4.0 Information Sharing Specifications and Data Exchange Formats
FHWA published an Interim Guidance on Information Sharing Specifications and Data Exchange Formats for the Real-Time System Management Information Program in October 2007 [Federal Highway Administration, Publication of Interim Guidance on the Information Sharing Specifications and Data Exchange Formats for the Real-Time System Management Information Program, in Federal Register. 2007: Washington, D.C. p. 58347 - 58379.]. The real-time information program recognized under Section 1201 of SAFTEA-LU was intended to institute a standard data format for the exchange of travel- and traffic-related data between State and local government agencies and the traveling public.
While ITS standards have been developed over the past decades, non-standardized interfaces and different versions of the standards have been deployed in various ITS applications such as traffic management, transit management, and emergency management systems causing difficulties in traffic data exchange among agencies. In addition, traffic information which is available to use for transportation operators and the traveling public is not always accessible due to the lack of standard interfaces. Thus the standardization of data exchange formats is key to establishing the Real-Time System Management Information Program.
There are a variety of traveler information systems in operation today. A system may cover a single metropolitan area, an entire state, or an even larger area such as a multi-state corridor. The types of information and the modes covered can also vary widely. A system might use data from a single transportation entity, a metropolitan transit operator, multiple agencies and/or private entities.
Data for use in a traveler information system are often collected for other purposes. However, these data have a valuable second use when they are processed and packaged in forms that can be used to influence travelers' trip-making decisions. Currently, both public agencies and private organizations are providing information to travelers in many ways. In addition, technological advances are expanding travelers' options on how information can be obtained: telephone, Internet, radio, TV, variable message signs, PDAs, and more. The following sections identify the data quality specifications for a sample real-time traveler information program.
4.1 Sample Applications for Data Quality Measures
Six primary interfaces and their associated applications were defined in the Interim Guidance on Information Sharing Specifications and Data Exchange Formats. These high-level specifications were mapped from the ATIS01-Broadcast Travel Information Market Package in the National ITS Architecture. Table 6 summarizes the six primary interfaces and the associated applications. These applications are described in more detail in the following sections.
Table 6. The Six Primary Interfaces and Associated Applications
Primary Interface |
Application Areas |
Sample ITS Applications |
|---|---|---|
Traffic Management Information |
• Road network conditions |
511, VMS, traffic.com traffic report, weather warning system, VMS |
Maintenance and Construction Management |
• Maintenance and construction work plans |
511, traffic.com roadwork report, VMS |
Transit Management Information |
• Emergency transit schedule information |
511, Web site Information from Transit Operators |
Information Service Provider Information |
• Broadcast information |
511 |
Parking Information |
• Parking locations |
Advance parking management system, ITS truck parking service |
Emergency Management Information |
• Evacuation information |
511, VMS, HAR |
4.2 ITS Applications using Real-Time Information
511
511 is a nationally available traveler information service that provides pre-trip and en-route traveler information. 511 programs provide information through cell phone, an internet web site, TV and radio programs, traveler information kiosks, in-vehicle devices, radios, and/or other wireless devices at local, metro, and statewide levels. Typical 511 services involve current travel information such as traffic delay/congestion, travel time, weather, roadwork, incident, transit, and/or event information allowing travelers to make better choices - choice of time, choice of mode of transportation, and choice of route of travel.
The sample data needs include:
- Delay/congestion information: travel time and speed
- Travel time information: travel time and speed
- Weather information: air temperature, visibility, and precipitation
- Roadwork/construction information: lane closure, duration, and location
- Incident Information: lane closure and estimated clearance time
- Transit Information: service disruption and service adherence
Traffic.com
Traffic.com is an independent private provider of traffic information and services in major U.S. cities. Traffic.com provides real-time traffic conditions, travel time, incident, construction, event, and mass transit information via a number of media, including the Internet, cell phones, radio, satellite radio, and television. The company utilizes four types of traffic data sources: digital traffic sensors, GPS/probe devices, commercial and government partners, and traffic operations center staff members including their own network of sensors to disseminate real-time traveler information. In addition, Traffic.com uses a variety of means such as police and fire scanners and monitoring traffic cameras to collect information and combines these to provide travelers with traffic information.
