Star Sign up now to gain access to all the features of this website. NEW — access is now open to ALL USERS! — Already registered? Please log in.

Questions on Pedestrian LOS methods at signals

Eric Royer
California Dept. of Transportation
Questions on Pedestrian LOS methods at signals was posted on June 19th, 2018 at 7:38 PM EDT
regarding Chapter 19: Signalized Intersections

I have some questions regarding the Pedestrian methods in chapter 19, mostly around intersections with Right turn islands. Sidewalk Width and Corner Radius; are these dimensions provided at the location of the right turn islands or at the outside edge of the intersection? Crosswalk Width; is a right turn island with pedestrian pass through considered a constraint on the width of a crosswalk? Crosswalk Length; is the dimension of this distance through a right turn island included, or only the length exposed to traffic? Number of lanes (Nd); does this number include lanes of traffic between the right turn island and the sidewalk? How are right turning volumes handled when right turn islands are present? Vrtor and Vrt in equations 19-63 and 19-76? Are these volumes zeroed out or still included? It would be useful if the example problem provided in Chapter 31 included these factors. I'm surprised to see what a large impact dimensions of the intersection have on LOS. I'm analyzing the intersection of SR 89 and US 50 in south lake tahoe (the south lake tahoe "wye") which is fairly large, but the hour I'm analyzing has only about 80 pedestrians per hour in total among all 4 legs and seems to work out to LOS C, but the example in chapter 31 has thousands of pedestrians per hour and is LOS B, and it seems to stem from only two lanes being crossed and wide crosswalks.

Thanks for any clarification that can be provided.

Eric Royer Caltrans District 3 Traffic Ops/Rural Highway Ops

Paul Ryus
Kittelson & Associates, Inc.
RE: Questions on Pedestrian LOS methods at signals was posted on June 23rd, 2018 at 10:57 AM EDT

Sidewalk Width and Corner Radius; are these dimensions provided at the location of the right turn islands or at the outside edge of the intersection?

Sidewalk width and corner radius are used to determine the area available for storing pedestrians on an intersection corner. In the case of this intersection, with channelizing islands with landscaping and areas with raised concrete not intended for pedestrian use, I would use the total walkway area within the island in lieu of the (W_a W_b - 0.215R2) component of Equation 19-51. On the outside edges of the intersection, you would need to make some assumptions about pedestrian holding area for the three corners where the sidewalks are not curb-tight.

Paul Ryus
Kittelson & Associates, Inc.
RE: Questions on Pedestrian LOS methods at signals was posted on June 23rd, 2018 at 10:59 AM EDT

Crosswalk Width; is a right turn island with pedestrian pass through considered a constraint on the width of a crosswalk?

Yes. Crosswalk width is based on the effective width. A reduced-width cut in the right-turn island at the end of the crosswalk would be treated the same as the reduced-width median cut given as an example in the text in the middle of page 19-75.

Paul Ryus
Kittelson & Associates, Inc.
RE: Questions on Pedestrian LOS methods at signals was posted on June 23rd, 2018 at 11:00 AM EDT

Crosswalk Length; is the dimension of this distance through a right turn island included, or only the length exposed to traffic?

Only the curb-to-curb length of a signalized crossing. Right-turn islands would not be included in the length, but a pass-through across a raised median would be included in the length. The variable is used for evaluating the crosswalk circulation area, which only applies to signalized crossings; it would not be used for the unsignalized crossings between the street corners and the right-turn islands.

Paul Ryus
Kittelson & Associates, Inc.
RE: Questions on Pedestrian LOS methods at signals was posted on June 23rd, 2018 at 11:01 AM EDT

Number of lanes (Nd); does this number include lanes of traffic between the right turn island and the sidewalk?

The HCM isn’t clear on this point and I had to go back to the original research, documented in NCHRP Report 616 and NCHRP Web-only Document 128 to find an answer, along with comparing the intersection characteristics used to develop the ped LOS regression equation to what exists in the field (for details, see my post farther down in this thread). The number of lanes is the total number of traffic lanes being crossed to get from one side of the street to the other, including any channelized right-turn lanes.

(Edited to consider the characteristics of the intersections used in the original research.)

