schemes and coordinated signal control strategies along signalized arterials, the benefits achieved by transit vehicles may be washed out to some extent. Our recent study for optimizing the South Snohomish Regional Transit Signal Priority (SS-RTSP) system operations found that transit vehicle delays over a signal-coordinated corridor can be lengthened by TSP treatments in some scenarios. A VISSIM-based simulation model is developed to emulate TSP system operations along the SR-99 arterial covering 13 intersections, in the City of Lynnwood, Washington. Various transit operation scenarios under diverse coordinated signal control plans are designed. Theoretical analysis is also provided to formulate the transit travel process. The research findings indicated that to achieve the best operational efficiency, the compatibility between TSP control schemes and signal control coordination should be strengthened to minimize transit disruption to signal coordination.