Application of Crash Modification Factors for Access Management, Volume 2: Research Overview (2021)

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.

56 Introduction Chapter 2 documented the results of a literature review, and Chapter 3 documented the results of an assessment of existing data sources. This chapter identifies areas where additional research is needed and is feasible. The purpose of the gap analysis is to focus the research effort on leveraging quality existing information and building on past research to fill priority voids and/or limitations in the current state of the knowledge. For example, there is an opportunity to make use of datasets used to develop SPFs and CMFs in the Highway Safety Manual (1st Edition) (AASHTO 2010) and supplement those datasets with additional variables to incorporate access management strategies and overcome data limitations from past studies. Summary of Gap Analysis and Possible Data Sources The gap analysis follows the structure of the literature review and focuses on the three primary levels of implementation (i.e., site, intersection, and corridor). The intent is to identify the avail- ability of quality CMFs (i.e., those CMFs rated four or five stars in the CMF Clearinghouse) for each strategy and implementation level. If SPFs are available that could be used to infer CMFs (by dividing the “with” SPF prediction by the “without” SPF prediction), this is also noted. If SPFs are available, but they are not calibrated using data from the same jurisdiction and time period, the difference between predictions could be due to factors other than those related to the site charac- teristics. This makes it challenging to infer CMFs from SPFs and is noted as such in the SPF-related summary tables in Chapter 2. Starting with the information in Chapter 2, the project team reviewed the high-priority strate- gies (based on the survey described in Chapter 3) to identify those where quality CMFs are not available and where estimating new CMFs may be feasible. Table 49 presents these priority strate- gies, indicating where CMF information is lacking and identifying potential data sources to sup- port CMF or SPF development. Table 50 presents strategies for which knowledge gaps exist and future research may be needed but which were not identified as high-priority strategies from the survey. Those data sources shown in boldface in the tables were the focus of the data collection effort (Task 7). The other potential data sources were considered. Prioritization of NCHRP Project 17-74 Research The previous section presents a summary of current knowledge on the safety effects of access management strategies as well as knowledge gaps and potential priority research topics. In a discussion of these results, the NCHRP Project 17-74 panel indicated a preference for focusing on enhancing the current Highway Safety Manual (1st Edition) Part C Predictive Method to include access management features. This included the calibration of existing SPFs to different C H A P T E R 4 Gap Analysis

Strategy Substrategy CMF Availability Missing CMFs Possible Data Sources Manage location, spacing, and design of median openings and crossovers Create directional median opening No CMFs at 3 or more stars CMFs for corridor and site levels. DTFH61-09-C-00026 Safety Evaluation of Access Management Policies and Techniques (includes data from California, Minnesota, and North Carolina) Brigham Young University Project: Safety Analysis of Access Management Implementation in Utah Regulate median opening density 5 CMFs at 3 stars but none at 4 or 5. CMFs are for corridor level. CMFs for corridor and site levels. Regulate median opening spacing 2 CMFs at 3 stars for site level. CMFs for corridor and site levels. Manage location and spacing of unsignalized access Establish spacing for unsignalized access CMFs at 3 stars for site and corridor levels exist. CMFs for corridor, intersection, and site levels. DTFH61-09-C-00026 Safety Evaluation of Access Management Policies and Techniques (includes data from California, Minnesota, and North Carolina) NCHRP Project 17-62 (includes data from Ohio, Minnesota, and North Carolina) Iowa State Project: Development of a Comprehensive Access Management Policy and Guidelines Brigham Young University Project: Safety Analysis of Access Management Implementation in Utah Install non- traversable medians, and accommodate left turns and U-turns Convert traversable median (non-TWLTL) to non-traversable median 2 CMFs at 3 stars exist for intersection level. 2 CMFs at 4 stars exist for corridor and site levels. CMFs for intersection level and improved CMFs for corridor and site levels. DTFH61-09-C-00026 Safety Evaluation of Access Management Policies and Techniques (includes data from California, Minnesota, and North Carolina) NCHRP Project 17-62 (includes data from Ohio, Minnesota, and North Carolina) Brigham Young University Project: Safety Analysis of Access Management Implementation in Utah Replace TWLTL with non-traversable median CMFs rated 3 stars exist at site and corridor levels. Higher rated CMFs for corridor, intersection, and site levels. DTFH61-09-C-00026 Safety Evaluation of Access Management Policies and Techniques (includes data from California, Minnesota, and North Carolina) NCHRP Project 17-62 (includes data from Ohio, Minnesota, and North Carolina) Brigham Young University Project: Safety Analysis of Access Management Implementation in Utah Manage the spacing of signalized and unsignalized access on crossroads in the vicinity of freeway interchanges Establish spacing criteria for interchange ramp terminals 3 CMFs rated 3 stars available at site level. CMFs for corridor, intersection, and site levels. DTFH61-09-C-00026 Safety Evaluation of Access Management Policies and Techniques (includes data from California, Minnesota, and North Carolina) Table 49. Gaps identified as high priority.

