1 post / 0 new
Xinbo Mi

It’s been a few days since the question was asked and there hasn’t been any discussion other than Sam’s response. Now I would like to add some of my thoughts.

I’ve been speaking up for modeling at small/medium-sized MPOs recently. Models are different between those from large metro areas and small ones. They have different emphases and answer different kinds of questions, or were asked to answer those questions even if not supposed to. This question Sharif brought up is a perfect example (probably not that perfect because this one does not have a good answer lol). Surprisingly (or unsurprisingly I would rather say), it did not initiate an active discussion from experts in our community who usually do modeling projects on large metro areas and we don’t care about these kinds of questions in large metro areas’ models, I assume.

Back to the question itself, I got these TWLTL questions too. I agree with Sam that perhaps there is no good answer to it, at least at macroscopic level. Per Highway Capacity Manual, “There is no formal methodology for evaluating the traffic operational effectiveness of a TWLTL on a two-lane highway”, let alone our travel demand models. The added capacity or reduced delay heavily depends on the proportion of left-turn vehicles and the volume of opposing traffic. Actually, I shouldn’t be using “added capacity” but instead it is really the “reduced capacity” for not having the TWLTL, comparing to the modeled capacity computed by [# of lanes] x [lane capacity]. For your example from 4-lane to 5-lane, adding the TWLTL only makes full use of the capacity (except for extreme circumstances) of those 4 thru lanes. Then you may ask, can I reduce my base scenario’s 4-lane capacity? I personally think yes you can but you need to be very careful because that will break the link consistency/homogeneity of your model. Given the fact that your model is well built with appropriate capacity constraint at intersections for arterials and with good sensitivity, you can try the approach above and you may end up with 2 situations: (1) the output difference is minimum, which has an easy answer; (2) the output has noticeable difference, and you should then analyze where the marginal traffic comes from/is diverted to. Then you will need to bring those affected links into your scope, take a look at their volume and capacity, make adjustment and engineering judgement, and rerun the model if necessary. However, I do not recommend doing this, unless you really-really have to for a specific project, and the only-only tool available is a macroscopic travel demand model. My official answer is still there is no good answer, or we shouldn’t deliberately address this issue in a travel demand model and expect the model giving us an accurate answer.

Now back to more spotlight on small/medium-sized MPOs, though I said this question is a good question but not with a good answer, I would still encourage folks answering these questions, even just by telling people there is no good answer. This actually benefits small/medium-sized MPOs a lot by making people (either technical or non-technical managers) understand what questions a travel demand model is supposed to answer and what is not. Don’t be surprised that our community still has so many folks not knowing or unsure about similar questions and similar answers. There are no silly questions nor silly answers. Please ask them and answer them. I truly hope we can embrace everyone into our community.

Btw lastly to add, one change you can definitely make for TWLTL in the model is to switch the link from undivided highway to divided highway, which is what I was doing, but it has minimal impact to the capacity as expected. :-D

Xinbo Mi
Senior Transportation Engineer
Evansville Metropolitan Planning Organization
1NW Martin Luther King Jr. Blvd. Room 316
Evansville, Indiana 47708
Office Main Line 812.436.7833
[new logo_smaller]

From: On Behalf Of Sam Granato
Sent: Thursday, January 19, 2023 6:33 AM
Subject: [TMIP] Re: [TMIP] TWLTL

**** This is an EXTERNAL email. Exercise caution. DO NOT open attachments or click links from unknown senders or unexpected email. ****

I don’t think there’s a good answer for that, and definitely not a one size fits all. Do you have any locations along a proposed TWLTL location where on average traffic in the through lanes gets backed up more than it does at the next downstream signal or stop sign?

Sent from my iPhone

> On Jan 18, 2023, at 3:08 PM, sharifullah wrote:
> Hi All,
> I am looking for any recommendations on how TWLTL impacts capacity of the through lanes. I want to add a TWLTL in a 4-lane segment and making it as a 5-lane segment. Any thoughts on how I can increase the capacity of thru lanes to address any capacity increase due to addition of the TWLTL?
> Thanks!
> Sharif
> --
> Full post:
> Manage my subscriptions:
> Stop emails for this post:
Full post:
Manage my subscriptions:
Stop emails for this post: