Вход на сайт
Просмотр новости

Найдите то, что Вас интересует

Prioritizing Traffic Light Penalties with Traffic Counts
Дата публикации: 16-09-2026 15:38:18


Impetus
Many routers, such as OSRM,1 set penalties for barriers like traffic lights or at-grade crossings, to estimate time lost at these points. For intersections with traffic lights, the standard procedure is to put a tag of highway=traffic_light at the intersecting node. For divided highway intersections, common practice is to tag each intersecting node also.
This quickly adds up to excess penalty for the simple router.2 On a cross-traffic turn (left turns in the United States), this will lead the router to count three penalties per intersection. But, time spent at these intersections is not near three times the normal intersection, especially considering a cross-traffic turn often has first priority and protected access along divided highways.
The more-detailed method of tagging removes the unnecessary penalties. Instead of tagging highway=traffic_light at every intersection, a node is placed at every stop line to the intersection.
Now, in every road-legal path, the route will pass through only one traffic light node, with the following tags:
OSM
highway = traffic_light
traffic_light = signal
traffic_light:direction = forward/backward
The directionality tag is not necessary on one-way segments, but it is better to keep explicit and consistent with two-way segments.
Although this tagging is ideal, it would be intensive for mappers to correct every intersection like this. An intersection of two undivided highways, for instance, need not be updated, since the node graph already adds only one traffic light penalty. But any intersection with at least one divided highway has a route that will pass through multiple intersecting nodes.
To prioritize the routes most in-need of review, I searched for intersections likely to accumulate the most delay from these penalties. The first method analyzed high-volume roadways, and the second analyzed intersections with the most accidents.
Highway traffic volume
My first method was to look at stretches of highway with the most annual average daily traffic (AADT). Due to being high volume, these roads are likely to be divided, and are also guaranteed to have vehicles passing through key intersections (resulting in lots of total excess penalty).
I downloaded the 2026 data on AADT from IDOT’s webpage.3 I removed all motorways without traffic lights and filtered the data for segments in Chicagoland with an AADT over 40,000 vehicles per day (sorted by distance from downtown), returning the following short list:
Cicero Ave (Ogden–87th),
Harlem (Stevenson–50th),
First Ave (Ogden–Stevenson),
North Ave (Mannheim–1st),
Mannheim Ave (I-194–Butterfield),
87th St (Roberts–Central),
La Grange Rd (143rd–191st),
Roosevelt Rd (Meyers–I 294),
Kingery Ave (Touhy–Stevenson),
Butterfield Rd (Veterans Tlwy–22nd),
North Ave (Gary–Kingery),
and a few small sections of road between Elgin and Addison)
I reviewed every intersection on these segments for divided highways, and tagged appropriately.
Intersection crashes
The second method was to analyze which intersections them receive the most crashes per year. By itself, an intersection with many crashes indicates that drivers already have some difficulty in navigating the roadway, so the extra detail is doubly appropriate.
IDOT also maintains a database of vehicle crash data by year.4 Based on the 2025 data, I found 70,287 intersection-related crashes in Chicagoland, out of 300,000 total crashes in Illinois. I clustered the individual crashes to a radius of 400 feet (the average length of a block in Chicago). Below is a table with the most total crashes, as well as their neighborhood.


Cluster name
Crashes in 2025
Neighborhood




Ohio Street off-ramp
224
Near North Side, Chicago


Congress Parkway off-ramp
157
Loop, Chicago


Wacker/Michigan Ave
141
Loop, Chicago


79th/Stony Island/South Chicago
127
South Shore, Chicago


Ohio/Michigan
108
Near North Side, Chicago


Ohio/Dearborn
100
Near North Side, Chicago


111th/Harlem
77
Worth, IL


Pershing/Western
72
McKinley Park, Chicago


95th/Stony Island
68
South Deering, Chicago


Stevenson Expwy/Cicero
60
Garfield Ridge, Chicago


I 57/Halsted
54
Washington Heights, Chicago


Madison/Cicero
54
Austin, Chicago


Ogden/Lincoln
52
Montgomery, IL


Golf Rd/Wolf Rd (Cumberland Circle)
58
Des Plaines, IL


The cluster radius of 400 feet turned out to be too large to analyze individual intersections, but it was still useful for generating high-crash areas.
Ohio Street off-ramp (Kennedy–LSD)
Michigan Ave (LSD–Balbo)
Ida B. Wells Dr
Roosevelt (Halsted–LSD)
Stony Island Ave (71st–Bishop Ford)
All of these stretches were reviewed for special traffic lights via CookOrtho2025 imagery, licensed from NearMaps.



Project OSRM, OSRM Backend, “penalty.feature.” ↩


I could not find a router which clusters close-together traffic lights into one. Either way, explicit tagging is better for all routing purposes, at the cost of being less clear for network maps. ↩


Illinois Department of Transportation, Annual Average Daily Traffic, 2026. ↩


Illinois Department of Transportation, Crashes - 2025. ↩





Основное содержимое страницы с новостью.

Impetus

Many routers, such as OSRM,1 set penalties for barriers like traffic lights or at-grade crossings, to estimate time lost at these points. For intersections with traffic lights, the standard procedure is to put a tag of highway=traffic_light at the intersecting node. For divided highway intersections, common practice is to tag each intersecting node also.

Example of excess traffic light penalty

This quickly adds up to excess penalty for the simple router.2 On a cross-traffic turn (left turns in the United States), this will lead the router to count three penalties per intersection. But, time spent at these intersections is not near three times the normal intersection, especially considering a cross-traffic turn often has first priority and protected access along divided highways.

