Recommended Guidelines for the Selection of Test Levels 2 Through 5 Bridge Railings (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.

150 SELECTION GUIDELINES The following section presents the recommended selection guidelines and the process for the application of the guidelines for the selection of MASH TL2 though TL5 bridge railings. A risk approach applicable to new construction, rehabilitation and retrofitting is recommended. An alternative cost-benefit approach is provided and discussed following the presentation of the process. A discussion of the process and the policy decisions necessary for implementation are also contained later in this section. Appendix B includes only the process, without any discussion. Appendix B is presented in a format that could be inserted directly into the AASHTO LRFD Bridge Design Specification or the Roadside Design Guide. Bridge Rail Risk Assessment Process The selection of the appropriate MASH test level bridge railing for new or rehabilitation construction is dependent on site-specific conditions and results may differ for each side of the bridge. This process, therefore, should be followed for each bridge edge. These selection procedures only apply to the bridge railing itself. Providing appropriate guardrail-bridge rail transitions, adequate guardrail approaches, and appropriate terminals and crash cushions are also important considerations in the complete safety performance of the bridge. Users should refer to the AASHTO Roadside Design Guide for guidance on appropriate transitions, approach guardrails and terminals. The following selection guidelines include six parts: (1) determine the anticipated construction year traffic volume (AADT); (2) estimate the total encroachments expected over the 30-year life of a 1,000-ft section of the bridge; (3) adjust the expected number of encroachments for site-specific conditions; (4) select the test level from the appropriate chart; (5) additional considerations; and (6) if guidelines do not apply. These steps are described in full below. 1. Traffic Conditions – Determine the anticipated construction year traffic volume (AADT) and percent trucks (PT). These selection guidelines assume an annual traffic growth rate of 2% per year and a design life of 30 years. If the anticipated growth rate or design life are significantly different, use the following equation to compute the equivalent construction year traffic volume for use in these selection guidelines: 𝐴𝐴𝐷𝑇 = 0.7430 ∙ 𝐴𝐴𝐷𝑇 ∙ (1 + 𝐺) / where: AADT0 = The anticipated construction year bi-directional traffic volume (use the one-way traffic volume for one-way roads and ramps), AADTEQ = The equivalent construction year bi-directional traffic volume, G = The anticipated annual traffic growth rate where 0≤G≤1. L = The design life of the bridge railing in years.

151 2. Encroachments – Estimate the total number of encroachments (NENCR) that will be experienced on a 1,000-ft section of the bridge railing during the life of the bridge railing by entering Table 68 or Figure 30 with the bi-directional construction year AADT from Step 1 and the highway type. a. Do not proportion the value of NENCR based on the length of the bridge. The entire method is based on a per 1,000-ft basis. b. If the AADT of interest falls to the right of the end of the curve for the desired highway type, the level of service for the highway is likely D or worse and these procedures cannot be used; refer to Step 6. 3. Site Conditions – Determine the site-specific adjustment factors for the bridge under consideration using the adjustment factors shown in Table 67. Multiply all the adjustments from Table 67 together to obtain fTOT. Find the modified total number of encroachments (NMOD ENCR) either by: a. Drawing a horizontal line in Figure 30 until the curve corresponding to fTOT is obtained (interpolation between lines is acceptable) then reading down to the horizontal axis for the value of the modified total number of encroachments (NMOD ENCR) on 1,000-ft of bridge railing over the 30-year life of the bridge railing or b. Multiplying the estimated encroachments (NENCR) from Table 68 by the total adjustments (fTOT) from Step 2 to obtain the modified total number of encroachments (NMOD ENCR) on 1,000-ft of bridge railing over the 30-year life of the bridge railing. 4. Test Level Selection – Characterize the hazard environment under the bridge as high, medium or low according to the following definitions: HIGH: A high-hazard environment below the bridge includes possible interruption to regional transportation facilities (i.e., high-volume highways, transit and commuter rail, etc.) and/or interaction with a densely populated area below the bridge. Penetrating the railing may limit or impose severe limitations on the regional transportation network (i.e., interstates, rail, etc.). Penetrating the railing also has the possibility of causing multiple fatalities and injuries in addition to the injuries associated with the vehicle occupants. A high- hazard environment is also present if penetration or rolling over the bridge railing could lead to the vehicle damaging a critical structural component of the bridge (e.g., a through-truss bridge). MEDIUM: A medium hazard environment below the bridge includes possible interruption to local transportation facilities, large water bodies used for the shipment of goods or transportation of people, and/or damage to an urban area which is not densely populated. Penetrating the railing would limit

