Abstract
Urban traffic data can be used for other purposes besides actuated signal control. It is however important that the data is of sufficient quality. This paper discusses a procedure for the detection of invalid traffic data produced by single loop detectors at signalized intersections. Basic quality checks -based on minimum and maximum flow thresholds- are used for the detection of erroneous data, whilst quality checks based on the principle of conservation of vehicles are introduced for the detection of inaccurate data. In most cases, the principle of conservation of vehicles implies that flow measurements have to be consistent between upstream and monitoring detectors within one intersection.
The basic quality checks are appropriate for the detection of erroneous data. Moreover, quality checks based on the principle of conservation of vehicles are a useful addition to basic quality checks, since they are appropriate for the detection of inaccurate traffic data that was not detected yet by the basic quality checks. However, since flows are inconsistent between upstream detectors mutually as well in some cases, it is not always clear whether the results of this quality check are reliable. These cases were analyzed further by means of field observations. Finally, since the designed control procedure detects more invalid data than the detector station itself it can be concluded that these macroscopic quality checks are a useful addition to the microscopic quality checks.
The basic quality checks are appropriate for the detection of erroneous data. Moreover, quality checks based on the principle of conservation of vehicles are a useful addition to basic quality checks, since they are appropriate for the detection of inaccurate traffic data that was not detected yet by the basic quality checks. However, since flows are inconsistent between upstream detectors mutually as well in some cases, it is not always clear whether the results of this quality check are reliable. These cases were analyzed further by means of field observations. Finally, since the designed control procedure detects more invalid data than the detector station itself it can be concluded that these macroscopic quality checks are a useful addition to the microscopic quality checks.
Original language | English |
---|---|
Title of host publication | Proceedings of the 85th Annual meeting of the Transportation Research Board, 22-26 January 2006, Washington DC, USA |
Place of Publication | Washington, DC |
Publisher | Transportation Research Board (TRB) |
Number of pages | 16 |
Publication status | Published - 22 Jan 2006 |
Event | 85th Transportation Research Board (TRB) Annual Meeting 2006 - Washington, United States Duration: 22 Jan 2006 → 26 Jan 2006 Conference number: 85 |
Conference
Conference | 85th Transportation Research Board (TRB) Annual Meeting 2006 |
---|---|
Country/Territory | United States |
City | Washington |
Period | 22/01/06 → 26/01/06 |
Keywords
- METIS-235350