Ramp counts for assignment validation

4 posts / 0 new
Last post
ewilke
Ramp counts for assignment validation

Hello - We have had some debate here recently about whether ramp counts should be included in assignment validation checks (%RMSE, flow:count ratios, etc.). The ramps are counted in this state regularly so we have trustworthy data to use, but we have heard of an old "rule of thumb" that ramp counts should not be used. We can't think of a specific reason why, however. After all - if we have good data, why not use it? 

Is anyone else familiar with this "rule of thumb", and what the reasoning behind it is? Is it outdated or still applicable today?

Also, I'd be interested in hearing if anyone has other examples of "rules of thumb" that may not be as relevant in today as they were in the past.

Thank you,

Eric Wilke, Iowa DOT

wwhite

Eric:

In my experience, every good piece of data you have should be used BUT interpretation is key. For example, a ramp count is also a turning movement count (of a sort). If you put in four ramp counts for each mainline count, the “rules” of what constitutes an acceptable region-wide RMSE may need to be revisited. Since ramps aren’t very long, if you’re using a macroscopic model, attempts to calibrate macro conditions to those micro counts also should be carefully executed. Even parcel-level loadings could have dramatic effects on any given ramp volume. However, by considering them as an additional performance indicator variable in the model validation process, the results can only get better.

One way to get an “apples-to-apples” use of ramp counts/TMCs in regional application is to use them for validation reviews but not present them in the regional validation performance summary results. In that way, they help fine-tune your macro-efforts efforts without distorting (i.e. overweighing) the expected relationship between limited access mainline counts, used in standard macroscopic validation criteria, and arterial/collector roadway counts.

All of that said, ramp/TMC counts should be used for subarea and corridor-level validations to the greatest extent possible. In summary, you should always use good data but always be equally careful in interpreting/applying the results. If you’re using “old” validation criteria to judge performance, the same applies. A good RMSE or VOL/CNT can be a positive indicator, but those indicators do not necessarily mean the model is suitable for any specific purpose.

Regards,
Wade

From: eric.wilke=iowadot.us@mg.tmip.org [mailto:eric.wilke=iowadot.us@mg.tmip.org] On Behalf Of ewilke
Sent: Thursday, February 15, 2018 3:08 PM
To: TMIP
Subject: [TMIP] Ramp counts for assignment validation

Hello - We have had some debate here recently about whether ramp counts should be included in assignment validation checks (%RMSE, flow:count ratios, etc.). The ramps are counted in this state regularly so we have trustworthy data to use, but we have heard of an old "rule of thumb" that ramp counts should not be used. We can't think of a specific reason why, however. After all - if we have good data, why not use it?

Is anyone else familiar with this "rule of thumb", and what the reasoning behind it is? Is it outdated or still applicable today?

Also, I'd be interested in hearing if anyone has other examples of "rules of thumb" that may not be as relevant in today as they were in the past.

Thank you,

Eric Wilke, Iowa DOT
--
Full post: https://tmip.org/content/ramp-counts-assignment-validation
Manage my subscriptions: https://tmip.org/mailinglist
Stop emails for this post: https://tmip.org/mailinglist/unsubscribe/12606

balpern2

Eric,

My take on this is: if you have a good, comprehensive set of highway ramp
counts - include them in your validation and evaluate them as a separate
facility type. Many models tend to substantially over-estimate ramp volumes
for a variety of reasons, even if the highway mainline volumes validate
well (basically, too many trips assigned via the highway for short
distances, and not enough for longer distances) but if you don't compare
the assigned ramp volumes vs. counts, you might have no other way to know
this. If your model has this problem, there are various ways to correct it
(each with their advantages and disadvantages), but I'm not going to get
into that now.