152 local transportation routes, however, detours would be possible and reasonable. Penetrating the railing has the possibility of causing at least one non-motor vehicle injury or fatality. LOW: A low-hazard environment below the bridge includes water bodies not used for transportation, low-volume transportation facilities, or areas without buildings or houses in the vicinity of the bridge. Penetrating a low hazard railing would have little impact on regional or local transportation facilities. A low hazard railing has no buildings or facilities in the area which present possible non-motor vehicle related victims of a rail penetration. Choose the hazard environment most applicable to the bridge under consideration. Enter the appropriate chart in Figure 32 for the hazard environment selected above, the modified lifetime encroachments per 1,000-ft of bridge edge (NMOD ENCR) from Step 3, and the PT from Step 1 to select the appropriate MASH test level for the bridge railing. If the point plots above the dashed risk boundary these charts cannot be used and the engineer should refer to Step 6. 5. Additional Considerations – The bridge railing selected using this process provides a solution where the risk of observing a severe or fatal injury crash over the design life of the bridge railing should be less than 0.01 when the specific site conditions evaluated (i.e., traffic volume and mix, geometry, posted speed limit, and access density) are considered. Engineering judgment should be used when unusual or difficult to characterize site conditions are encountered when selecting a bridge railing. Limited numbers of crash tested bridge railings are available at some test levels, therefore, it is possible that the recommended test level barrier for the evaluated site conditions may not be the best choice for some site conditions not explicitly addressed in these selection guidelines. For example, the particular layout of the barrier at the end of a ramp may influence ISDs and require the use of engineering judgment in designing the interchange to determine an appropriate barrier as it approaches the intersection. Another example might be the presence of pedestrians or bicyclists which might benefit from a higher or different type of railing or the use of sidewalks. Some of the factors that should also be considered are: a. TL5 bridge railings may be appropriate for specially designated hazardous material or truck routes. b. ISD obstructions created by higher test level bridge railings at the ends of ramps or bridges should be considered and the bridge railings may require transitioning to a lower height approaching the intersection. c. Stopping sight distance on bridges where the radius and design speed plot below the dashed line in Figure 31 may limit the use of higher test level bridge railings. d. The presence of pedestrians, bicyclists, snowmobiles, all-terrain vehicles and other recreational vehicles may affect the choice of bridge railing. e. Crash history especially as it relates to heavy vehicle crashes or bridge rail penetrations may justify higher performance bridge railings.

153 f. Regional concerns about snow removal, hydrological impact of flood waters flowing over the bridge, and maintaining scenic views may also play a role in the selection of bridge railings beyond these selection guidelines. g. The capacity of the bridge deck may limit the choices available for higher test level bridge railings on rehabilitation projects. 6. Guidelines Do Not Apply – There are some situations where these guidelines should not be used, namely: a. The traffic conditions violate the free traffic flow assumption used in developing the guidelines such that the estimate of the number of encroachments is not reliable. Generally, this results from a plot point in Figure 30 that is to the right of the end of the highway type line. This indicates that the level of service may be D or worse and the basic assumptions of the method are invalid. b. The user may find that the selection plots above the boundary of Figure 32. In such a case the following options should be considered: i. Can the traffic operational conditions (i.e., AADT and PT) be reduced? ii. Are the roadway characteristics (e.g., horizontal curvature, grade, etc.) resulting in large adjustments to the NENCR? Can the geometry be modified to reduce the adjustments? iii. Can the deck and superstructure support a TL6 bridge railing? These situations require a more detailed analysis of the site conditions that examines a broader range of alternatives beyond just the bridge railing test level selection. A solution will probably require the collaboration of traffic operations, geometric design and bridge railing design engineers to either modify the traffic or geometry conditions of the bridge such that these guidelines can be used or perform a crash history investigation to determine the actual performance of the existing bridge railing.