Data Needs:
- Delay/congestion information: travel time and speed
- Travel time information: travel time and speed
- Weather information: air temperature, visibility, and precipitation
- Roadwork/construction information: lane closure, duration, and location
- Incident Information: lane closure and estimated clearance time
- Transit Information: service disruption and service adherence
Real-Time Weather Information System
More accurate and accessible weather information could improve road maintenance and decrease fatal crashes. The real-time weather and road information is collected from Doppler weather radar, the national lightening detection network, a road sensor network, the agricultural weather network, satellite data, and reports from state highway personnel and state patrol officers. Wind speed and direction, cloud thickness, precipitation, air temperature, dew point and humidity, and radar depiction are/or provided for weather information and overall road conditions, pavement temperature, pavement condition (dry, wet, icy) , road dew point, road freeze point, and/or road snow depth are provided. In addition to these parameters, some agencies provide high resolution road images and/or video images to help the user to inform the driving public of real-time roadway conditions. Many RWIS systems are built up with various types of a map-based system that allows the user to select weather and pavement condition parameters for a specific area within an entire coverage area. The real-time information is typically updated every a 30 to 60 minutes.
Data Needs:
- Weather information: wind speed and direction, cloud thickness, precipitation type and intensity, air temperature, dew point and humidity, and radar depiction
- Road information: overall roadway condition, visibility or visible distance, pavement temperature, pavement condition (dry, wet, icy) , road dew point, road freeze point, and/or road snow depth
Advanced Parking Management Systems
Real-time parking information enhances mobility by avoiding parking problems and traffic congestion. The search for parking often keeps vehicles on the road needlessly and may cause lengthy queues that block adjacent streets. Slowing and stopped vehicles in travel lanes may create safety hazards. Advanced parking systems can save total travel time, improve safety, shift demand to other modes/destinations, and reduce traveler frustration and anxiety.
There are three major areas where parking information is generally in high demand – downtown areas, airports/terminals, and park-and-ride lots. Real-time en-route parking information systems provide information in the form of real-time parking availability, parking locations, and shuttle/transit service. While parking information can be provided through the Internet before a person embarks on a trip, a more common method of dissemination is through dynamic message signs that allow travelers to make a decision en-route. Entry/exit counters and space occupancy detectors are the most frequently utilized method to check the real-time parking availability [Federal Highway Administration, Advanced Parking Management Systems: A Cross-Cutting Study. 2007: Washington, D.C.].
Data Needs:
- Parking lot availability: entry/exit counters and space occupancy detectors
- Parking lot location: parking lot availability, parking lot location information, and transit/shuttle service
Incident Management Systems
Incident management systems can reduce the effects of incident-related congestion by decreasing the time to detect incidents, the time for responding vehicles to arrive, and the time required for traffic to return to normal conditions. A variety of surveillance and detection technologies including inductive loop or acoustic roadway detectors and camera systems providing frequent still images/ full-motion video can help detect incidents quickly. Real-time Information dissemination systems help travelers safely navigate around incidents on the roadway. While most 511 systems include a real-time incident information system, most traffic management centers also share real-time incident information with travelers through technologies deployed as part of incident management programs, such as dynamic message signs, highway advisory radio as well as internet web sites.
Data Needs:
- Incident Information: lane closure, estimated clearance time, and detour information
4.3 Data Quality Measures
Real-time travel information applications require various traffic-related parameters. For instance, traffic incident applications involve the location of incidents, estimated clearance time, speed, travel time, and other parameters. Also transit service information includes routes, schedules, schedule adherence, and fare information as well as transit service information during emergency evacuation. Each application requires a unique set of traffic-related parameters, different levels of data flows, and database management. This section investigates the data quality measures of three of the most widely utilized traffic-related parameters, travel time, speed, and weather information.
