Download TTI Report 1752-2 and more Lecture notes Database Management Systems (DBMS) in PDF only on Docsity!
Technical Report Documentation Page
- Report No. 2. Government Accession No. 3. Recipient's Catalog No. FHWA/TX-98/1752-
- Title and Subtitle 5. Report Date ITS DATA MANAGEMENT SYSTEM: YEAR ONE ACTIVITIES August 1997
- Performing Organization Code
- Author(s) 8. Performing Organization Report No. Shawn M. Turner, Robert E. Brydia, Jyh C. Liu, and William L. Eisele Research Report 1752-
- Performing Organization Name and Address 10. Work Unit No. (TRAIS) Texas Transportation Institute The Texas A&M University System College Station, Texas 77843-
- Contract or Grant No. Study No. 0-
- Sponsoring Agency Name and Address 13. Type of Report and Period Covered Texas Department of Transportation Research Report: Research and Technology Transfer Office September 1996 - August 1997 P. O. Box 5080 Austin, Texas 78763-
- Sponsoring Agency Code
- Supplementary Notes Research performed in cooperation with the Texas Department of Transportation and the U.S. Department of Transportation, Federal Highway Administration. Research Study Title: TransLink ®Research Program
- Abstract
This report documents research conducted in the development of an ITS data management system, hereafter referred to as “ITS DataLink.” The objective of the ITS DataLink system is to retain, manage, share, and analyze ITS data for a variety of transportation analyses. A survey of selected traffic management centers found a number of issues related to the retention and sharing of ITS data, including institutional communication and data management issues. Initial sections of this report include these and other survey findings. Background information is also provided on the definition and selection of transportation performance measures.
Later sections of this report contain specific information on the development of the ITS DataLink system. Based upon the literature review, the report summarizes applicable performance measures that can easily be calculated with ITS data. The report includes examples of performance measure calculation using ITS data from the TransGuide center in San Antonio, Texas. The report documents the desirable features of an ITS data management system. The final sections of the report contain documentation of the web browser-based ITS DataLink system, which has been undergoing online testing and refinement since early 1997. The report concludes with findings on ITS data management systems and contains recommendations for future refinements of the ITS DataLink system.
- Key Words 18. Distribution Statement Data Warehouse, Data Management, Performance No restrictions. This document is available to the Measures, Traffic Management Centers, Loop public through NTIS: Detectors, ITS, ATMS National Technical Information Service 5285 Port Royal Road Springfield, Virginia 22161
- Security Classif.(of this report) 20. Security Classif.(of this page) 21. No. of Pages 22. Price Unclassified Unclassified 66 Form DOT F 1700.7 (8-72) Reproduction of completed page authorized
v
DISCLAIMER
The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the data presented herein. The contents do not necessarily reflect the official views or policies of the Texas Department of Transportation or the Federal Highway Administration. This report does not constitute a standard, specification, or regulation. Shawn Turner (Texas certification number 82781), Robert Brydia, Jyh (Steve) Liu, and William Eisele prepared this report.
vii
TABLE OF CONTENTS
PAGE
LIST OF FIGURES......................................................... ix
viii
TABLE OF CONTENTS (Continued)
- 1 INTRODUCTION LIST OF TABLES x
- 1.1 Problem Statement
- 1.2 Research Objectives and Scope
- 1.3 Organization of Report
- 2 BACKGROUND
- 2.1 Review of ITS Data Retention and Management Practices
- 2.1.1 Phoenix Traffic Operations Center
- 2.1.2 Los Angeles District 7 Traffic Management Center
- 2.1.3 San Francisco Bay Area Traffic Management Center
- 2.1.4 Georgia DOT Advanced Transportation Management System
- 2.1.5 Illinois Traffic Systems Center
- 2.1.6 Montgomery County Transportation Management Center
- 2.1.7 Detroit, Michigan ITS Center
- 2.1.8 Minneapolis, Minnesota Traffic Management Center
- 2.1.9 TRANSCOM
- 2.1.10 INFORM
- 2.1.11 MetroCommute
- 2.1.12 TranStar Traffic Management Center
- 2.1.13 TransGuide Advanced Traffic Management System
- 2.1.14 North Seattle Advanced Traffic Management System (ATMS)
- 2.1.15 Toronto’s COMPASS
- 2.1.16 Summary of Findings
- 2.2 Summary of Data Retention and Management Practices
- 2.3 Performance Measures Based Upon Goals and Objectives
- 2.4 Previous Research on Transportation Performance Measures
- 2.5 Summary of Performance Measure Research
- 3 APPLICATIONS AND PERFORMANCE MEASURES USING ITS DATA
- 3.1 Identifying Needs and Potential Uses for ITS Data
- 3.2 Estimating Performance Measures from ITS Data
- 3.3 Example Calculation of Performance Measures from ITS Data