154 Table 67. Encroachment Adjustments. Posted Speed Lane Width Horizontal Curve Radius Po st ed S pe ed Li m it (m i/h r) U nd iv id ed D iv id ed a nd O ne -w ay A vg , L an e W id th (f t) U nd iv id ed D iv id ed a nd O ne -w ay H or iz on ta l C ur ve R ad iu s at th e C en te rl in e ( ft) A ll H ig hw ay Ty pe s <65 1.42 1.18 9≥ 1.50 1.25 950≥R1 4.00 ≥65 1.00 1.00 10 1.30 1.15 1910>R1>950 (2228.2-R1)//318.3 11 1.05 1.03 1910≤R1 1.00 12≤ 1.00 1.00 950≥R2 2.00 For roads with unposted speed limits use the adjustment for <65 mi/hr 1910>R2>950 (R2-2864.8)/954.9 1910≤R2 1.00 • If road curves toward vehicle on the barrier side use R1. • If road curves away from the vehicle on the barrier side use R2. fPSL = fLW = fHC = Access Density Lanes in One Direction Grade N um be r of A cc es s P oi nt s on B ri dg e or w ith in 2 00 ft of ei th er en d U nd iv id ed D iv id ed a nd O ne -w ay N o. T hr ou gh L an es in O ne D ir ec tio n U nd iv id ed D iv id ed a nd O ne -w ay Pe rc en t G ra de A ll H ig hw ay T yp es 0 1.00 1.00 1 1.00 1.00 -6≥G 2.00 1 1.50 2.00 2 0.76 1.00 -6>G>-2 0.5-(G/4) 2≤ 2.20 4.00 3≤ 0.76 0.91 -2≤G 1.00 fACC = fLN = fG = fTOT=fACC∙fLN∙fLW∙fG∙fHC∙fPSL =

155 Table 68. AADT – Lifetime Encroachments per 1,000-ft of Bridge Railing. AADT 4 LN DIV 2 LN UNDIV 1 LN ONEWAY AADT 4 LN DIV 2 LN UNDIV 1 LN ONEWAY 500 0.8 1.2 1.7 33,000 11.6 4.3 1,000 1.6 1.9 3.4 34,000 11.8 4.4 2,000 3.1 3.2 6.0 35,000 12.1 4.6 3,000 4.4 3.6 8.1 36,000 12.4 4.7 4,000 5.5 3.6 9.6 37,000 12.6 4.8 5,000 6.5 3.4 10.6 38,000 12.9 5.0 6,000 7.4 3.1 11.4 39,000 13.2 5.1 7,000 8.1 2.7 11.8 40,000 13.5 5.2 8,000 8.8 2.3 12.0 41,000 13.9 5.4 9,000 9.3 2.0 12.0 42,000 14.2 5.5 10,000 9.7 1.9 11.9 43,000 14.5 5.6 11,000 10.1 1.8 11.7 44,000 14.9 5.8 12,000 10.4 1.8 11.6 45,000 15.2 5.9 13,000 10.6 1.8 46,000 15.6 6.0 14,000 10.8 1.9 47,000 15.9 15,000 10.9 2.0 48,000 16.2 16,000 11.0 2.1 49,000 16.6 17,000 11.0 2.2 50,000 16.9 18,000 11.0 2.4 51,000 17.2 19,000 10.9 2.5 52,000 17.6 20,000 10.9 2.6 53,000 17.9 21,000 10.8 2.7 54,000 18.3 22,000 10.7 2.9 55,000 18.6 23,000 10.6 3.0 60,000 20.3 24,000 10.6 3.1 65,000 22.0 25,000 10.5 3.3 70,000 23.7 26,000 10.6 3.4 75,000 25.4 27,000 10.7 3.5 80,000 27.1 28,000 10.8 3.7 85,000 28.7 29,000 10.9 3.8 90,000 30.4 30,000 11.0 3.9 95,000 31,000 11.2 4.1 100,000 32,000 11.4 4.2 105,000 33,000 11.6 4.3 110,000 LOS ≥ D LOS ≥ DLOS ≥ D LOS ≥ D

156 Figure 30. AADT – Lifetime Encroachments/1,000-ft of Bridge Railing Nomograph. 110100 NMO D ENCR 1 10 100 100 1,000 10,000 100,000 N EN C R AADT0 or AADTEQ ) 4 LANE DIVIDED 2 LANW UNDIVIDED 1 LANE ONE WAY

157 Figure 31. Minimum Horizontal Curve Radius Based on Barrier Obstruction to the Stopping Sight Distance Compared to AASHTO Exhibit 3-14. 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 25 30 35 40 45 50 55 60 65 70 75 M in im um C ur ve R ad iu s ( ft) Design Speed (mi/hr) AASHTO Ex. 3-14 with e<=10% Min. Radius Based on Barrier Obstruction to SSD

158 LOW MEDIUM HIGH Figure 32. Test Level Selection Nomograph (Risk<0.01 in 30 years for 1000 ft of bridge railing). TL2 or TL3 TL2 or TL3 TL2 or TL3 TL4 TL4 TL4 TL5 TL5 TL5 Risk Boundary Risk Boundary Risk Boundary