(Edited August 23rd, 2018 at 12:10 PM EDT)
Paul Ryus
Kittelson & Associates, Inc.
RE: Questions on Pedestrian LOS methods at signals was posted on June 23rd, 2018 at 11:03 AM EDT

How are right turning volumes handled when right turn islands are present? Vrtor and Vrt in equations 19-63 and 19-76? Are these volumes zeroed out or still included?

Right-turn volumes are always included.

Paul Ryus
Kittelson & Associates, Inc.
RE: Questions on Pedestrian LOS methods at signals was posted on June 23rd, 2018 at 11:04 AM EDT

I'm surprised to see what a large impact dimensions of the intersection have on LOS.

Exposure to traffic is what drives pedestrian intersection LOS at signalized intersections. Note that if all you want is LOS as a result, you don’t need to perform Steps 1 and 2 (corner and crosswalk circulation area) of the procedure. In other words, crosswalk width and corner radius aren’t used in determining LOS. Crosswalk length for the purposes of determining LOS is represented by the number of traffic lanes crossed. Steps 1 and 2 are used primarily for sizing crosswalks and pedestrian holding areas.

Eric Royer
California Dept. of Transportation
RE: Questions on Pedestrian LOS methods at signals was posted on June 24th, 2018 at 8:16 PM EDT

Paul, Thank you very much for your insight. I'll have to make some adjustments to my spreadsheet, I really appreciate your help, on the weekend no less.

Eric

Eric Royer
California Dept. of Transportation
RE: Questions on Pedestrian LOS methods at signals was posted on August 7th, 2018 at 3:00 PM EDT

Paul, regarding Nd it still doesn't seem clear to me. in 616 there is no clarification that I can find and in 128 the section where it does mention through traffic lanes on page 30 is for a multimodal analysis which includes segment analysis.it doesn't seem to make sense to me, when analyzing pedestrian LOS at an intersection, to exclude a left or right turn lane, since they all result in an increased crossing distance. Would a T-intersection have no through lanes on the intersecting road? What about a 4 way intersection, where there is a large number of left turns and one of the lanes becomes a "trap" lane for the left turn?

Regards, Eric Royer

Paul Ryus
Kittelson & Associates, Inc.
RE: Questions on Pedestrian LOS methods at signals was posted on August 8th, 2018 at 12:05 PM EDT

Eric, I based my earlier answer on Web-only Document 128. Exhibit 12 lists the required data for all the multimodal methods, including the pedestrian method. These data include "cross-street through lanes at intersections" and "presence of right-turn channelization islands at intersections", but nothing related to the number of cross-street turn lanes. Similarly, on page 32, the description of pedestrian intersection LOS describes the turning volumes needed to compute pedestrian LOS as additional required data, but not turning lanes.

Nevertheless, I've now gone one step farther and compared the characteristics of some of the intersections used in the original research (listed in Exhibit 43 in NCHRP Report 616) to what exists in the field. For example, the south side of the intersection of Fowler Ave. & 56th St. in Tampa, FL is described in the exhibit as crossing nine lanes. According to Google Maps, the crossing at the actual intersection crosses 3 SB through lanes, 3 NB left-turn lanes, 2 NB through lanes, and 1 (non-channelized) NB right-turn lane, for a total of 9 lanes. This indicates to me that the regression equation for ped LOS was developed using the total number of lanes crossed and not the number of through lanes as suggested in Web-only Document 128. At another intersection in Tampa, Tampa Bay Drive at Dale Mabry Highway, the exhibit indicates the east crossing crosses 7 lanes. In the field, the east crossing would involve crossing a channelized NB right-turn lane, 2 EB through lanes, 1 WB left-turn lane, 2 WB through lanes, and 1 channelized WB right-turn lane, for a total of 7 lanes. This indicates to me that the channelized right-turn lanes should be counted as part of the number of lanes crossed.

To sum up, Nd should be the total number of travel lanes crossed when crossing the street, including any channelized right-turn lanes. Finally, note that according to pages 24-25 of Web-only Document 128, the method was not developed for multiple-lane free right-turns, nor for high-speed free right-turns (such as onto a freeway ramp). I've edited my answer above to avoid future confusion.

Paul

Eric Royer
California Dept. of Transportation
RE: Questions on Pedestrian LOS methods at signals was posted on August 8th, 2018 at 12:49 PM EDT

Thanks again Paul, for going the extra mile. Much appreciated. makes much more sense now.

Sign in to add a reply. Don't have an account? No problem, sign up for free.