Guidelines for Treatments to Mitigate Opposite Direction Crashes (2022)

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.

20 Reduce Likelihood of Head-On Crashes As noted in Chapter 3, the preferred way to eliminate opposite direction crashes is to find ways to help keep vehicles in their travel lanes. If a driver is not fully alert and inadvertently strays from the travel lane, an increased separation between opposing directions of travel can help to reduce the likelihood that an opposite direction crash will occur. Table 11 summarizes countermeasures that can be deployed to help achieve additional physical separation between opposing lanes. The cost of treatments tends to require some infrastructure enhancements, and so these countermeasures are generally more expensive to implement. C H A P T E R 4 Countermeasure Project Cost Crash Type Facility Type Install centerline buffer area Moderate Head-on, opposite direction sideswipe Two-lane undivided highway Widen median High Head-on & opposite direction sideswipe (in median) Divided highway Install alternating / periodic passing lanes High Head-on, opposite direction sideswipe, & same direction sideswipe Two-lane undivided highway Install SafetyEdgeSM treatment Low Head-on, opposite direction sideswipe, SVROR Two-lane, multilane, and freeway Pave or widen shoulders High Head-on, opposite direction sideswipe, SVROR Two-lane, multilane, and freeway Table 11. Countermeasures to reduce likelihood of head-on crash. Install Centerline Buffer Area Treatment/Countermeasure: Centerline Buffer Area Project Type/Cost: Crash Type: Head-on and opposite direction sideswipe Facility Type/Characteristics: Two-lane undivided highways Low Moderate High WHAT (Introduction) The MUTCD (11) indicates that centerline pavement marking for two-way, two-lane high- ways should consist of passing and no passing zone markings, and that a single solid yellow

Reduce Likelihood of Head-On Crashes 21   line shall not be used to represent the centerline marking. In some instances, transportation agencies have observed operational and safety benefits by providing a narrow centerline buffer area that is separated by longitudinal pavement markings (Figure 8). This additional buffer area shifts the lateral placement of opposing direction vehicles. WHY (Safety State of the Practice) The use of a centerline buffer area introduces a greater physical separation between approach- ing vehicles. As shown in Figure 8, this configuration can accommodate passing zones while also maintaining a physical buffer between the active opposing lanes of travel. This treatment is largely untested; however, a recent analysis evaluated the centerline buffer area scenario for Texas highways (13). The study evaluated two-lane and four-lane rural highway locations with and without centerline buffer strips present. This research determined that the safety benefits decline and are not statistically significant for four-lane highway locations. For these multilane highways, the buffer area had little noticeable effect for SVROR crashes or opposite direction crashes. For the two-lane highways, however, the centerline buffer area can be used to help reduce the number of opposite direction crashes. As the centerline buffer width increases, the percentage of opposite direction crashes also decreases (see Figure 9). General Observations As shown in Figure 9, the number of opposite direction crashes decreases as the centerline buffer area width increases. As an example, the CMF for a 2 ft wide buffer is approximately 0.65 (or a 35 percent reduction in opposite direction crashes). Similarly, a buffer median width of 10 ft is equivalent to a CMF value of 0.1 (or a 90 percent reduction). WHERE (Application Issues) Design The construction of centerline buffer areas can be accommodated at locations where pavement widths are wide enough to accommodate the narrow buffer strip without other- wise compromising traffic operations. To accomplish this, the road should consist of paved shoulders, lanes widths that are (ideally) 12 ft in width, and a narrow buffer area. Construc- tion of the centerline buffer area may occur as part of a widening or resurfacing project. If significant widening is required, the cost of implementing this treatment is likely to increase significantly. (Photograph provided by Texas A&M Transportation Institute) Figure 8. Centerline buffer area.

22 Guidelines for Treatments to Mitigate Opposite Direction Crashes Operations The construction of a centerline buffer area can potentially result in an increase in vehicle speeds by shifting the lateral position of opposing vehicles. This operational effect may occur during daytime or nighttime conditions. The increase in vehicle speeds, however, can be associ- ated with an increase in SVROR crashes, though this increase does not appear to be statistically significant. Widen Median Treatment/Countermeasure: Widen Median Project Type/Cost: Crash Type: Median crashes (head-on, opposite direction sideswipe) Facility Type/Characteristics: Divided highway Low Moderate High WHAT (Introduction) Divided highways, by definition, include medians to physically separate opposing directions of traffic. Medians can take multiple forms, but generally fall into one of three categories: raised (e.g., curbed median island), depressed (e.g., grass/turf median with drainage), or flush (e.g., pavement markings only) (Figure 10). When the required ROW is available, widening a median increases the distance between opposing directions of traffic. This increased median width can reduce the likelihood of opposite direction crashes. WHY (Safety State of the Practice) Studies on the effect of median width have shown that increasing width reduces cross- median crashes, but the amount of reduction varies across studies. Key findings are summarized in Table 12. Study sites are separated into full access control at urban and rural locations. The preferred method for studying crash effects and developing crash modification factors is to conduct a before–after study in which the treatment date is known, and the changes in Figure 9. Centerline buffer area CMF for opposite direction crashes (2-lane rural).