4.3.1 Travel Time
The travel time data collection handbook (1998) defines travel time as “the time required to traverse a route between any two points of interest”. Accurately estimating travel time or Travel Time Estimation (TTE) is a critical component of a traveler information system. Traveler information users may alter their route of travel, change their mode, or cancel their trip based on the information provided to them. However, there is no perfectly accurate ATIS travel time estimate that computes the duration of any trip option considering different departure times. Furthermore, it is not possible to estimate perfectly accurate travel times given that some form of prediction is required [Toppen, A. and K. Wunderlich, Travel Time Data Collection for Measurement of Advanced Traveler Information Systems Accuracy. 2003, Prepared for Federal Highway Administration: Washington D.C.].
Travel time is typically estimated using various mathematical models using traffic data collected from a variety of technologies. TTE is a cross-subject that needs various advanced technologies in computer-aided engineering, electronic engineering, automatic control engineering, and telecommunication engineering, all of which are applied in data collection, information transmission, and signal processing. The major technologies currently being used in TTE include loop detectors, probe vehicle technologies, license plate matching, test vehicle technologies, GPS, Automated Vehicle Location (AVL) using transponders or toll tags, cell phones as probes, and aerial surveys.
Accuracy
The accuracy of travel time depends on the data collection and analysis methods. Although each data collection method/technology has some specific advantages, each also has some particular shortcomings, causing the device or technology to work improperly under certain circumstances. For example, loop detectors have trouble measuring low-speed vehicles and only provide point time-mean speeds to estimate link travel times. Probe vehicle technology cannot provide around-the-clock traffic data collection. License plate matching is time consuming and needs large handling efforts that still cannot avoid errors. Test vehicle technology requires massive labor efforts, with slow data processing.
Furthermore, although every travel-time estimation method has its own advantage, none of the methods can provide consistently satisfying outcomes on different common cases. Currently, there is no unanimously satisfying methodology that can estimate travel time with certain accuracy that could be used in traveler information and transportation management applications. Therefore a method’s application accuracy will be questioned until it has yielded consistently from various applications. The accuracy of data collection methods is also a key aspect that has been discussed by many researchers as part of real-time travel time estimation algorithms [Zhang, W., Freeway Travel Time Estimation Based on Spot Speed Measurements, in Civil Engineering. 2006, Virginia Tech: Blacksburg, VA., Cheevarunothai, P., Y. Wang, and N.L. Nihan, Identification and Correction of Dual-Loop Sensitivity Problems. Transportation Research Record, 2006(1945): p. 9., Hablas, H.E., A Study of Inclement Weather Impacts on Freeway Free-Flow Speed, in Civil Engineering. 2007, Virginia Tech: Blacksburg, VA., Weijermars, W.A.M. and E.C.V. Berkum, Detection of Invalid Loop Detector Data in Urban Areas. Transportation Research Record, 2006(1945): p. 7.]. For example, the data derived from inductance loop detectors are typically screened in order to enhance the quality of the data used in transportation applications. The most commonly used approach is to determine minimum and maximum acceptable values of volume, speed, and/or occupancy. Any data outside these ranges are regarded as invalid. For example, Coifman [Coifman, B., Using Dual Loop Speed Traps To Identify Detector Errors. Transportation Research Record, 1999(1683): p. 12.] introduced a method that identifies detector errors using dual loop speed trap data at both the upstream and downstream detectors in the same lane. Hablas [Hablas, H.E., A Study of Inclement Weather Impacts on Freeway Free-Flow Speed, in Civil Engineering. 2007, Virginia Tech: Blacksburg, VA.] attempted to investigate the impact of detector failure frequency and failure duration on the accuracy of loop detector speed, flow, and density measurements using a Monte Carlo simulation approach and developed regression models to relate loop detector accuracy to detector failure data.
