2 HIN Intro
There are many established ways to examine crashes to better understand traffic safety patterns. Hotspot analyses have long been used to address high crash locations by retrospectively identifying the greatest concentrations of reported crashes over a determined period of time. Hotspot analysis is a valuable method to visualize locations with historic crash issues, but it is less effective at identifying locations with latent crash risk factors. In this way, it can be described as reactive. Additionally, hotspots may be less effective for analyzing bicyclist safety if crash frequencies are low due to geographic sparsity, which can exacerbate issues related to regression to the mean. Conversely, a systemic analysis is effective for identifying roadways with risk factors for crashes, independent from their crash history. For example, a wide arterial with a 45-mph posted speed limit, high traffic volumes, no bike facility, and few trip-attracting land uses may not have any reported bike crashes. However, the roadway and operational characteristics of that arterial are associated with higher bicycle crash risk. The absence of crashes is therefore not a reflection of low crash risk, but a reflection of lack of exposure that hotspot analyses cannot adequately convey. Systemic analysis is largely proactive; it allows planners and engineers to find locations that may warrant safety improvements before crashes have occurred there.
High injury networks strike a balance between entirely retrospective and entirely proactive methods. Using spatial patterns of crash history, a High Injury Network identifies areas on the road network where crashes have been concentrated in sequence. A stretch of arterial roadway with crashes occurring at every other intersection might not show up on a traditional hotspot analysis because no one location has multiple crashes happening in the same place. However, the pattern of crashes all along the corridor suggests a larger safety issue. Further, the entire corridor likely shares similar characteristics that could be addressed systemically – even the intersections along the corridor that have not yet had crashes.
This section describes the development of a statewide High Injury Network and the results of the related High Injury Network analysis. The High Injury Network was built from a standard sliding windows analysis, which measures severity-weighted crash density by mode along segments on the network.
| name | install_date | uninstall_date | obs_duration | modes |
|---|---|---|---|---|
| 2nd Ave PBL | 4/23/16 | 4 years 11 mons 30 days 23:00:00 | Bike | |
| 39th Ave NE Greenway | 1/1/14 | 5/31/18 | 0 years 4 mons 29 days 23:00:00 | Bike |
| BGT | 1/1/14 | 4 years 11 mons 30 days 23:00:00 | Bike and Ped | |
| Broadway PBL | 1/1/14 | 10/30/21 | 3 years 9 mons 29 days 23:00:00 | Bike |
| Chief Sealth | 1/1/14 | 10/30/21 | 1 year 4 mons 29 days 23:00:00 | Bike |
| Elliot Bay | 5/11/13 | 12/11/22 | 4 years 11 mons 10 days 23:00:00 | Bike and Ped |
| Fremont Bridge | 10/8/12 | 4 years 11 mons 30 days 23:00:00 | Bike | |
| MTS or I-90 Trail | 12/18/13 | 12/31/21 | 3 years 11 mons 30 days 23:00:00 | Bike and Ped |
| NW 58th St | 1/1/14 | 7/31/22 | 4 years 6 mons 30 days 23:00:00 | Bike |
| Spokane Bridge | 11/24/13 | 4 years 11 mons 30 days 23:00:00 | Bike |