Travel-Time Estimates Obtained from Intelligent Transportation Systems and Instrumented Test Vehicles: Statistical Comparison

Abstract

Accurate estimation of travel time is necessary for monitoring the performance of the transportation system. Often, travel times are estimated indirectly by using instantaneous speeds from inductance loop detectors and making a number of assumptions. Although these travel times may be acceptable estimates for uncongested conditions, they may have significant error during congested periods. Travel times also may be obtained directly from intelligent transportation systems (ITS) data sources such as automatic vehicle identification (AVI). In addition, mobile cellular telephones have been touted as a means for obtaining this information automatically. Data sources that collect travel-time estimates directly provide travel-time data for both real-time and off-line transportation system monitoring. Instrumented test vehicle runs are often performed to obtain travel-time estimates for system monitoring and other transportation applications. Distance measuring instruments (DMIs) are a common method of instrumentation for test vehicles. DMI travel-time estimates are compared with AVI travel-time estimates by using a variety of statistical approaches. The results indicate that the travel-time estimates from test vehicles instrumented with DMI are within 1% of travel-time estimates from AVI along the study corridor. These results reflect that DMI is an accurate instrumented test vehicle technology and, more important, AVI data sources can replace traditional system monitoring data collection methods when there is adequate tag penetration and infrastructure. A method for identifying instrumented test vehicle drivers who may require additional data collection training is provided. The described procedures are applicable to any instrumented vehicle technique (e.g., the Global Positioning System) in comparison to any ITS data source that directly estimates travel time (e.g., mobile cellular telephones).

Validating Dynamic Message Sign Freeway Travel Time Messages with Ground Truth Geospatial Data

“…Difficulties can arise when communications are disrupted or when individual detectors or stations periodically malfunction, report inaccurate data, or are placed in a way that masks actual traffic conditions on key links or provides limited coverage of the freeway system. Various algorithms have been proposed or used to estimate travel times by using inductive loops (5)(6)(7)(8)(9)(10)(11). When inductive loop data are used for incident detection, for example, it is important to have control over errors to avoid false alarms.…”

Statistical Methods for the Evaluation and Monitoring of Traveler Information System Data Quality

“…Travel time can be directly measured by observing the time it takes for a vehicle to traverse a link or it can be inferred from measurements of related variables such as spot-speed, volume, and density. [18] The following sections briefly discuss the difference between inference and direct measurement of travel time.…”