- 4 DEVELOPMENT OF AN ITS DATA MANAGEMENT SYSTEM PAGE
- 4.1 Background
- 4.1.1 Data Storage
- 4.1.2 Database Construction
- 4.1.3 Access to Data
- 4.1.4 Data Versus Information
- 4.1.5 Privacy
- 4.2 Data Management System Features
- 4.2.1 Data Storage
- 4.2.2 Database Access
- 4.2.3 User Interface
- 4.2.4 Data Aggregation and Summarization
- 4.2.5 Calculate Performance Measures
- 4.2.6 Tabular and Graphical Output Formats
- 4.3 Documentation of the Prototype Data Management System
- 4.3.1 System Architecture
- 4.3.2 ITS DataLink System Cost
- 5 FINDINGS AND CONCLUSIONS
- 6 RECOMMENDATIONS
- 7 REFERENCES
x
LIST OF TABLES
PAGE
Table 2-1. Telephone Survey Used at Selected TMCs................................ 3 Table 2-2. Summary of TMCs Interviewed in the Study.............................. 4 Table 2-3. Level of Permanent Data Aggregation at Selected TMCs in North America...... 14 Table 2-4. Recommended Measures of Effectiveness for Various Objectives.............. 19 Table 2-5. Criteria for Goal-Achievement Evaluation............................... 20 Table 2-6. Classification of Criteria............................................. 21 Table 2-7. Performance Measures for Congestion Management Systems................. 24 Table 2-8. Candidate Measures of Effectiveness................................... 26 Table 2-9. Preferred Measures for Evaluating the Performance and Impact of Transportation Improvements............................. 27 Table 3-1. Example Matrix: Level of Aggregation for ITS Data Applications............. 30 Table 3-2. Example of Five-Minute Point Summaries of Twenty-Second Loop Detector Data............................................... 36 Table 3-3. Example of Fifteen-Minute Link Summaries of Loop Detector Data............ 37 Table 3-4. Example of Peak Hour Corridor Summaries from Loop Detector Data......... 38 Table 4-1. Approximate Storage Space and Costs for One Complete Year of TransGuide Loop Detector Data.................................... 45 Table 4-2. Performance Measures in ITS DataLink................................. 47
1 INTRODUCTION
Many intelligent transportation system (ITS) components are being installed in U.S. urban areas, and vast amounts of data are currently available through these ITS components. Vehicle detectors (e.g., inductance loop, video, infrared, sonic, or radar) collect information about vehicle volumes, speeds, and lane occupancies. Probe vehicle systems utilize technologies such as the global positioning system (GPS) and radio/cellular phone triangulation to determine vehicle positions at frequent time intervals. Automatic vehicle identification (AVI) uses transponder polling to collect vehicle information at instrumented locations, enabling the collection of travel times and average speeds along roadway sections. This ITS data is often used only for real-time operations and then discarded. Some traffic management centers (TMCs) save the data, but few have a mechanism for sharing the data resources among other transportation groups or agencies within the same jurisdiction. Even when TMCs save data, they face issues related to transforming the vast amounts of data into useful information for adjusting operating strategies, evaluating system performance, or making decisions about future transportation investments.
1.1 Problem Statement
Many transportation analysts and researchers struggle to obtain accurate, reliable data about existing transportation performance and patterns. Models rely heavily on existing conditions for calibration purposes, and decision-makers rely on models and the existing transportation performance to make decisions about transportation investments. The importance of accurate, reliable data in transportation analyses is paramount to sound decisions in planning, designing, operating, and maintaining the transportation system.
An opportunity exists to utilize the vast amounts of data available through ITS for a wide variety of planning, design, operation, and evaluation purposes. The problem lies in developing a framework for retaining, managing, sharing, and analyzing the data (i.e., transforming the “mountains” of data into useful information). The information can then be used for various transportation analyses.
1.2 Research Objectives and Scope
The primary objective of this research is to explore several issues relating to the development of ITS data management systems. These issues include the following:
- What are the needs and potential uses of ITS data beyond real-time applications?
- What performance measures are needed for these applications, and can the available ITS data be used to calculate these performance measures?
- What computer hardware and software does an ITS data management system require?
- What are the considerations for storing and aggregating ITS data?
Data Management Questionnaire for Traffic Management Centers (TMCs)
- How large is the monitored system in centerline-miles? What data are being collected (e.g., volume, speed, occupancy, classification)?