Cheevarunothai [Cheevarunothai, P., Y. Wang, and N.L. Nihan, Identification and Correction of Dual-Loop Sensitivity Problems. Transportation Research Record, 2006(1945): p. 9.] presented an algorithm and its implementation for identifying and correcting loop sensitivity problems. Loop sensitivity-level discrepancies between two single loops forming a dual-loop detector and unsuitable sensitivity levels of the single loops are two major causes of quality degradation in dual-loop data. The proposed algorithm identifies dual-loop sensitivity problems using individual vehicle data extracted from loop event data and corrects dual-loop sensitivities to enhance the reliability of dual-loop detectors and improves the quality of traffic speed and bin volume data. Hellinga [Hellinga, B. and L. Fu, Assessing Expected Accuracy of Probe Vehicle Travel Time Reports. Journal of Transportation Engineering, 1999. 125(6): p. 7.] examined the issue of the accuracy of mean travel times as estimated from probe vehicles. The study concluded that the reliability of probe-based average link travel times is highly affected by sampling bias.
While the accuracy of the travel time information is heavily dependent on the data collection technique and travel time estimation methods, Toppen [Toppen, A. and K. Wunderlich, Travel Time Data Collection for Measurement of Advanced Traveler Information Systems Accuracy. 2003, Prepared for Federal Highway Administration: Washington D.C.] recommended the range of 13-21% travel time error is acceptable for ATIS applications using example applications in Los Angeles. The author concluded that an accuracy drop below a critical point deems relying on experience more efficient. Also at the highest levels of accuracy, little is gained by making further improvements and if the accuracy error is below 5%, it makes little sense to invest in improved accuracy.
Meanwhile, a recent study [Tarnoff, P.J., S.E. Young, J. Crunkleton, and N. Nezamuddin, Guide to benchmarking operations performance measures: Preliminary draft, Final report. 2008, University of Maryland, Center for Advanced Transportation Technology: College Park, Maryland.] also recommended that the acceptable accuracy range of travel time is between 10 and 20 percent for travel information applications. The study also found that if the error exceeds 20 percent, the public lose confidence in the information source, undermining the support and usefulness of the system. In addition, the application does not necessarily benefit from an increased accuracy below the specified range. The study also introduced different error ranges for different applications. For example, traffic engineering and traffic management applications requires errors between 5 and 10 percent to travel time systems, while transportation planning applications, including any type of planning or long-range monitoring activity, require a 5 to 15 percent error range for travel time data.
The report Traffic Data Quality Measurement proposed that the accuracy of travel times for traveler information systems be maintained within a 10 to 15 percent error range [Battelle, Traffic Data Quality Measurement. 2004, Prepared for Federal Highway Administration: Washington D.C.]. In addition, the report developed methods and tools to enable traffic data collectors and users to determine the quality of traffic data. While various studies recommend a 5 to 21 percent error range of travel time for real-time traveler information applications, the range of 10 to 17 percent travel time error would be a reasonable target for traveler information applications.
Coverage
Fujito [Fujito, I., R. Margiotta, W. Huang, and W.A. Perez, Effect of Sensor Spacing on Performance Measure Calculations. Transportation Research Record, 2006. 2006(1945): p. 11.] investigated the impact of sensor spacing along freeway corridors on the computation of performance measures using a travel time index. The study evaluated the effectiveness of loop detector spacing of 0.3, 0.6, 1.0, 2.0, 3.0, and 4.0 miles using data from Atlanta and 0.5, 1.0, 2.0, and 3.0 mile spacing using data from Cincinnati. The study found that increasing the sensor spacing led to over- or underestimating the travel time index relative to the baseline condition while no evidence was found that the travel time index measure became “worse” as the sensor spacing increased. However, the results appear to suggest that more sensors are not necessarily “better,” depending on the usage of the data. It does appear that, as a general rule, detector spacing of up to 1.0 mi should provide a reasonable estimate of performance measures for tracking congestion. The actual spacing between two adjacent detectors may be narrower or wider, depending on the local highway geometry (e.g., interchange locations). The study also showed that the actual location of the sensors is important in estimating the travel time index for a corridor. Thus strategically located sensors could significantly improve the performance measures of traffic collection systems.
The report Traffic Data Quality Measurement estimated “100 % area coverage” is recommended for the coverage measures. The coverage includes highway sections and major arterials.
