TY - JOUR
T1 - Integrating intersection traffic signal data into a traffic monitoring program
AU - Guin, Angshuman
AU - Hunter, Michael
AU - Rodgers, Michael
AU - Anderson, James
AU - Susten, Scott
AU - Wiegand, Kiisa
PY - 2016
Y1 - 2016
N2 - There are ongoing efforts to leverage the large volume of data collected to support real-time traffic operations for various non-real-time uses. However, adapting data streams to a different purpose can be challenging, especially if different applications use the data in different ways. Although traffic operations and traffic monitoring programs often use similar technology for vehicle detection, they may have different sensitivities to potential errors. Although small errors in traffic counts may not be critical in traffic operations, for which data are refreshed every few seconds, the same error levels could become much more significant in traffic monitoring, for which data, and the corresponding errors, are typically aggregated over longer periods of time. This study investigates if and how traffic volume data from detectors at signalized intersections could be appropriately used in a traffic monitoring program that supports the Highway Performance Monitoring System (HPMS). For this purpose, this study evaluates both the accuracy and representativeness of these traffic signal detector data in comparison to standard HPMS-type traffic count data obtained by portable pneumatic counters under different traffic flow conditions and intersection geometries. The results, although varying by site, indicate that these traffic signal detectors can produce 15-min aggregate traffic counts of comparable quality to results from portable pneumatic tube counters and provide 90% accuracy at a 95% level of confidence for 15-min aggregate counts. Through an examination of the representativeness of the data under different conditions, the study develops a set of eligibility criteria that can be used to identify intersections that are suitable for performance monitoring data collection.
AB - There are ongoing efforts to leverage the large volume of data collected to support real-time traffic operations for various non-real-time uses. However, adapting data streams to a different purpose can be challenging, especially if different applications use the data in different ways. Although traffic operations and traffic monitoring programs often use similar technology for vehicle detection, they may have different sensitivities to potential errors. Although small errors in traffic counts may not be critical in traffic operations, for which data are refreshed every few seconds, the same error levels could become much more significant in traffic monitoring, for which data, and the corresponding errors, are typically aggregated over longer periods of time. This study investigates if and how traffic volume data from detectors at signalized intersections could be appropriately used in a traffic monitoring program that supports the Highway Performance Monitoring System (HPMS). For this purpose, this study evaluates both the accuracy and representativeness of these traffic signal detector data in comparison to standard HPMS-type traffic count data obtained by portable pneumatic counters under different traffic flow conditions and intersection geometries. The results, although varying by site, indicate that these traffic signal detectors can produce 15-min aggregate traffic counts of comparable quality to results from portable pneumatic tube counters and provide 90% accuracy at a 95% level of confidence for 15-min aggregate counts. Through an examination of the representativeness of the data under different conditions, the study develops a set of eligibility criteria that can be used to identify intersections that are suitable for performance monitoring data collection.
UR - http://www.scopus.com/inward/record.url?scp=84976314692&partnerID=8YFLogxK
U2 - 10.3141/2593-08
DO - 10.3141/2593-08
M3 - Article
AN - SCOPUS:84976314692
SN - 0361-1981
VL - 2593
SP - 74
EP - 84
JO - Transportation Research Record
JF - Transportation Research Record
ER -