User declared design distance vs. average measured

Questions and discussions about operating Tru-Traffic

Moderator: bullock

User declared design distance vs. average measured

Postby socaltrafficpadawan » Thu Nov 30, 2017 1:15 am

Hello Greg,

While processing some trip logs to create a travel time & delay report, I came across the following message in the attachment named "capture". By carefully analyzing all of the runs, I found out that two (2) of my runs are the runs that are triggering such warning message.

I am curious to know how to tell exactly what could be the reason behind these two runs giving me a warning message. To provide a little more background perspective to this situation, I conducted about 300+ runs with the same equipment and same network model and only two runs (sequential runs; one after the other) have triggered the message. Additionally, the weather conditions during these two runs were adequate (mostly cloudy, but a few other runs were collected that same day without any issues). I am assuming some GPS noise is the cause for these two runs to be reason for this issue.

Is there a way to correct the discrepancy in the distance between the two intersections shown in the warning message?
Capture.PNG (22.97 KiB) Viewed 489 times
Posts: 1
Joined: Thu Nov 16, 2017 10:19 pm

Re: User declared design distance vs. average measured

Postby bullock » Fri Dec 01, 2017 10:01 pm

That's a good question.

To find the underlying cause of the discrepancy, first open both the Network View and the GPS View to the Trip Logs tab.

Right-click on the Network View and choose Properties. In the section Show for the Trip Logs, check the boxes Trip Log Traces and Speed Color-coded Trip Log Points. Zoom the Network View as needed to resolve the link between Telegraph Canyon Rd. and Miracosta Cir. to, say, 50 feet or less.

In the GPS View, check the box for the Trip Logs with the two special runs. While watching the Network View, you may want to alternately check and uncheck the other runs to observe what's different about the two special runs.

If GPS noise is causing the discrepancy, then those two runs will show an erratic path along the link between Telegraph Canyon Rd. and Miracosta Cir., like you were drunk driving. If so, then you can disable the noisiest points by either
  • right-clicking on a point a choosing Trip Log: Disable Point from Trip Log, or
  • Alt-right-clicking on a point.

Another possible cause for the discrepancy is that the true travel distance along that link is longer than 855 feet.

If so, then use the Network View and GPS View to note the runs that correctly describe the true travel path, with no GPS noise and nicely resolved bends.

Open the Intersection Parameters for Miracosta Cir. Next to the box for the Distance from Telegraph Canyon Rd., click the button Get from Trip Logs.... Click the header button over the Travel distance column to group the runs in the list. The outliers will be at the ends of the lists, with one end showing any runs that under-resolve the bends, and the other end showing the noisy runs (with erratic paths). You can use click and Shift-click to select all the runs in a group, then check or uncheck a box in the selected group to include or exclude them as desired. You'll want to exclude the outliers and include only the clean runs.

If the link between Telegraph Canyon Rd. and Miracosta Cir. is a straight line, then check the run with the shortest distance and uncheck all the rest. Or equivalently, check all the runs and change the statistic in the drop-down list near the bottom to show "Minimum".

If the link between Telegraph Canyon Rd. and Miracosta Cir. has bends, then uncheck all the outliers, both those with erratic paths and those that under-resolve the bends. Check all the nice runs, especially the ones identified and correctly describing the true travel path along the link. Set the statistic in the drop-down list near the bottom to, save, "Average".

When you click OK, your declared travel distance should be correct.

If you want to either upload your file or email it to me, I'd be happy to look it over.

Site Admin
Posts: 209
Joined: Thu May 06, 2004 6:51 pm
Location: Pacific Grove, CA

Return to Program Operation

Who is online

Users browsing this forum: Bing [Bot] and 1 guest