159 Discussion Implementation The LRFD Bridge Design Specifications address the issue of the selection of the appropriate test level for bridge railings in Chapter 13 Section 13.7.2 and the Roadside Design Guide addresses the same issue in Chapter 7 Sections 7.3 and 7.5.[AASHTO11, AASHTO12] Section 13.7.2 of the AASHTO LRFD Bridge Design Specification provides general principles for selecting the appropriate test level for bridge railings. The general principles involve using higher capacity barriers for situations where there are more trucks, higher traffic volumes or particularly hazardous conditions. There is no precise definition of what constitutes a “high” traffic volume or a suitably larger percentage of trucks so it is left to the designer to make a subjective decision regarding how to apply the definitions of Section 13.7.2 to the selection for a particular site. Chapter 7 of the Roadside Design Guide generally defers to the AASHTO LRFD Bridge Design Specifications with respect to the selection of an appropriate bridge railing. Section 7.3 provides general guidance on bridge railing test level selection and Section 7.5 provides somewhat more detailed descriptions. Section 7.5 provides the following five factors that should be considered in selecting a bridge railing: 1. Performance, 2. Compatibility, 3. Cost, 4. Field Experience and 5. Aesthetics. Of the five factors, only the first and third (i.e., performance and cost) directly affect the bridge railing selection with respect to the test level. The Roadside Design Guide, in Section 7.5.1, refers to the FHWA policy which requires the use of bridge railings that are crash tested according to Report 350 or subsequent FHWA-approved guidelines such as MASH. [FHWA97a] The FHWA policy also states that “the minimum acceptable bridge railing will be a TL3 … unless supported by a rational selection procedure.” [FHWA97a] The selection procedures recommended herein should satisfy the requirement for a “rational selection procedure” for using TL2 bridge railings in particular locations on the NHS with low volumes and low percentages of trucks. The selection guidelines developed in this project were formulated such that they can be inserted into Section 13.7.2 of the AASHTO LRFD Bridge Design Specification. These new selection guidelines are consistent with the existing wording of Section 13.7.2 and could be inserted after the definitions of the six test levels. It is expected that the AASHTO RDG will continue to defer to the AASHTO LRFD Bridge Specification and refer users to that document although the same guidelines could be inserted into RDG Section 7.5.1 if desired.

160 Critical Values for Design The recommended guidelines use a risk-based method where the risk of observing a severe or fatal crash (i.e., A+K) during the design life of the bridge railing was less than 0.01 per 1000-ft of bridge railing. Several policy decisions are inherent in this choice: • Use of risk rather than benefit-cost, • A critical value of 0.01 risk of a severe or fatal crash during the 30-year design life of each 1,000-ft of bridge railing, • 2 percent annual traffic growth and • The 30-year design life of the bridge railing. While the recommendations assumed a risk-based method, a critical risk of 0.01 and a design life of 30 years, these values can be changed relatively easily within the context of the procedure outlined above. For example, switching from a risk to a benefit-cost approach is accomplished simply by changing the version of Figure 32; likewise, changing from a risk of 0.01 to 0.02 is accomplished simply by using the appropriate figure in place of Figure 32. Figures that can be used to modify the method type (i.e., risk or benefit-cost) and critical value (i.e., various risk or BCR values) are contained in Appendix A. Regardless of which figure is chosen for the final procedures, the process outlined in the last section is exactly the same; only Figure 32 needs to be changed in order to transform the selection procedures from a risk-base to a benefit-cost based procedure or to change the critical design values. If AASHTO SCOBS and TCRS desire to change the recommendations herein, the following replacements can be easily made: For New Construction • Risk of 0.005  Exchange Figure 36 for Figure 32. • Risk of 0.01  Use Figure 32 as shown (i.e., the recommended selection guidelines). • Risk of 0.02  Exchange Figure 37 for Figure 32. • BCR of 1  Exchange Figure 38 for Figure 32. • BCR of 2  Exchange Figure 39 for Figure 32. • BCR of 3  Exchange Figure 40for Figure 32. For Rehabilitation Construction • Risk of 0.005  Exchange Figure 36 for Figure 32 (i.e., same as for new construction). • Risk of 0.01  Use Figure 32 as shown (i.e., the recommended selection guidelines). • Risk of 0.02  Exchange Figure 37 for Figure 32(i.e., same as for new construction). • BCR of 1, 2, or o Consideration of upgrading from R350 TL4  Exchange Figure 41 for Figure 32. o Consideration of upgrading from R350 TL3 Exchange Figure 42 and/or Figure 43for Figure 32. Notice that the selection guidelines are exactly the same for new and rehabilitation construction if a risk-based procedure is used. If a benefit-cost procedure is chosen there will be different versions of the selection figure for new construction versus rehabilitation construction.

