Systemic Pedestrian Safety Analysis (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

71 Background and Motivation To help complement their traditional hot spot analysis, California DOT (Caltrans) has devel- oped a quasi-systemic safety approach for identifying high risk crash locations. The approach, dubbed the Pedestrian Systemic Monitoring Approach for Road Traffic Safety (PEDSMARTS), focuses on developing strategies to reduce pedestrian and bicycle injuries along urban arterials. Caltrans sought to use this program as a means for more effectively incorporating pedestrian and bike projects into their safety funding, as these roadway users were typically underrepresented through traditional hot spot analyses due to lack of exposure. PEDSMARTS takes a holistic approach to safety by screening the entire state highway system (∼15,000 miles) and focusing on facility types of interest, rather than locations of interest. The systemic analyses matrix tool seeks to represent key roadway types and crash types in a single table to inform the decision-making process; this helps to provide a big-picture snapshot of the issues rather than one focused on a single area or facility type. Step 1: Define Study Scope The Caltrans process was initiated to complement the traditional hot spot analysis and focused on high risk locations along urban arterials. Step 2: Compile Data This approach used primarily crash data and basic roadway data, including presence of traffic signals, number of lanes, and AADTs. C H A P T E R 1 2 Case Example 4: California Department of Transportation Key Takeaways • Employs a matrix of facility types and crash types to identify high crash frequency scenarios across a roadway network. • Connects each high crash scenario (or matrix cell) to a potential countermeasure for further consideration.

72 Systemic Pedestrian Safety Analysis Step 3: Determine Risk Factors The process started by defining specific facility types of interest, in addition to crash types. Crash count data is used to identify facility types with high crash frequencies. The roadway characteristics were predetermined by the University of California, Berkeley team and Caltrans from prior analyses and included the following: • Signalized or unsignalized locations • Number of lanes for both main and cross streets • AADT for both main streets and cross streets Risk factors are inferred using a matrix tool that reflects the combinations of crash types and facility types that have the highest crash counts (see Figure 9). The matrix includes categories for location type across the horizontal axis and crash types across the vertical axis, in which cell (i, j) represents the number of crash type i experienced at locations of type j (e.g., in Cell 3, 3 in Figure 9, there were 98 Type 3 crashes at locations of Type 3). The matrix cell with the highest number of crashes represents a systemic hot spot type, and the corresponding location type and crash type are entered into a systemic countermeasures matrix (see Step 5). Step 4: Identify Potential Treatment Sites Each of the systemic hot spot cells is associated with a set number of sites. For example, the 98 crashes identified in the red cell in Figure 9 occurred at 51 different sites at Location Type 3. How- ever, all sites with the set of characteristics of Location Type 3 are potential candidates for treat- ment of this crash type. Additional work (in the field, using local knowledge or aerial imagery, or some combination of these) would be required to identify which subset of these sites, which include many locations with zero prior crashes, are priorities for treatment. Step 5: Select Potential Countermeasures Figure 10 shows an example systemic countermeasure matrix. This consists of counter- measures specifically identified to have the potential to reduce the crash type i for location type j in Caltrans’ countermeasure toolbox. The matrix identifies countermeasures 2 and 4 as the most 71 59 10 70 28 67 56 82 66 67 39 55 98 10 80 73 11 61 2 36 30 1 2 3 4 5 77 22 23 22 238 338 282 183 174 280 258 273 171 233 39 116 51 25 289 1 2 3 Location type number of sites Crashes on urban arterials Location type is based on features of the site. Example: Intersection; ADT<10,000; speed  45 mph; 3 or 4 lanes; traffic signal not present. Crash type is based on features of the crash. Example: Turning vehicle 258 pedestrian crashes across all Type 2 locations. 98 pedestrian crashes for Crash Type 3 and Location Type 3. 25 sites of Location Type 4 “Systemic hot spot” C ra sh t yp e 4 5 1215 Figure 9. Example of systemic hot spot identification matrix (Grembek et al. 2013).

Case Example 4: California Department of Transportation 73 applicable to address Crash Type 3 at locations of Type 3). Once a countermeasure is selected, safety practitioners can then implement the countermeasure across all locations of that type (e.g., across all 51 locations of Type 3) or across a subset (identified in Step 4). The methodology includes consideration of dozens of countermeasure options and lists countermeasures linked to specific pedestrian crash types. Some of the countermeasure options include the following: • Roadway lighting • Raised median island • Traffic calming treatments (e.g., curb extensions, mini-circles) • Countdown signals • Right turn on red restrictions • Advanced stop bar • Overpass/underpass • Pedestrian hybrid beacon • Crosswalk marking and enhancements • Transit stop improvements • Roundabouts • Curb ramps Other Steps and Lessons Learned to Date The inherent challenges with this approach lie in performing expert analyses to develop the crash type and location type matrices, and then in accounting for differences in pedestrian activity/exposure across different facility types as they pass through different area types during prioritization. Since the matrix does not take these types of data or pedestrian volume into account in the initial analysis, more work must go into Step 4 to assess the appropriateness of each site for treatment. However, this research has been ongoing since 2013, and Caltrans is continually seeking to modify and improve the procedures for their systemic program. This methodology has allowed Caltrans to more readily incorporate systemic projects into its annual safety programming and to develop a more comprehensive safety improvement plan. An article related to this methodology (Grembek et al. 2013) can be found at http://citeseerx.ist.psu.edu/ viewdoc/download?doi=10.1.1.448.85&rep=rep1&type=pdf. Location type is based on features of the site. Example: Intersection; ADT<10,000; speed  45 mph; 3 or 4 lanes; traffic signal not present. Crash type is based on features of the crash. Example: Turning vehicle Values in this table represent the possible countermeasures to reduce crash type i for location j. To reduce Type 3 crashes at Type 3 locations, apply countermeasures 2 or 4, across all of the Type 3 locations (n = 51) of the arterial (i.e., systemic). 1,2 3,4 1,2 2,6 7,8,9 2 3,5,6 4 6,7 4,7,9 1,3,4 3 2,4 6 4,5 4 5 1,2,3 2,3,6 5,7 2 1 2 3 4 5 5,6 3 3,6 5,7 1 2 3 Location type Possible countermeasure C ra sh t yp e 4 5 Figure 10. Example of systemic countermeasure matrix (Grembek et al. 2013).