Validity
TransGuide is an ATIS application which was designed by the San Antonio District of the Texas Department of Transportation (TxDOT). TransGuide provides the traveling public with real-time traveler information on traffic conditions, travel times, accidents, and construction. Turner [Turner, S., L. Albert, B. Gajewski, and W. Eisele, Archived Intelligent Transportation System Data Quality: Preliminary Analyses of San Antonio TransGuide Data. Transportation Research Record, 2000(1719): p. 8.] investigated the data quality analyzing loop detector data from the TransGuide system in San Antonio. The study utilized three attributes of data quality that are relevant to ITS data archiving: suspect or erroneous data, missing data, and data accuracy. The study found that in the analysis of TransGuide data, missing data accounted for about 25 percent of all data records. Error detection rules were developed to screen for suspect or erroneous data. It was concluded that data quality procedures are essential for ITS data applications.
Hablas [Hablas, H.E., A Study of Inclement Weather Impacts on Freeway Free-Flow Speed, in Civil Engineering. 2007, Virginia Tech: Blacksburg, VA.] investigated the relationships between the failures of loop detector and the accuracy of loop detector measurements using a Monte Carlo simulation approach. The research concluded that the errors of performance measures such as density, flow, and space mean speed increase as the frequency and duration of failures increase. The study also developed regression models that predict the accuracy of loop detector measurements using input parameters such as the failure frequency, the failure duration, and the traffic stream density. The report Traffic Data Quality Measurement projected that less than 10 % of the detector failure rate is acceptable for real-time travel time information.
Timeliness, Completeness, and Accessibility
The report Traffic Data Quality Measurement proposed that for the timeliness measure “the data is required close to real-time”. Also Tarnoff [Tarnoff, P.J., Getting to the INFOstructure. 2002, Prepared for TRB Roadway INFOstructure Conference.] suggested delay should be less than 1 minute for local implementation and less than 5 minutes for national implementation for the timeliness measure. The reports also suggested that 95 to 100 percent coverage was required for real-time travel time information.
For the accessibility measure, the real-time travel information application can be adequately serviced with access times in the 5 to 10 minute range while predictive traffic flow methods should access the information within 30 seconds. However, the traffic information from some private service providers shows a warning sign if the traffic information is more than 5 minutes old. Also sensor networks typically update real-time speed and volume information every few minutes.
4.3.2 Speed
Traffic stream speeds are typically measured in the field using a variety of spot speed measurement technologies. The most common of these spot speed measurement technologies is a presence-type loop detector, which identifies the presence and passage of vehicles over a short segment of roadway (typically 5 to 20 meters long). When a vehicle enters the detection zone, the sensor is activated and remains activated until the vehicle leaves the detection zone.
The average traffic stream speed can be computed in two different ways: a time-mean speed and a space-mean speed. The difference in speed computations is attributed to the fact that the space-mean speed reflects the average speed over a spatial section of roadway and thus is weighted by the traffic stream density, while the time-mean speed reflects the average speed of the traffic stream passing a specific stationary point. In other words, time-mean speed is the arithmetic mean of the speeds of vehicles passing a point on a highway during an interval of time. Alternatively, the space-mean speed is the harmonic mean of the speeds of vehicles passing a point on a highway during an interval of time.
Specifically, Daganzo [Daganzo, C.F., Fundamentals of Transportation and Traffic Operations. 3rd ed. 2003, Oxford, UK: Elsevier Science.] demonstrates that the space-mean speed is a density weighted average speed, while the time-mean speed is a flow weighted average speed. The space-mean speed reflects the spatial dimension of speed and thus is utilized in the standard speed-flow-density relationships.
