A spatial editing and validation process for short count traffic data : final report - Page 41 |
Previous | 41 of 57 | Next |
|
small (250x250 max)
medium (500x500 max)
Large
Extra Large
large ( > 500x500)
Full Resolution
All (PDF)
|
This page
All
|
Hughes- Oliver, Heo, McDonald July 2006
A Spatial Editing and Validation Process for Short Count Traffic Data
— 41 —
Figure 12. Prediction interval flag map for Region 7, the test area. Flag colors indicate the following:
red— level 2 alert, blue— level 1 alert, yellow— no alert.
The segmentation approach, while computationally feasible, can create misleading results due to edge
effects. More specifically, after segmentation stations near the border of a region do not benefit from
information available in the adjoining region and this could impact either predictions or prediction
uncertainties. To investigate the degree of impact, we focused on two counties, Johnston and Wake,
placed them in a newly defined segment, and re- obtained their prediction intervals and flags. Of the 489
PTC stations in Johnston County, only two of them ( 0.4%) had changes in their flag values. Of the 1,531
PTC stations in Wake County, only 38 of them ( 2.5%) had changes in their flag values. Table 8 identifies
those stations for which there were changes. The small percentages of affected stations and the fact that
there were never any changes from a “ no alert” flag to a “ level 2 alert” flag suggested that edge effects
may be very small.
Table 8. PTC stations in Johnston ( 2 of 489) and Wake ( 38 of 1,531) Counties whose flag vlaues changed
as a consequence of being placed in a different region.
County
UNIQ_ I
D
Flag from
Region 7
Flag from
other region Difference
Johnston 500144 1 0 1
Johnston 500543 1 0 1
Wake 910616 2 1 1
Wake 1080316 2 1 1
Wake 1080588 2 1 1
Object Description
Description
| Title | A spatial editing and validation process for short count traffic data : final report - Page 41 |
| Pres Local File Path-M | \Preservation_content\StatePubs\pubs_borndigital\images_master\ |
| Full Text | Hughes- Oliver, Heo, McDonald July 2006 A Spatial Editing and Validation Process for Short Count Traffic Data — 41 — Figure 12. Prediction interval flag map for Region 7, the test area. Flag colors indicate the following: red— level 2 alert, blue— level 1 alert, yellow— no alert. The segmentation approach, while computationally feasible, can create misleading results due to edge effects. More specifically, after segmentation stations near the border of a region do not benefit from information available in the adjoining region and this could impact either predictions or prediction uncertainties. To investigate the degree of impact, we focused on two counties, Johnston and Wake, placed them in a newly defined segment, and re- obtained their prediction intervals and flags. Of the 489 PTC stations in Johnston County, only two of them ( 0.4%) had changes in their flag values. Of the 1,531 PTC stations in Wake County, only 38 of them ( 2.5%) had changes in their flag values. Table 8 identifies those stations for which there were changes. The small percentages of affected stations and the fact that there were never any changes from a “ no alert” flag to a “ level 2 alert” flag suggested that edge effects may be very small. Table 8. PTC stations in Johnston ( 2 of 489) and Wake ( 38 of 1,531) Counties whose flag vlaues changed as a consequence of being placed in a different region. County UNIQ_ I D Flag from Region 7 Flag from other region Difference Johnston 500144 1 0 1 Johnston 500543 1 0 1 Wake 910616 2 1 1 Wake 1080316 2 1 1 Wake 1080588 2 1 1 |
