Annual Average Daily Traffic (AADT) is important information for various transportation research areas such as travel model calibration and validation, pavement design, roadway design, air quality compliance, and so forth. Specifically for model calibration and validation in a long-range transportation planning, a base year model requires quite a number of count locations across the study region. Sometimes, those count locations are not fully available on time or the base year model requires more counts at unobserved locations, mostly on non-freeway facilities. To provide better prediction of AADT for unobserved locations on non-freeway facilities, this study considers a spatial dependency because, in general, the traffic volume at one monitoring station is correlated with the volumes at its neighboring stations. The spatial regression model takes into account both spatial trend (mean), and spatial correlation, which is modeled by one of the geostatistical approaches called kriging. In this study, the spatial regression model is applied to AADT in Wake County, North Carolina. From the result, it is found that the overall predictive capability of the spatial regression model is much better than that of ordinary regression model. In addition, the urban area has more reliable prediction than the rural area. Finally, we expect that the spatial regression model will provide better predictions for unobserved locations on non-freeway facilities where AADT counts are needed for a given analysis, but where budget restricts the number of field counts that can be collected.
Samenvatting