- How are the data collected (e.g., loops, AVI, CCTV, other types of technology)?
- How often are data sent to the center or “polled” from the center? Is aggregation of the data performed at local controller units (i.e., a central or distributed system)?
- Are data saved for future use? If so, for what future uses are data saved?
- If data are saved, are they aggregated? What is the time increment over which the data are aggregated? How was this increment determined?
- What are the most common uses of the data? Who are the most common users of the data?
- What are the typical agencies (or people) making requests for the data (e.g., MPOs, universities)?
- Does the TMC have policies established for data storage and/or data sharing?
- What are the related privacy concerns that have been encountered or are anticipated with the data storage or sharing?
2 BACKGROUND
This chapter provides a review of ITS data retention practices at selected traffic management centers (TMCs) and summarizes previous research on the selection and use of performance measures. Researchers surveyed selected TMCs on their retention and use of ITS data for non-real-time applications. The first part of this chapter summarizes the results of this TMC survey. The second part of the chapter presents various research on selection and use of performance measures in transportation analyses.
2.1 Review of ITS Data Retention and Management Practices
To better understand the current practices of retaining ITS data for future transportation analyses, the research team surveyed selected TMCs across North America. The survey questions shown in Table 2-1 were developed and administered by telephone to determine the data management strategies in existence and the opportunities that may exist for utilizing ITS data.
Table 2-1. Telephone Survey Used at Selected TMCs
Table 2-2 shows the TMCs that the researchers interviewed in this study. The table provides the center name, location, web site, and contact information. Figure 2-1 shows the geographic location of the TMCs contacted for the study.
Table 2-2. Summary of TMCs Interviewed in the Study
Location Name Internet Site (^) Person/PhoneContact Agency
Phoenix, Traffic Operations Phil Carter Arizona Center http://www.azfms.com^ (602) 255-7754 Arizona DOT Los Angeles, Los Angeles District http://www.scubed.com/ David Lau California 7 TMC caltrans/la/la_transnet.html (213) 897-4385 Caltrans Oakland, San Francisco Bay http://www.dot.ca.gov/ Jack Allen California Area TMC dist4/links.htm (510) 286-5761 Caltrans Atlanta, http://www.georgia- Dennis Reynolds Georgia Georgia DOT ATMS^ traveler.com (404) 635-1031 Georgia DOT Oak Park, Traffic Systems http://www.ai.eecs.uic.edu/ Tony Cioffi Illinois Center GCM/GCM.html (708) 524-2145 Illinois DOT
Rockville, Montgomery County John Riehl County Dept. of Maryland TMC http://www.dpwt.com (301) 217-2190 Public Works and
Montgomery
Transportation Detroit, http://campus.merit.net/ Ross Brehmer Michigan Michigan ITS Center^ mdot/its.html (313) 256-9800 Michigan DOT Minneapolis, http://www.dot.state.mn.us/ Ron Dahl Minnesota Minnesota TMC^ tmc (612) 341-7269 Minnesota DOT
Jersey City, Sanjay Patel New Jersey TRANSCOM^ csnj/issues/^ (201) 963-4033 TRANSCOM
http://www.travroute.com/ apr96transcom.html Long Island, http://metrocommute.com/ Ray Schiemes New York INFORM^ LI/inform.html (516) 952-6872 New York DOT New York City, Dan Broe New York MetroCommute^ http://metrocommute.com^ (212) 406-9610 MetroCommute Houston, Carlton Allen Texas TranStar TMC^ http://traffic.tamu.edu^ (713) 881-3285 Texas DOT San Antonio, http://www.transguide.dot. David Kingery Southwest Texas TransGuide ATMS state.tx.us/overview.html (210) 731-5154 Research Institute
Seattle, Mahrokh Arefi Washington North Seattle ATMS^ regions/northwest/nsatms/^ (206) 440-4462 Washington DOT
http://www.wsdot.wa.gov/ atmsmain.htm Toronto, David Tsui Ontario Ministry Ontario, Canada COMPASS^ http://compass.gov.on.ca (416) 235-3538 of Transportation
The developers of the TOC in Phoenix realized that the 20-second data would inevitably be valuable information for transportation applications and research. In addition, since the Center has not been in operation long, data storage has not become a significant concern. Currently, a majority of the users of the data are DOT or TMC personnel; however, some requests have been made by universities for research purposes. Generally, operational requests desire five-minute data. Fifteen-minute summaries of the data are often desired for simple traffic counts to estimate construction impacts and for typical design work.