Reduce Likelihood of Head-On Crashes 23   (Photograph by Texas A&M Transportation Institute) Figure 10. Wide median treatment. Study Location(s) Year Road Type Change in Median Width Percent Reduction in Crashes (%) Source Crash Severity Cross-Median Crash Type From (ft) To (ft) All (KABCO) All (KABC) 4-lane (KABCO) ≥ 5-lane (KABCO) Urban Locations – Full Access Control NR 2008 NR 10 20 NR NR 11 11 (23) 30 20 21 40 29 29 50 36 37 60 43 44 70 49 50 80 54 55 90 59 60 100 64 65 FL 2016 Arterial NA NA a = -0.0048* a = -0.0051* NR NR (24) 10 20 5 5 30 9 10 40 13 14 50 17 18 60 21 23 Rural Locations – Full Access Control NR 2008 NR 10 20 NR NR 14 NR (23) 30 26 40 37 50 46 60 54 70 60 80 66 90 71 100 75 CA, NC, OH, PA, WA 2014 Freeway NR +1 ft NR NR 2 2 (25) NR – not reported *Values represent the “a” coefficients in the following equation: CRF = 100(1- e[a x (MW – Base Width)] Where: MW = Proposed median width; Base Width = Existing median width. Note: The project selection process included projects with three-star ratings or more at the CMF Clearinghouse. ) Table 12. Studies that evaluated wider medians at access controlled segments.

24 Guidelines for Treatments to Mitigate Opposite Direction Crashes crashes before and after this date can be tracked; however, widening a median is a treatment that is not always “installed” in a manner that allows for a before–after study. It is unlikely that the median width on a highway will ever be changed without making other significant changes to the geometric cross section (23). In this case, the fact that there is a significant change other than the change in median width makes it more difficult to isolate the effects of the change in width in a before–after evaluation. General Observations Users can expect to see a reduction in cross-median crashes of 11 percent or more (CMF = 0.89 or lower). As median width increases, total crashes are associated with a reduction of at least 4 percent in total crashes, at least 3 percent in fatal and major injury crashes, and at least 2 percent in all other types of injury crashes. It should be noted that the studies listed in Table 12 generally found that wider medians led to fewer crashes, but there was a decreasing effect with incremental increases in median width; that is, widening a median from 10 ft to 30 ft typically had more of an effect than further widening a 30 ft median to 40 ft. WHERE (Application Issues) Design Widening a median requires sufficient ROW to accommodate the wider cross section of the highway, unless travel lanes are removed as part of the redesign of the highway. Median widening can occur in conjunction with adding travel lanes in one or both directions. In addi- tion to the construction cost of the highway, acquisition of ROW is typically a substantial cost in such projects. Environmental concerns (e.g., drainage, wildlife migration, etc.) may also present themselves during the design process. Operations Locations with wider medians may experience increased vehicle speeds. At nighttime, the additional lateral separation may reduce headlight glare from approaching vehicles. Install Alternating/Periodic Passing Lanes Treatment/Countermeasure: Install Alternating/Periodic Passing Lanes Project Type/Cost: Crash Type: Segment-only crashes (head-on, opposite direction sideswipe, same direction sideswipe) Facility Type/Characteristics: Two-lane undivided highway Low Moderate High WHAT (Introduction) Periodic passing lanes are lanes added to one or both directions of a two-lane undivided highway to provide opportunities for passing of slower vehicles and the dispersal of traffic platoons (Figure 11). These passing lane configurations, also referred to as Super 2 corridors, can be implemented on existing two-lane roadways where there is a significant amount of slow-moving traffic, there is limited sight distance for passing, and/or the existing traffic volume has exceeded the two-lane highway capacity, creating the need for vehicles to pass on a more frequent basis. Passing lanes can help reduce crashes associated with passing on two-lane roads, such as head-on and sideswipe crashes.