The more-detailed method of tagging removes the unnecessary penalties. Instead of tagging highway=traffic_light at every intersection, a node is placed at every stop line to the intersection.

Now, in every road-legal path, the route will pass through only one traffic light node, with the following tags:

OSM highway = traffic_light traffic_light = signal traffic_light:direction = forward/backward

The directionality tag is not necessary on one-way segments, but it is better to keep explicit and consistent with two-way segments.

Although this tagging is ideal, it would be intensive for mappers to correct every intersection like this. An intersection of two undivided highways, for instance, need not be updated, since the node graph already adds only one traffic light penalty. But any intersection with at least one divided highway has a route that will pass through multiple intersecting nodes.

To prioritize the routes most in-need of review, I searched for intersections likely to accumulate the most delay from these penalties. The first method analyzed high-volume roadways, and the second analyzed intersections with the most accidents.

Highway traffic volume

My first method was to look at stretches of highway with the most annual average daily traffic (AADT). Due to being high volume, these roads are likely to be divided, and are also guaranteed to have vehicles passing through key intersections (resulting in lots of total excess penalty).

I downloaded the 2026 data on AADT from IDOT’s webpage.3 I removed all motorways without traffic lights and filtered the data for segments in Chicagoland with an AADT over 40,000 vehicles per day (sorted by distance from downtown), returning the following short list:

  • Cicero Ave (Ogden–87th),
  • Harlem (Stevenson–50th),
  • First Ave (Ogden–Stevenson),
  • North Ave (Mannheim–1st),
  • Mannheim Ave (I-194–Butterfield),
  • 87th St (Roberts–Central),
  • La Grange Rd (143rd–191st),
  • Roosevelt Rd (Meyers–I 294),
  • Kingery Ave (Touhy–Stevenson),
  • Butterfield Rd (Veterans Tlwy–22nd),
  • North Ave (Gary–Kingery),
  • and a few small sections of road between Elgin and Addison)

I reviewed every intersection on these segments for divided highways, and tagged appropriately.

Intersection crashes

The second method was to analyze which intersections them receive the most crashes per year. By itself, an intersection with many crashes indicates that drivers already have some difficulty in navigating the roadway, so the extra detail is doubly appropriate.

IDOT also maintains a database of vehicle crash data by year.4 Based on the 2025 data, I found 70,287 intersection-related crashes in Chicagoland, out of 300,000 total crashes in Illinois. I clustered the individual crashes to a radius of 400 feet (the average length of a block in Chicago). Below is a table with the most total crashes, as well as their neighborhood.

Cluster name Crashes in 2025 Neighborhood
Ohio Street off-ramp 224 Near North Side, Chicago
Congress Parkway off-ramp 157 Loop, Chicago
Wacker/Michigan Ave 141 Loop, Chicago
79th/Stony Island/South Chicago 127 South Shore, Chicago
Ohio/Michigan 108 Near North Side, Chicago
Ohio/Dearborn 100 Near North Side, Chicago
111th/Harlem 77 Worth, IL
Pershing/Western 72 McKinley Park, Chicago
95th/Stony Island 68 South Deering, Chicago
Stevenson Expwy/Cicero 60 Garfield Ridge, Chicago
I 57/Halsted 54 Washington Heights, Chicago
Madison/Cicero 54 Austin, Chicago
Ogden/Lincoln 52 Montgomery, IL
Golf Rd/Wolf Rd (Cumberland Circle) 58 Des Plaines, IL

Downtown clusters of crash data.

The cluster radius of 400 feet turned out to be too large to analyze individual intersections, but it was still useful for generating high-crash areas.

  • Ohio Street off-ramp (Kennedy–LSD)
  • Michigan Ave (LSD–Balbo)
  • Ida B. Wells Dr
  • Roosevelt (Halsted–LSD)
  • Stony Island Ave (71st–Bishop Ford)

All of these stretches were reviewed for special traffic lights via CookOrtho2025 imagery, licensed from NearMaps.

  1. Project OSRM, OSRM Backend, “penalty.feature.” ↩

  2. I could not find a router which clusters close-together traffic lights into one. Either way, explicit tagging is better for all routing purposes, at the cost of being less clear for network maps. ↩

  3. Illinois Department of Transportation, Annual Average Daily Traffic, 2026. ↩

  4. Illinois Department of Transportation, Crashes - 2025. ↩

Схожие новости

#Наименование новостиТональностьИнформативностьДата публикации
1What 689 cycling routes look like when you measure them against OpenStreetMap07.4716-09-2026
2What 689 cycling routes look like when you measure them against OpenStreetMap07.4716-09-2026
3Tile server for toll roads, cycling restrictions and cobblestone overlays0529-06-2026
4Fehlende smootness-Daten in OSM013.3510-09-2026
5Overpass Ultra06.0805-08-2026
6La communauté mondiale OpenStreetMap s’est réunie à Paris, nous y étions !!010.415-09-2026
7About OSM for Cities0729-06-2026
8В России ускорили моделирование дорожного трафика в десятки раз2602-07-2026
9Stauprognose: Weiterhin oft Stillstand auf den Autobahnen0506-07-2026
10Stauprognose: Volle Autobahnen in alle Richtungen01003-08-2026

Классификация: . Схожих патентов: 0. Схожих новостей: 10. Тональность: 0. Информативность: 7.7. Источник: www.openstreetmap.org.