TOC personnel discussed several issues relating to data management considerations and uses for transportation analyses. The first consideration is that the large amounts of data are very difficult to manage with most spreadsheet and database software. In one year, the Phoenix TOC generates 15 million lines of data. This often requires that the disaggregate (e.g., 20-second) data be put into an aggregated form (e.g., 5 or 15 minutes) to perform analyses. The issue of adequate analysis tools should be considered in the planning stage of the data management system.
Another important point made was the interest in a larger system of ITS data collection, aggregation, and dissemination. The point was made that a National ITS database (e.g., similar to the Highway Performance Monitoring System, or HPMS) may be more appropriate than one that simply evaluates a single location or metropolitan area. Although local systems are clearly valuable, a National database or repository would allow for a central storage of data from several urban areas. Eventually, such a system might allow for comparisons to be made between urban areas.
2.1.2 Los Angeles District 7 Traffic Management Center
The Los Angeles Traffic Management Center (TMC) has been in operation for several years and monitors 1,204 directional km (748 directional miles) of roadway. The TMC is primarily used to monitor and verify incidents in the freeway coverage area. Both inductance loops and CCTV are in operation in the monitoring system. DOT personnel use the CCTVs for verifying incidents and do not record the video for future use due to motorist privacy and liability concerns. Two television stations have a connection to the CCTVs as well. The media do not have control over the cameras; they simply have a connection to the CCTVs for simple viewing.
The TMC polls the loop stations every 30 seconds and data from these inductance loops are saved for future use. Over 1,000 loop stations placed at approximately 0.8 km (0.5 mile) spacings in each lane provide volume and lane occupancy data. The loop stations consist of single loops, and speed is estimated from the volume and lane occupancy data. The TMC saves three days of 30-second data and four days of five-minute summaries into temporary storage. The TMC has saved the 30-second data onto circular tapes since the opening of the Center. The data saved to temporary storage is much easier to retrieve than the data stored on tapes on the mainframe computer. DOT personnel developed special software that is necessary for data retrieval from the mainframe computer.
Typical users of the data include a local cable channel that provides a map of the freeways with incident information. In addition, this information is posted to the World Wide Web site. Local universities have also used the data for research purposes.
2.1.3 San Francisco Bay Area Traffic Management Center
In the near future, the San Francisco Traffic Management Center (TMC) will cover all the freeways in the nine-county bay area. The TMC uses loops to obtain volume, speed, and lane occupancy data. They estimate vehicle class based upon the volume, speed and lane occupancy data. The TMC does not currently utilize CCTVs but anticipates them to be part of the future system for monitoring freeway conditions. Loop data are sent to the TMC every 30 seconds in a binary format and eventually converted to an ASCII-text format.
The TMC in San Francisco does not archive data. The San Francisco Bay Area TMC personnel recognize the importance of the data for future transportation applications. Unfortunately, the contract for the TMC development did not provide consideration for the saving of data. An effort is underway within the TMC to develop procedures for archiving and summarizing the data in a desired format. This effort is being lead by the Office of Environmental Engineering and Modeling and Forecasting. Currently, their staff is evaluating methods to aggregate the data into the five-minute level and to peak hours per lane. They wish to have the system aggregating and producing five- and fifteen-minute summaries within the next six months. The TMC personnel commented that for planning and modeling purposes, five- minute data summaries are most desirable. Requests at the TMC from researchers tend to be for much more detailed data. Other area transportation agencies, including the Metropolitan Transportation Commission (MTC), have expressed an interest in the development of an ITS data management system.
One difficulty in the development of the system appears to be the large quantity of data available for use. The size alone makes the large amount of data difficult to manage. Determining the type of aggregation level to provide is also a concern since different users appear to desire different levels of aggregation for different applications.
2.1.4 Georgia DOT Advanced Transportation Management System
The Georgia DOT opened their Advanced Transportation Management System (ATMS) in May 1996. This system utilizes 360 video detection cameras to monitor 101 km (63 miles) of freeway in the Atlanta metropolitan area. The cameras obtain vehicle volume and lane occupancy data and estimate speed information. The video detection cameras are located approximately every 0.54 km (0.33 mile) in the monitored system.
Data are sent to the ATMS from the loop stations every 20 seconds and are not currently archived at the Center. DOT personnel use the data for monitoring the system and investigating incidents and accidents. The DOT has begun efforts to save the data at hourly and 15-minute
summaries of volume data. Due to budgetary constraints, this is the only data that is being saved at the Center.