Reduce Likelihood of Head-On Crashes 25   WHY (Safety State of the Practice) Several studies documented reductions in crashes with the installation of passing lanes on two-lane highways. Key findings are summarized in Table 13. Information specific to the opposite direction crash types, however, is not explicitly available. General Observations Users can expect to see a reduction ranging from 33 up to 42 percent (CMF = 0.58 to CMF = 0.67) in total crashes and a reduction of 29 to 42 percent (CMF = 0.58 to CMF = 0.71) in fatal and injury crashes. A 35 percent reduction in segment-only crashes can be expected. This is equivalent to a CMF equal to 0.65. WHERE (Application Issues) Design The designer should consider the existing width of ROW, terrain, and structures to evaluate the feasibility of a Super 2 corridor and determine the best locations to install passing lanes with a minimum of ROW requirements, grading, and structure widening. It is preferable to avoid locating high-traffic intersections and driveways within the boundaries of a passing lane. Where passing lanes are terminated, the designer should confirm that the sight distance is suitable and that any conflicts with oncoming traffic can be avoided. The design should also accommodate constraints such as guard rail, guard fences, or narrow bridges (29). (Photograph by Texas A&M Transportation Institute) Figure 11. Alternating passing lanes. Study Location(s) Percent Reduction in Crashes (%) Source Crash Type / Location Crash Severity Segment-Only Crashes (KABCO) Total Crashes (KABCO) Fatal and Injury Crashes (KABC) MI 2012 33 29* (26) TX 2012 NR 42 (27) WY 2016 42 NR (28) NR – not reported *Data set did not contain any fatal crashes. Note: The project selection process included projects with three-star ratings or more at the CMF Clearinghouse. NR 35 NR Year Table 13. Summary of passing lanes studies for rural two-lane highways.

26 Guidelines for Treatments to Mitigate Opposite Direction Crashes Operations Where terrain, available budget, and other considerations allow, the addition of another passing lane is preferred over adding length to an existing passing lane. Incremental benefits are minimal for passing lane lengths over two miles. Install SafetyEdgeSM Treatment Treatment/Countermeasure: Install SafetyEdgeSM treatment Project Type/Cost: Crash Type: Head-on, opposite direction sideswipe, and SVROR Facility Type/Characteristics: Two-lane highways, multilane highways, and freeways Low Moderate High WHAT (Introduction) SafetyEdgeSM is a pavement edge treatment designed to mitigate roadway departure crashes (Figure 12). Rather than a vertical face with a potentially steep drop-off, the SafetyEdgeSM treatment shapes the edge of the pavement to a 30-degree slope, providing a surface more conducive to drivers correcting their paths and re-entering the roadway safely. The treatment involves the use of a specially designed but commercially available “shoe” device attached to the paver. Asphalt is extruded under the shoe, resulting in an edge with the desired shape that also provides more durability to resist raveling. Designed primarily to prevent roadway departure overcorrection crashes on roadways with asphalt surfaces, the SafetyEdgeSM may be installed on any paved road. The SafetyEdgeSM can be installed on all road types including two-lane highways, multilane highways, and freeways. For divided highways, a SafetyEdgeSM can be installed on the right (outside) shoulder and the left (median) shoulder to facilitate a vehicle’s return to the travel lanes. WHY (Safety State of the Practice) SafetyEdgeSM treatments have been studied in a variety of settings. Several pilot projects focused on lessons learned from the installation procedure, but more recent studies have also focused on safety performance. Common target crashes include opposite direction crashes; however, in some studies this specific crash type is included in the total crash or ROR category. Key findings for several of these studies are summarized in Table 14. General Observations Users can expect to see reductions of approximately 19 percent (CMF = 0.81) for total head- on and sideswipe crashes. Fatal and injury crash reductions range from 6 to 16 percent (CMF = 0.84 to 0.94) following the installation of a SafetyEdgeSM treatment at rural two-lane highways. This is equivalent to an average crash reduction of 11 percent (CMF = 0.89). WHERE (Application Issues) Design Ideally, the pavement edge should be thick enough so that there is adequate pavement depth to form the angled surface unique to the SafetyEdgeSM treatment. As with conventional paving, the graded material adjacent to the SafetyEdgeSM should be brought flush with the top of the pavement following paving, and this activity should be scheduled as part of a regular main- tenance effort. The SafetyEdgeSM concept is that when drop-offs recur, they will not be vertical,

Reduce Likelihood of Head-On Crashes 27   and the pavement edge will retain a shape that will not induce tire scrubbing. The SafetyEdgeSM is often included as part of projects to rebuild or rehabilitate existing paved shoulders in need of maintenance, but this treatment can also be used at locations where there is currently little or no paved shoulder width and the responsible agency plans to pave or overlay the road. Operations SafetyEdgeSM is particularly applicable for roadways during resurfacing, even if they have been resurfaced multiple times and have uneven edges of pavement. This treatment enables drivers traveling at highway speeds to more easily return to the travel lane after one or more tires drop off the paved surface. Pave or Widen Shoulders Treatment/Countermeasure: Pave or Widen Shoulder Project Type/Cost: Crash Type: Head-on, opposite direction sideswipe, SVROR Facility Type/Characteristics: Two-lane highways, multilane highways, and freeways Low Moderate High Study Location(s) Percent Reduction in Crashes (%) Source Crash Type Crash Severity All Crash Types (KABCO) ROR Crashes (KABCO) Head-On and Sideswipe (KABCO) All Crash Types (KABC) ROR Crashes (KABC) GA, IN 2011 9 14 NR 17 23 (30) 2011 7 5 NR 6 20 (30) IA 2017 13** 12 NR 16** 8 (31) KS 2017 NR NR NR NR 35 (31) FL, IA, NC, OH, PA 2017 1 21 19 11 NR (31) NR - represents crash reductions that were not reported or did not meet the three-star filter criteria. * Values based on unpaved shoulder locations only. Unless otherwise indicated, the study assessed paved shoulder locations. ** Iowa crash types were defined as non-intersection crashes only. Note: The selection process included studies with three-star ratings or more at the CMF Clearinghouse. For multistate studies, the selected values extend across more than one state and typically have larger sample sizes. GA, IN* Year Table 14. Summary of SafetyEdgeSM studies for two-lane segments. (Photograph provided by Texas A&M Transportation Institute) Figure 12. SafetyEdgeSM prior to grading the earthen shoulder.