Strategy Substrategy CMF Availability Missing CMFs Possible Data Sources Convert two-way to one-way Convert two-way operation to one -way operation 8 CMFs rated 3 stars for site level. CMFs for corridor, intersection, and site levels. DTFH61-09-C-00026 Safety Evaluation of Access Management Policies and Techniques (includes data from California, Minnesota, and North Carolina) NCHRP 17-58 “Safety Prediction Models for Six-Lane and One-Way Urban and Suburban Arterials” Establish corner clearance criteria Driveways at signalized intersections 15 CMFs rated 3 stars for intersection level. CMFs for corridor and intersection levels. DTFH61-09-C-00026 Safety Evaluation of Access Management Policies and Techniques (includes data from California, Minnesota, and North Carolina) Safety Effects of Corner Clearance at Signalized Intersections. FHWA DCMF Project (includes data from California and North Carolina) NCHRP Project 17-62 (includes data from Ohio, Minnesota, and North Carolina) Install traversable medians Convert undivided to divided by traversable median 1 CMF rated 3 stars for both site and corridor levels. CMFs for corridor, intersection, and site levels. Manage spacing of traffic signals Establish traffic signal density criteria 3 CMFs rated 3 stars for both site and corridor levels. CMFs for corridor and site levels. DTFH61-09-C-00026 Safety Evaluation of Access Management Policies and Techniques (includes data from California, Minnesota, and North Carolina) NCHRP Project 17-62 (includes data from Ohio, Minnesota, and North Carolina) Establish traffic signal spacing criteria 7 CMFs rated 3 stars for both site and corridor levels. CMFs for corridor and site levels. Right-turn treatment Channelize right-turn lane 3 CMFs rated 1 star for intersection level. CMFs for intersection and site levels. Not identified Not identified Table 50. Gaps not identified as high priority.

Gap Analysis 59   access management conditions, developing adjustment factors for existing SPFs to capture the effects of access management strategies, and developing CMFs for specific access management strategies. This section presents opportunities to fill primary gaps in CMFs and SPFs related to access management. Specifically, this section identifies opportunities to develop CMFs that are com- patible with and complementary to the Part C Predictive Method in the Highway Safety Manual (1st Edition), including 1. A summary of which access management features are directly or indirectly addressed in the Highway Safety Manual (1st Edition) Predictive Method. 2. A summary of the priority access management features that are candidates for CMF develop- ment in this project (based on a discussion between the NCHRP Project 17-74 research team and project panel members). Summary of Access Management Features Directly or Indirectly Addressed in the Highway Safety Manual (1st Edition) The Part C Predictive Method in the Highway Safety Manual (1st Edition) was reviewed to identify which access management features are addressed directly or indirectly through SPFs and CMFs. The list of access management features is the same list that was used to categorize these features for the literature review and gap analysis. Table 51 lists the access management features and indicates which are accounted for in the Part C Predictive Method. A check mark indicates that the Highway Safety Manual (1st Edition) provides a CMF for that strategy. “Indirect—Use of Multiple SPFs Possible” means that the High- way Safety Manual (1st Edition) has SPFs for the alternate site characteristics. These SPFs may be used to infer a CMF provided that each of the SPFs can first be successfully calibrated to the same jurisdiction of interest, and the sites to which the CMF would apply are indeed represented by the SPFs in question in terms of site characteristics and traffic volumes. The shaded cells in Table 51 indicate the highly ranked access management strategies, with rankings based on the survey of practitioners. Priorities included median type, corner clearance, access density, access spacing, turn lanes, and turn restrictions. There are limited features currently accounted for in the SPFs and CMFs within the Highway Safety Manual (1st Edition) and even fewer are explicitly provided. While analysts can infer the effect of some strategies from SPFs, caution should be exer- cised in doing so because differences in crash predictions from two SPFs representing two levels of a feature could be due to factors, other than the feature of interest, which are not accounted for in the SPFs. CMFs that are directly or indirectly accounted for in the Highway Safety Manual (1st Edition) are briefly described in the following. Illustrative examples of inferring CMFs are provided for two features: (1) install non-traversable median on undivided highway and (2) establish density for unsignalized access (e.g., maximum driveway density). In both cases, CMFs are provided in Part D. These can be compared to CMFs that could be inferred from the SPFs provided in Part C. Convert a Four-Legged Intersection to Two Three-Legged Intersections The effect on safety of converting a single four-legged intersection to two three-legged inter- sections can be inferred from SPFs. Chapters 10, 11, and 12 of the Highway Safety Manual (1st Edition) have SPFs for four-legged and three-legged intersections for both signalized and unsignalized control. Chapters 10 and 11 provide SPFs for four-legged intersections for both signalized and unsignalized control and SPFs for three-legged intersections for unsignalized con- trol. Chapter 12 provides SPFs for four-legged and three-legged intersections for both signalized and unsignalized control.