There are several users for the data including internal DOT requests (e.g., planning and operations divisions), nearby universities, and the Southeast Michigan Council of Governments (the local Metropolitan Planning Organization).
A major concern at the Michigan ITS Center is the accuracy of the loop data that are being collected. This issue is not unique to the Michigan Center as detector reliability is often questioned and maintenance is high. Algorithms have been created and implemented into the programs that create the flow maps by comparing determined or reported speeds to adjacent sections. Software that provides a reasonable estimate of “smoothed” speeds along a roadway is a useful technique to address this concern.
2.1.8 Minneapolis, Minnesota Traffic Management Center
The Minnesota Traffic Management Center (TMC) monitors 282 km (175 miles) ( percent) of the freeways in the twin cities metropolitan area. Approximately 3,000 loop detectors are in place at 0.8 km (0.5 mile) spacings. In addition, there are 180 CCTVs throughout the system at a 0.6 km (1.0 mi) spacing. TMC personnel use the CCTVs to monitor incidents and verify algorithms. The loop detectors collect average volume per lane across lanes at a given location, occupancy (presence), and vehicle length. Average speed is estimated from this data. Center personnel note that this estimated speed is an adequate estimate for most uses, including speed maps on the World Wide Web.
Loop stations are polled from the Center every 30 seconds as an average across lanes. Every five minutes, the data are aggregated and saved. The TMC software compress the five- minute data from 2.3 MB to 1.5 MB and save it on the hard disk every day. The process is automatic and the data have been saved since the Center opened in 1993. In addition, the 30- second mainlane loop data are also archived and logged. A seven-day “wrap around” file is kept in readily accessible computer disk storage for easy access. The decision to save the data at the 30-second level was made since the space was available and the lowest level of disaggregation seemed necessary for the requests of the data being made. In addition, summaries of 5 minutes, 15 minutes, or other time periods can easily be obtained from the 30 second data. Most requests are from the DOT for traffic analysis, construction impact determination, and planning applications. Researchers from local universities often request the data as well. Due to many data requests from agencies and individuals external to the Center, a data distribution service (DDS) has been developed. This service allows interested individuals and agencies to receive the 30-second data stream through the Internet.
2.1.9 TRANSCOM
TRANSCOM (the Transportation Operations Coordinating Committee) opened in 1986 to coordinate traffic management in New York, New Jersey, and Connecticut. TRANSCOM was one of four TMCs in the United States to be selected for funding for the Model Deployment Initiative. Several agencies in the tri-state area supply traffic management information to TRANSCOM.
A project of particular interest is TRANSCOM’s System for Managing Incidents and Traffic (TRANSMIT). The TRANSMIT system is using vehicles with automatic vehicle identification (AVI) transponders installed on the vehicles. These “E-Z Pass” tags are used for toll collection and also allow for using the vehicles as probes of the traffic flow. The TRANSMIT system collects speed and travel time data for use in incident detection. The TRANSCOM system also uses CCTVs for verification of the AVI system and incident management, but does not save CCTV video. The tag “reads” come into the Center in real-time as they occur in the field. The TMC software aggregates and saves this information into 15- minute time periods. The system does not save real-time tag reads for each vehicle. Furthermore, privacy concerns are eliminated since readings of the tag account numbers are not saved by the system. A tag number is read; however, it is a dummy tag number that cannot be used to trace individual vehicles. The system has been in operation for six months and TRANSCOM operators are pleased with its performance. TRANSCOM personnel claim that the system costs much less than loop installation and maintenance, especially since lane closure is not necessary for routine repairs and maintenance. Requests for the AVI data have been for origin-destination surveys and planning for construction.
The AVI data is the only data that TRANSCOM saves. Centers throughout the tri-state area simply send information regarding incidents to TRANSCOM but not the raw data that is used in their algorithms. TRANSCOM personnel save text files which describe the incidents that have occurred and been managed by the Center.
2.1.10 INFORM
INFORM (INformation FOR Motorists) monitors 56 km (35 miles) of the central corridor of the Long Island Expressway (Route 495). The Center polls the 2,400 loop detectors 60 times per second. The data are then aggregated to 1/4 second, one minute, and 15 minutes at the Center. The loops are located at 0.8 km (0.5 mile) spacings, one per lane, and collect volume, lane occupancy, and vehicle length data. Vehicle speeds are estimated from this data. The 15- minute data are saved every day in a compressed ASCII text file in a 15-minute format. These data are archived for a three month period and then the tapes are overwritten.
The data management system was developed due to budget and space constraints. Typical data requests are internal and are from planning, operations, and design groups. Of particular interest is the fact that the planning department would like the data in a five minute