28 Guidelines for Treatments to Mitigate Opposite Direction Crashes WHAT (Introduction) Shoulders provide a place where a driver can intentionally steer the vehicle off the travel lanes for a variety of reasons (e.g., maintenance or refuge of a disabled vehicle, use of a personal communications device, rest for a fatigued driver, etc.). Shoulders also provide a place for drivers who unintentionally leave the travel lanes to correct their travel path and safely return to the road. Shoulder paving is recognized (32) as a positive countermeasure to reduce a shoulder drop-off hazard that will help stray vehicles resume normal travel, accommodate stopped vehicles to avoid encroachment of the travel way, provide access to emergency vehicles, protect pavement structural integrity, and facilitate highway maintenance work (Figure 13). WHY (Safety State of the Practice) Studies on shoulder width and shoulder paving have shown there are benefits to having a nominal paved shoulder width, though that benefit could decrease as paved width increases beyond a certain value. Key findings from selected studies are summarized in Table 15. Though the findings differ for facility type and location, widening narrow shoulders to provide additional recovery space for an errant vehicle can help to reduce the number of crashes. General Observations Based on previous studies, widening existing shoulders produces incremental benefits, though that incremental benefit may decline for very wide shoulder widths. WHERE (Application Issues) Design Paving an existing width of unpaved shoulder typically does not have a large design impact, other than ensuring appropriate drainage and accounting for adjacent roadside appurtenances or objects (e.g., culverts, guardrails, etc.). Appropriate cross-slope transitions need to be main- tained for drainage without causing a tripping or rollover hazard for vehicles using the shoulder. Adding width to an existing shoulder requires the necessary ROW be available, so acquisition costs could be a sizable portion of the economic component of these widening projects. Adding width may also require moving roadside objects or otherwise accommodating their presence (e.g., lengthening culverts, relocating guardrails, etc.). If extended shoulder paving along the entire length of the road segment is not feasible, select widening at locations with horizontal (Photograph provided by Texas A&M Transportation Institute) Figure 13. Shoulder pavement candidate.

Reduce Likelihood of Head-On Crashes 29   curvature may be more cost effective since vehicles are more likely to off-track at curved locations. Operations Although the original intention of shoulder paving is to enhance highway safety, wider widths of paved shoulders are likely to be associated with higher speeds of travel. As such, the net safety effect of shoulder paving is a combination of several possibly confounding effects: the benefits of allowing for the safe recovery of stray vehicles, and the detrimental tendencies of inviting voluntary shoulder stops, faster travel, and occasional shoulder use for travel (32). Study & Location Year Before (ft) After (ft) Percent Reduction in Crashes (%) Source Total Crashes (KABCO) Injury Crashes (KABC) Urban – Divided Median – widen inside paved shoulders for principal arterials, other freeways and expressways at interchange influence areas (8′ outside paved shoulder) FL 2012 4 6 24 NR (32) 8 36 10 22 12 -4 Urban – Divided Median – widen inside paved shoulders for principal arterials, other freeways and expressways at interchange influence areas (10′ outside paved shoulder) FL 2012 4 5 43 NR (32) 6 44 53 8 -22 22 10 2 34 11 -29 -17 12 5 1 Rural Two-Lane Roads – opposite direction crashes (including SVROR) TX 2019 1 to < 4 ≥ 4 10 50 (13, 33) Rural Two-Lane Roads – previously unpaved shoulders – opposite direction crashes TX 2019 0 2 to 3 28 64 (13, 33) 0 ≥ 4 -1 36 NR – not reported Note: For 2012, the Florida study (32) included studies that received a three-star rating or more at the CMF Clearinghouse. The NCHRP Project 17-66 team developed the other 2019 value as part of that project. The CMF currently does not have a star rating assignment. Table 15. Summary of shoulder pavement studies.