Bernie Alpern
Transportation Planner
Eyal Highway and Traffic Engineers, Ltd.
Jerusalem, Israel

On Thu, Feb 15, 2018 at 10:07 PM, ewilke wrote:

> Hello - We have had some debate here recently about whether ramp counts
> should be included in assignment validation checks (%RMSE, flow:count
> ratios, etc.). The ramps are counted in this state regularly so we have
> trustworthy data to use, but we have heard of an old "rule of thumb" that
> ramp counts should not be used. We can't think of a specific reason why,
> however. After all - if we have good data, why not use it?
>
> Is anyone else familiar with this "rule of thumb", and what the reasoning
> behind it is? Is it outdated or still applicable today?
>
> Also, I'd be interested in hearing if anyone has other examples of "rules
> of thumb" that may not be as relevant in today as they were in the past.
>
> Thank you,
>
> Eric Wilke, Iowa DOT
> --
> Full post: https://tmip.org/content/ramp-counts-assignment-validation
> Manage my subscriptions: https://tmip.org/mailinglist
> Stop emails for this post: https://tmip.org/mailinglist/unsubscribe/12606
>

Guy Rousseau

Hi Eric, ramp geometry is something else to take into consideration during assignment validation. For instance, when we looked at the NPMRDS data a few years ago to develop new free-flow speed lookup tables, we noticed some differences in free-flow speeds, which resulted in speeds that did not match the observed data very well. As such, several steps were undertaken in order to update the assumed model free-flow speeds, including the identification of loop ramps, flagged as a network attribute (a simple binary code, so 0 or 1, with 1 indicating a loop ramp, denoting a particular pre-set free-flow speed). Our regional travel demand model network features 3 streamlined types of ramps, system-to-system ramps, exit ramps and entrance ramps (back in the days we had on-ramps, off-ramps, low-speed ramps, medium-speed ramps and high-speed ramps). In our regional model, ramps as a distinct facility type also warrant their own VDF curve parameters. So anyhow, with all that being said, if you have processed your ramp counts via a thorough QA/QC process with random ground counts field verifications, you should be ok to leverage those for assignment validation, also feel free to consider NPMRDS data as an additional data source (as well as observed VMT), as a way to further inform your assignment validation procedure. Seasonal adjustment factors applied to the ramp counts might be something else to look into (ADT vs AADT), same for count classification (trucks, # of axles, etc.).

Guy Rousseau
Travel Surveys & Transportation Model Development Manager
Atlanta Regional Commission
229 Peachtree St NE, Suite 100
Atlanta, Georgia 30303
P | 470.378.1565
M| 678.986.4344
grousseau@atlantaregional.com
atlantaregional.com
________________________
CONFIDENTIALITY NOTICE: This e-mail and any attachments are intended solely for the use of the named recipient or recipients. Any dissemination of this e-mail by anyone other than an intended recipient is strictly prohibited. If you are not a named recipient, you are prohibited from any further viewing of the e-mail or any attachments or from making any use of the e-mail or attachments. If you believe you have received this e-mail in error, notify the sender immediately and permanently delete the e-mail and any attachments, and all copies.

From: eric.wilke=iowadot.us@mg.tmip.org [mailto:eric.wilke=iowadot.us@mg.tmip.org] On Behalf Of ewilke
Sent: Thursday, February 15, 2018 3:08 PM
To: TMIP
Subject: [TMIP] Ramp counts for assignment validation

Hello - We have had some debate here recently about whether ramp counts should be included in assignment validation checks (%RMSE, flow:count ratios, etc.). The ramps are counted in this state regularly so we have trustworthy data to use, but we have heard of an old "rule of thumb" that ramp counts should not be used. We can't think of a specific reason why, however. After all - if we have good data, why not use it?

Is anyone else familiar with this "rule of thumb", and what the reasoning behind it is? Is it outdated or still applicable today?

Also, I'd be interested in hearing if anyone has other examples of "rules of thumb" that may not be as relevant in today as they were in the past.

Thank you,

Eric Wilke, Iowa DOT
--
Full post: https://tmip.org/content/ramp-counts-assignment-validation
Manage my subscriptions: https://tmip.org/mailinglist
Stop emails for this post: https://tmip.org/mailinglist/unsubscribe/12606