Strategy Substrategy Chapter 10 (Rural 2- lane) Chapter 11 (Rural Multi lane) Chapter 12 (Urban/Suburban arterials) Chapter 18 Freeways Chapter 19 Ramps Alternative intersection and interchange design Convert a 4-legged intersection to two 3-legged intersections ü Indirect – Use of Multiple SPFs Possible No SPF for 3 -leg signal ü Indirect – Use of Multiple SPFs Possible No SPF for 3 -leg signal ü Indirect – Use of Multiple SPFs Possible -- Install roundabout at roadway intersection ü Indirect – Use of Multiple SPFs Possible Updates from NCHRP 17-70 have SPFs ü Indirect – Use of Multiple SPFs Possible Updates from NCHRP 17-70 have SPFs ü Indirect – Use of Multiple SPFs Possible Updates from NCHRP 17-70 have SPFs ü Indirect – Use of Multiple SPFs Possible Updates from NCHRP 17-70 have SPFs Provide median acceleration lane -- -- -- -- Replace direct left turn with grade-separated interchange ü Indirect – Use of Multiple SPFs Possible SPFs for interchange elements available in new freeway chapter material ü Indirect – Use of Multiple SPFs Possible SPFs for interchange elements available in new freeway chapter material ü Indirect – Use of Multiple SPFs Possible SPFs for interchange elements available in new freeway chapter material ü Indirect – Use of Multiple SPFs Possible NCHRP 17-45 has SPFs Superstreet (RCUT, J -Turn) -- -- -- -- Jug handle intersections -- -- -- -- Install continuous green T intersection at signalized 3 - legged intersection -- -- -- -- Control driveway design elements Change class/type of driveway -- -- ü Indirect – Use of Multiple SPFs Possible SPFs in current Highway Safety Manual (1st Edition). NCHRP 17-62 recommends driveway density though -- Change movement restriction -- -- -- -- Require design of driveways with the appropriate return radii, throat width, channelization, number of lanes, and throat length for the type of traffic to be served -- -- -- -- Control intersection design elements Control turning radius -- -- -- -- Convert two-way to one- way Convert two-way operation to one-way operation -- -- ü Indirect – Use of Multiple SPFs Possible SPFs may be available from NCHRP 17 -58 -- Establish corner clearance criteria Driveways at signalized intersections -- -- -- -- Table 51. Summary of access management strategies in Highway Safety Manual (1st Edition).

(continued on next page) Improve cross connectivity Allow vehicles to access adjacent properties without returning to the mainline -- -- -- -- Install continuous TWLTL on undivided highway Non-road diet scenarios ü Direct – CMF or SPF adjustment factor available -- ü Indirect – Use of Multiple SPFs Possible SPFs available -- Road diet scenarios -- -- ü Indirect – Use of Multiple SPFs Possible SPFs available -- Install non-traversable medians, and accommodate left turns and U-turns Convert traversable median (non-TWLTL) to non - traversable median -- -- -- -- Install isolated median barriers -- -- -- -- Install non -traversable median on undivided highway -- ü Indirect – Use of Multiple SPFs Possible SPFs available ü Indirect – Use of Multiple SPFs Possible SPFs available -- Replace TWLTL with non - traversable median -- -- ü Indirect – Use of Multiple SPFs Possible SPFs available -- Install service or frontage roads Change proportion of primary roadway with frontage road -- -- -- -- Install frontage road to provide access to individual parcels -- -- -- -- Install traversable medians Convert undivided to divided by traversable median -- -- -- -- Left-turn treatment Change storage capacity of existing left-turn deceleration lane -- -- -- -- Channelize left -turn lane -- -- -- -- Control/improve design elements of left -turn lanes -- -- -- -- Install left -turn deceleration lanes at roadway intersections -- -- -- -- Prohibit left turn -- -- -- -- Manage location and spacing of unsignalized access Establish density for unsignalized access (e.g., maximum driveway density) ü Direct – CMF or SPF adjustment factor available -- ü Indirect – Use of Multiple SPFs Possible SPFs available in NCHRP Web-Only Document 295 ( Ivan et al. 2021) -- Establish spacing for unsignalized access (e.g., driveways) -- -- -- --