Accuracy
The traditional practice for estimating speeds using single loop detectors is based on the assumption of a constant average effective vehicle length. Studies, however, have shown that this assumption provides speed estimates that are sufficiently inaccurate as to severely limit the usefulness of these speed estimates for real-time traffic management and traveler information systems [Hellinga, B.R., Improving Freeway Speed Estimates from Single-Loop Detectors. Journal of Transportation Engineering, 2002. 128(1): p. 10.]. In addressing these issues researchers have investigated the use of filtering techniques. For example, Dailey [Dailey, D.J., A Statistical Algorithm for Estimating Speed from Single Loop Volume and Occupancy Measurements. Transportation Research. Part B: Methodological, 1999. 33(5): p. 313.] developed a Kalman filter on vehicle length estimates while Hellinga [Hellinga, B.R., Improving Freeway Speed Estimates from Single-Loop Detectors. Journal of Transportation Engineering, 2002. 128(1): p. 10.] used exponentially smoothed adjacent dual loop detector vehicle length measurements to enhance the speed estimates of single loop detectors. Hellinga demonstrated that the exponential smoothing of 20-s average vehicle length measurements from adjacent dual loop detectors enhanced the accuracy of the speed estimates by approximately 20 percent. Wang and Nihan [Wang, Y. and N.L. Nihan, Freeway Traffic Speed Estimation with Single-Loop Outputs. Transportation Research Record, 2000(1727): p. 7.] used screening procedures to remove intervals with long vehicles and space-mean speed estimates were derived from the intervals with passenger cars only. Alternatively, researchers have investigated the use of median as opposed to mean statistics in order to enhance the robustness of the statistics by ensuring that the measures are not influenced by outlier observations. For example, Lin [Lin, W.-H., J. Dahlgren, and H. Huo, Enhancement of Vehicle Speed Estimation with Single Loop Detectors. Transportation Research Record, 2004(1870): p. 6.] used the median vehicle passage time as opposed to the mean passage time to estimate speeds from single loop detectors. Similarly, Coifman [Coifman, B., S. Dhoorjaty, and Z.-H. Lee, Estimating Median Velocity Instead of Mean Velocity At Single Loop Detectors. Transportation Research. Part C, Emerging technologies, 2003. 11(3): p. 12.] computed the median speed from the median occupancy in order to reduce speed estimate errors when a wide range of vehicle lengths are present in the traffic stream.
While the speed estimation is a key aspect of the accuracy of speed information for real-time traveler information, studies have recommended the speed accuracy requirements for a variety of applications. Tarnoff [Tarnoff, P.J., Getting to the INFOstructure. 2002, Prepared for TRB Roadway INFOstructure Conference.] proposed that for traveler information applications 20% or less error range is adequate for both local and national level implementations while the study recommended a 5 -10 % error range for traffic management applications. A recent study [Tarnoff, P.J., S.E. Young, J. Crunkleton, and N. Nezamuddin, Guide to benchmarking operations performance measures: Preliminary draft, Final report. 2008, University of Maryland, Center for Advanced Transportation Technology: College Park, Maryland.] recommended the following thresholds for speed accuracy:
- 1 – 5%, for traffic engineering applications
- 2 – 10%, for transportation planning applications
- 5 – 10%, for traffic management applications, and
- 5 – 20%, for traveler information applications.
Table 7. Speed Accuracy Requirement for Transportation Applications
Applications |
Recommended Accuracy Levels |
|---|---|
Free Flow Link Speeds for Planning Applications |
15 – 20 % |
Congested Link Speeds for Planning Applications |
At V/C < 1, 10 mph |
Transit Vehicle Speeds for Planning Applications |
15 – 20 % |
Free Flow Link Speeds for Traffic Simulation Applications |
5.0 % |
Congested Link Speeds for Traffic Simulation Applications |
2.5 % |
Corridor-level Vehicle Speeds for Congestion Management Applications |
5.0 % |
Completeness, Validity, Timeliness, Coverage, and Accessibility
The attributes of speed data is highly correlated with the travel time. Thus, similar to the real-time traveler information applications, the real-time speed information applications require more than 90% valid data, a less than 10% failure rate, real-time data, 100% area coverage, and 5-10 minute access time.
Additionally, the report Traffic Data Quality Measurement proposed acceptable levels of data quality for various applications, as summarized in Table 8 [Battelle, Traffic Data Quality Measurement. 2004, Prepared for Federal Highway Administration: Washington D.C.]. The table demonstrates that most speed applications require a 90 – 100 % validity check and 100 % coverage of the study area excluding transit vehicle speed applications, which require more than 95% completeness and 10 – 15% failure rates. Also the timeliness of each application varies from 6 months to three years.
