摘要: |
Each state in the United States can be viewed as a universe of road segments. For each road segment in each state, it is desired to know various traffic characteristics based on count data, classification count data, and weigh-in-motion data. These data are absolutely essential for highway design, maintenance, safety, and planning. Given no cost constraints, each road segment would be continuously monitored every day of the year. However, in practice, a few road segments are monitored continuously every day of the year to produce annual characteristics of traffic flow. The remaining road segments are monitored for one or two days each year, and this resulting data are `adjusted` (using factors based on data collected from the continuously monitored road segments) to produce estimates of annual characteristics. With this general approach, each state strives to provide estimates of annual characteristics for each road segment within its jurisdiction. In 1985, the Federal Highway Administration (FHWA) published the Traffic Monitoring Guide to assist states in achieving this end. As with almost any data collection effort, the monitoring data suffers from errors from many sources. In this paper, we report some empirical findings in a research project sponsored by the FHWA. This research project studied the variability in the traffic data from the continuously monitored road segments from state(s) and, the extent to which this variability is transferred to and affects the precision of the data produced from the road segments which are monitored only one or two days each year. The ultimate hope is that states will eventually be able to not only publish an estimate of a characteristic such as Average Annual Daily Traffic (AADT) for each road segment, but also that each estimate will be accompanied by a statement expressing how good the estimate is in terms of its estimated variability or precision, which will likely be expressed as a coefficient of variation. |