Manage spacing of traffic signals Establish traffic signal density criteria -- -- -- -- Establish traffic signal spacing criteria -- -- -- -- Create directional median opening -- -- -- -- Install U-turns as an alternative to direct left turns -- -- -- -- Regulate median opening density -- -- -- -- Regulate median opening spacing -- -- -- -- Replace full median opening with median designed for left turns from the major roadway -- -- -- -- Manage the spacing of signalized and unsignalized access on crossroads in the vicinity of freeway interchanges Establish spacing criteria for interchange ramp terminals -- -- -- -- Provide adequate sight distance at access points Manage design elements to improve sight distance -- -- -- -- Manage the location and placement of parking (e.g., replace curb parking with off- street parking or restrict on- street parking near driveways or intersections to improve sight distance) -- -- -- -- Manage vegetation to improve sight distance (e.g., in landscaped medians or sight triangles) -- -- -- -- Right-turn treatment Channelize right-turn lane -- -- -- -- Control/improve design elements of right-turn lanes -- -- -- -- Install right-turn deceleration lane at roadway intersections -- -- -- -- ü indicates a CMF is available, -- indicates no CMF is available. median openings and crossovers Manage the location, spacing, and design of Strategy Substrategy Chapter 10 (Rural 2- lane) Chapter 11 (Rural Multi lane) Chapter 12 (Urban/Suburban arterials) Chapter 18 Freeways Chapter 19 Ramps Table 51. (Continued).

Gap Analysis 63   Install Roundabout at Roadway Intersection The effect on safety of converting an intersection to a roundabout can be inferred from the SPFs available. Roundabout SPFs have been developed in NCHRP Project 17-70 and will be included in the Highway Safety Manual (2nd Edition). As mentioned previously, SPFs are available for almost all intersection types that may exist prior to roundabout conversion. Replace Direct Left Turn with Grade-Separated Interchange The effect on safety of converting a direct left turn to an interchange can be inferred from the SPFs available. SPFs for freeways and interchanges have been developed in NCHRP Project 17-45 and will be included in the Highway Safety Manual (2nd Edition). Change Class/Type of Driveway Chapter 12 of the Highway Safety Manual (1st Edition) predicts crashes by type of driveway. CMFs for changing the class/type of driveway can be inferred from the SPFs provided in the Highway Safety Manual (1st Edition). Convert Two-Way Operation to One-Way Operation NCHRP Project 17-58 is developing SPFs for one-way roads that could be used with SPFs from Chapter 12 of the Highway Safety Manual (1st Edition) to infer a CMF for converting to one-way operation. Install Continuous Two-Way Left-Turn Lane on Undivided Highway Chapter 10 of the Highway Safety Manual (1st Edition) includes a CMF for this treatment. In Chapter 12, there are SPFs for two-lane roads and four-lane roads as well as SPFs for three- or five- lane roads including a TWLTL. A CMF can be inferred by comparing the prediction for adding a TWLTL without taking away through-traffic lanes to the prediction for not adding the TWLTL. Install Non-Traversable Median on Undivided Highway SPFs in Chapters 11 and 12 of the Highway Safety Manual (1st Edition) can be used to infer a CMF. For example, Chapter 12 provides SPFs for multivehicle driveway and non-driveway crashes per mile, considering the annual average daily traffic (AADT), the number and type of driveways, and whether the arterial has a median or not. Table 52 presents inferred CMFs for providing a median on urban and suburban arterials, where the CMFs depend on AADT, drive- way density, and driveway type. In principle, these CMFs may be preferred to the CMFs that are explicitly provided in Part D that do not account for changes in AADT, driveway density, and driveway type. AADT Base Model (no driveways) Major Commercial Major Residential 40/mile 20/mile 40/mile 20/mile 5,000 0.635 0.456 0.635 0.623 0.847 10,000 0.648 0.471 0.657 0.645 0.871 15,000 0.656 0.481 0.671 0.659 0.884 20,000 0.662 0.488 0.681 0.668 0.894 25,000 0.666 0.494 0.689 0.676 0.901 30,000 0.670 0.499 0.695 0.682 0.907 35,000 0.673 0.503 0.701 0.687 0.911 40,000 0.676 0.506 0.705 0.692 0.915 Table 52. CMFs for providing a median based on the Highway Safety Manual (1st Edition) Part C Predictive Method for urban and suburban four-lane undivided and divided arterials.