Table 8. Data Quality Requirement for Transportation Applications
Applications |
Completeness |
Validity |
Timeliness |
Coverage |
|---|---|---|---|---|
Free Flow and Congested Link Speeds for Planning Applications |
90-100% validity for instrumented floating car data collection |
90-100% validity for instrumented floating car data collection |
Within three years of model validation year |
100% freeway |
Transit Vehicle Speeds for Planning Applications |
Less than 5% - one peak and one off-peak route |
Up to 15% failure rate for 48 hours counts |
Within three years of model validation year |
100% of study area |
Free Flow Link Speeds for Traffic Simulation Applications |
90-100% validity for instrumented floating car data collection |
90-100% validity for instrumented floating car data collection |
Within one year of study |
100% of study area |
Congested Link Speeds for Traffic Simulation Applications |
90-100% validity for instrumented floating car data collection |
90-100% validity for instrumented floating car data collection |
Within one year of study |
100% of study area |
Corridor-level Vehicle Speeds for Congestion Management Applications |
90-100% validity for instrumented floating car data collection |
90-100% validity for instrumented floating car data collection |
Within six month year of study |
100% of study area |
4.3.3 Weather Data
Low visibility, precipitation, high winds, and extreme temperature can affect driver capabilities, sight distances, vehicle performance, and infrastructure characteristics. ITS applications allow traffic managers to disseminate advisory and regulatory traveler information to motorists. These systems also facilitate sharing of road weather data among managers in multiple agencies and neighboring jurisdictions. To improve traffic operations under adverse environmental conditions, traveler information may be furnished through roadside warning systems, web-based applications, interactive telephone systems such as 511, and roadway weather information web sites. The goals of the systems are to provide decision support information to travelers in a manner that may enhance efficiency and safety. One key feature in this decision support information is to improve access to real-time and forecasted weather conditions [Pisano, P.A. and L.C. Goodwin. Surface Transportation Weather Applications. in the ITE Annual Meeting and Exhibit. 2002. Philadelphia, PA, USA., Osborne, L.F. and M.S. Owens, Evaluation Of The Operation And Demonstration Test of Short-Range Weather Forecasting Decision Support within an Advanced Rural Traveler Information System. 2000, Prepared for Federal Highway Administration: Washington D.C.].
511 typically provides information on current and changing travel conditions and forecasts for upcoming weather events that are likely to impact the ability to travel. Weather information for 511 on a segment-by-segment basis needs to be focused on the travel impact of weather conditions. Segments need to be defined at a logical length to reflect the possible weather conditions and variation in conditions along segments. 511 weather service offers the following weather related information for a segment or location depending on the system location: temperature, wind speed and direction, precipitation rate, sky condition, visibility in miles and eighths of a mile once visibility is below a mile, accumulation (for snow events), air quality, and pavement temperature and condition(dry, wet, icy) [511 Deployment Coalition, America’s Travel Information Number: Deployment Assistance Report #6, Weather and Environmental Content on 511 Services. 2003.].
Many state DOTs also provide textual and graphical road weather information on the internet [Fayish, A.C. and P.P. Jovanis, Usability of Statewide Web-Based Roadway Weather Information System. Transportation Research Record, 2004(1899): p. 11.]. The most advanced is the Washington State DOT traffic and weather information web site that collects data from a variety of sources, and displays current and forecasted pavement and weather conditions on a color-coded statewide map. The DOT accesses real-time data from meteorological observing networks, a CCTV surveillance system, mountain pass reports, and various satellite and radar images. Also interactive voice response technology to provide route-specific road condition reports and six-hour weather forecasts to drivers on highways is utilized as Weather Information Systems in many other states [Pisano, P.A. and L.C. Goodwin. Surface Transportation Weather Applications. in the ITE Annual Meeting and Exhibit. 2002. Philadelphia, PA, USA.]. The data quality measures of weather information are described in the following sections [511 Deployment Coalition, America’s Travel Information Number: Deployment Assistance Report #6, Weather and Environmental Content on 511 Services. 2003., Fayish, A.C. and P.P. Jovanis, Usability of Statewide Web-Based Roadway Weather Information System. Transportation Research Record, 2004(1899): p. 11.].
