[ITS UW
Logo]

 TDAD Frequently Asked Questions

What is the range of data available?

The data collection was begun on October 13, 1998, and has been operating continuously since then.  There may be occasional gaps in the data collection due to system failures, but the gaps sum to less than 8% of the total coverage.  The database contains measurements from all of the sensors, taken at 20-second intervals.

What are the differerent sensor types?

The "loop" sensor type is the physical vehicle detection sensor which is capable of monitoring a single lane of highway.  Every cabinet contains one or more loop sensors.  In addition, for each cabinet location, one or more virtual sensors may be defined algorithmically:  "station" sensors sum the measurements taken by loops across all lanes in one direction, and "speed trap" sensors calculate an average speed estimate.

What is the naming convention for the cabinets and sensors?

first seven characters of the sensor name identify the cabinet in which it is found.  The final character of the cabinet name can be either D or R, to distinguish between ramp-type cabinets (R), used for ramp metering, and freeway-type cabinets (D). Some more recent cabinets have an M in this position, meaning that the data collectors are microwave (radar) detectors.

The sensor name contains 7 characters that describe what the sensor is:

For example, ES-070R:_MNH__5 is a loop sensor in a ramp-type cabinet, measuring traffic on a mainline northbound road, in the fifth lane, which is an HOV lane. (More info about loop naming conventions.)

What is the meaning of the DATA_TIME column in the data output file?

This column encodes the date and time at which a given measurement was taken.  The values at the beginning and end of an output file should correspond to the time boundaries specified with the query interface.  Each sensor report describes a 20 second interval, so the DATA_TIME values should increase in 20-second increments.  The date-time is encoded in a character string with the following format: YYYYMMDDHHmmSSsss, where Y=year, M=month, D=day, H=hour, m=minute, S=second, s=millisecond.

What is a scan count? How is occupancy calculated?

The data comes from loop detectors embedded in the pavement, each 6 feet square. The loops operate in presence mode (that is, they turn on and stay on as long as a vehicle is over, or occupying, the loop). The field microprocessor (an 8-bit 6808-based machine) checks, or scans, the loop 60 times each second. Every time the scan reports that the loop is occupied, the scan counter is incremented. And, 60 scans-per-second multiplied by 20 seconds-per-period equals 1200. Thus, the scan count divided by 1200 and multiplied by 100 equals the occupancy in percentage. Or, you can just divide scan count by 12 to get percentage directly. (This comes from WSDOT documentation.)

What is volume?

Volume is a measurement of the number of vehicles that passed in the 20-second interval.

What is the meaning of the "flag" field?

For loop and station sensors, the "flag" field contains the value 0 if the sensor is correctly functioning. Speed trap sensors are more complex and implement various checks on the reasonableness of their data.  If any of these checks fail, a non-zero value is found in the "flag1" field.  As well, the value 8 will be found in the "flag2" field if the sensor is functioning correctly.  Therefore, valid rows of data from a speed trap sensor are those for which flag1 = 0, and flag2 = 8 or higher. The details of the loop flags and speed trap flags are available from the source code.

Is there any further information on the loop sensor layouts?

Yes, WSDOT's web site includes a summary of 2000 traffic volumes, with explanations of loop sensor configurations.

[Blue Bar]

TDAD Project | ITS UW Home

webmaster@its.washington.edu
(March 1, 2002)