64 Application of Crash Modification Factors for Access Management Replace Two-Way Left-Turn Lane with Non-Traversable Median SPFs in Chapter 12 of the Highway Safety Manual (1st Edition) can be used to infer a CMF by comparing the prediction for a four-lane road with a TWLTL (5T) to the prediction for a four-lane divided road. Establish Density for Unsignalized Access (e.g., maximum driveway density) Chapter 10 of the Highway Safety Manual (1st Edition) provides a CMF and Chapter 12 pro- vides SPFs that can be used to infer a CMF. For example, Table 53 presents CMFs inferred from the Chapter 12 SPFs. As before, SPFs are provided for multivehicle driveway and non-driveway crashes per mile, considering the AADT, the number and types of driveways, and whether or not the arterial is divided. Thus, the inferred CMFs for changing driveway density also depend on these factors. In principle, these CMFs are preferred to the Part D CMFs, at least within the range of the data used to estimate the SPFs from which they are inferred because the impact on crashes is disaggregated by AADT, the number and type of driveways, and whether the arterial is divided. Summary of Priority Access Management Features with Potential for CMF Development Table 54 identifies the priority access management features for NCHRP Project 17-74 based on the literature review, the survey of practitioners, and discussion with the project panel. For these features, there were also no CMFs rated four or five stars in the CMF Clearinghouse at the time of the research project. AADT Reduce Driveways from 40 to 30/mile Reduce Driveways from 20 to 10/mile Major commercial Major residential Major commercial Major residential 4U 4D 4U 4D 4U 4D 4U 4D 5,000 0.817 0.886 0.853 0.922 0.712 0.895 0.791 0.937 10,000 0.823 0.922 0.859 0.950 0.725 0.908 0.804 0.945 15,000 0.826 0.927 0.863 0.954 0.733 0.914 0.812 0.949 20,000 0.829 0.930 0.866 0.956 0.739 0.919 0.817 0.952 25,000 0.830 0.933 0.868 0.958 0.743 0.922 0.821 0.954 30,000 0.832 0.935 0.870 0.959 0.747 0.925 0.825 0.956 35,000 0.833 0.936 0.872 0.961 0.750 0.927 0.827 0.957 40,000 0.835 0.938 0.873 0.962 0.753 0.929 0.830 0.958 4U = 4-lane undivided roadway segment, 4D = 4-lane divided roadway segment. Table 53. Driveway density CMFs inferred from the Highway Safety Manual (1st Edition) Part C Predictive Method for multivehicle crashes on urban four-lane undivided and divided arterials. Table 54. High-priority strategies. Strategy Substrategy CMF Availability Manage location, spacing, and design of median openings and crossovers Create directional median opening No CMFs rated 3 or more stars Regulate median opening density 5 CMFs rated 3 stars. CMFs are for corridor level. Regulate median opening spacing 2 CMFs rated 3 stars for site level. Manage location and spacing of unsignalized access Establish spacing for unsignalized access 13 CMFs rated 3 stars. CMFs are for corridor and site levels. Manage the spacing of signalized and unsignalized access on crossroads in the vicinity of freeway interchanges Establish spacing criteria for interchange ramp terminals 3 CMFs rated 3 stars available at site level. Establish corner clearance criteria Driveways at signalized and unsignalized intersections 15 CMFs rated 3 stars for intersection level. Manage spacing of traffic signals Establish traffic signal density criteria 3 CMFs rated 3 stars for both site and corridor levels. Establish traffic signal spacing criteria 7 CMFs rated 3 stars for both site and corridor levels. Right-turn treatment Channelize right-turn lane 3 CMFs rated 1 star for intersection level.