Accuracy
Weather Information should contain information that matches actual conditions. Hourly comparison between FAA and National Weather Service weather sensor observations and road–weather sensor observations in close geographical proximity should be considered.
The accuracy of weather information is also dependent on data collection technologies such as weather sensors. Visibility is based on light scattering. The visibility sensor projects a beam of light over a very short distance, and the light that is scattered is detected by a receiver. The amount of light scattered and then received by the sensor is converted into a visibility value. A one-minute average visibility is calculated and the value is stored for the next 10 minutes. A harmonic mean is used rather than an arithmetic mean because it is more responsive to rapidly decreasing visibility conditions. The location of the visibility sensor is critical. The sensor should be located in the area of most concern. For aircraft navigation, most primary visibility sensors are placed near the touchdown zone of the primary instrumented runway. The sensor must be located at least 10 feet above ground level. Visibility is reported in quarter-mile increments up to two miles, then at 2.5 miles, then at every mile to a maximum of 10 miles. Visibilities greater than 10 miles are still reported as 10 miles. Values less than a quarter mile are reported as a quarter mile [Hablas, H.E., A Study of Inclement Weather Impacts on Freeway Free-Flow Speed, in Civil Engineering. 2007, Virginia Tech: Blacksburg, VA.].
The Precipitation Identifier (PI) sensor distinguishes between rain (RA) and snow (SN) while the Freezing Rain (FZRA) sensor detects freezing rain. The Freezing Rain sensor measures accumulation rates as low as 0.01 inches per hour. The PI sensor reports data every minute as a 10-minue moving average and stores the data in memory for 12 hours. If more than three data items are missing, the algorithm reports “missing”, if an equal number of different precipitation types are reported in the last 10 minutes the heavier is reported. After the determination of the precipitation type the algorithm calculates the intensity (light-moderate-heavy) and it is determined from the highest common intensity derived from three or more samples of data [Hablas, H.E., A Study of Inclement Weather Impacts on Freeway Free-Flow Speed, in Civil Engineering. 2007, Virginia Tech: Blacksburg, VA.].
Completeness
100% of weather data (24 hr/7 days) should be prepared to provide a reliable stream of information to travelers.
Timeliness
Weather information is recommended to be timely to the greatest extent possible in accordance to the speed of the weather change anticipated. In many rapidly evolving situations this can imply an hourly update with change notices of weather variations at hourly intervals. While it is recognized that non-urban areas will have more difficulty collecting, inserting and updating information quickly, it is recommended that every attempt be made in both urban and non-urban areas to update information as soon as there is a known deviation from the current route segment report.
Coverage
100% of functional weather/roadway sensors of study area should be covered.
Other attributes
It is also recommended that weather reports use the same, or similar, terminology to describe conditions. Lack of consistent terminology leads to misunderstanding and confusion and consistent terminology will make the system more usable as users move from one system to another.
4.3.4 Summary of Data Quality Measures
The previous sections described data quality measures for sample parameters including: travel time, speed, and weather information. Table 9 presents a summary of recommended data quality measures for each of the sample parameters.
Table 9. Recommended Data Quality Measures for Real-time Travel Information Applications
|
Travel Time |
Speed |
Weather Information |
|---|---|---|---|
Accuracy |
10-17% error range for data collection and travel time estimation |
5-20% error range for speed measurement or estimation |
Recommended to contain information that matches actual conditions |
Completeness |
95-100% temporal coverage |
95-100% temporal coverage |
100% (24 hr/7 days) |
Validity |
90-100% validity for sensor or instrumented car data collection |
90-100% validity for sensor or instrumented car data collection |
90-100% validity for sensor |
Timeliness |
Less than 1 minute for local implementation and less than 5 minutes for national implementation |
Real-time |
An hourly update |
Coverage |
Sensor spacing of 1 mile and 100% area coverage |
100% area coverage |
100% area coverage |
Accessibility |
Less than 5 minutes and warning system if traffic information is more than 5 minutes old |
Less than 5 minutes and warning system if traffic information is more than 5 minutes old |
5-10 minutes |