161 The cost-benefit rehabilitation selection figures assume that a Report 350 TL3 or TL4 bridge railing is already in place and must be demolished and replaced by either a MASH TL4 or TL5 bridge railing. Test Levels Considerations Currently, FHWA requires that roadside hardware developed and tested after January 1, 2011 be evaluated according to the AASHTO MASH but still allows the use of hardware designed, tested and accepted under Report 350. [AASHTO09] In developing bridge railing selection guidelines, therefore, there is some ambiguity since new hardware will be evaluated under the MASH criteria but existing hardware tested under Report 350 can and likely will still be used on new or retrofit construction. Table 69 shows a list of the TL2 through TL5 impact conditions for both the Report 350 and MASH longitudinal barrier crash tests arranged in order of increasing impact severity. One of the difficulties resolved by MASH was that the nominal impact severity of TL3 and TL4 in Report 350 had converged to about 100 ft-kips under Report 350. The MASH TL4 tests were increased in severity, particularly for TL4, in order to provide a broader range of selection options. One of the results is that MASH TL4 barriers generally need to be at least 36-inches tall rather than the 32-inch height that was common for Report 350 TL4. Table 69. Comparison of Impact Conditions for Report 350 and MASH ordered by Impact Severity. Test Vehicle Mass Speed Angle Nominal Impact Severity Typical Barrier Height lbs mi/hr deg ft-kips in. R350 TL2 4,409 44 25 50 24 MASH TL2 5,004 44 25 57 Unk R350 TL4 17,637 50 15 98 32 R350 TL3 4,409 62 25 102 27 MASH TL3 5,004 62 25 116 31 MASH TL4 22,046 56 15 155 36 R350 TL5 79,367 50 15 441 42 MASH TL5 79,367 50 15 441 42 One of the interesting features of developing both the benefit-cost and risk-based selection guidelines is that Report 350 TL2 and TL3 and MASH TL3 bridge railings tend to overlap each other. The reason that they overlap is that there is a great deal of performance overlap and relatively little difference in cost. As a result, the TL2 bridge railings generally always appear since the performance difference is small and the cost is slightly less (i.e., even in the risk-based criteria, the least costly railing that meets the risk criteria is the preferred alternative). There are few TL2 bridge railings that were designed and specifically crash tested to the Report 350 TL2 conditions and none at this time to the MASH TL2 conditions. For example, the AASHTO-ARTBA-AGC Guide to Bridge Railings currently contains one concrete and one

162 wood railing that were designed for Report 350 TL2.[TF1313] The construction costs of these new TL2 railings are not well documented since there have only been a handful of installations constructed. Similarly, there is virtually no field crash data experience available so the values used in RSAPv3, while reasonable estimates, cannot be validated with field crash data. There are also a number of older bridge railings that were tested under the AASHTO Guide Specifications for Bridge Railings for PL1 that were “grandfathered” into TL2 when Report 350 was adopted. TL2, as shown in these recommended selection guidelines, can be interpreted as any crash tested bridge railing with a height of less than 27 inches. The recommended selection guidelines offered earlier show TL2 in the recommendations, although SCOBS T-7, TCRS and FHWA may rather change it to MASH TL3 as a policy matter until more data is available on the field performance of these bridge railings. The Risk Line Figure 32, as well as its alternate versions in Figure 36 and Figure 37, show a dashed line that indicates the point where even a MASH TL5 bridge railing does not satisfy the selected risk criteria. The line was included so that it is clear to users that the desired criteria could not be met even with a TL5 bridge railing. It is tempting to consider the risk line a de facto indication that a MASH TL6 bridge railing should be used when the point plots above the risk line. In fact, however, a point plotting above the risk line can be interpreted in several different ways including: • The assumptions built into the development of the process may have been violated, • There are traffic or roadway conditions that may need to be examined more closely before making a bridge railing selection or • A TL6 bridge railing may be appropriate. Knowing which of these situations apply requires some further examination as described in the following sections. Regardless of whether a point plots above the risk line or outside the boundary of the figure chosen for the LRFD, Step 6 in the selection guidelines was added to provide some guidance to engineers who encounter situations where the guidelines may not be appropriate for use. The selection figures were developed using RSAPv3 and the encroachment model in RSAPv3 assumes that traffic is generally in a free-flowing condition. This has been interpreted to mean that the level of service is at least C or better. Of course sites with poor levels of service do not operate at those levels at all hours. A particular highway might have a level of service of D or F in peak hours but operate at B or even A conditions at the off-peak hours. At this time it simply not known what the effect of degraded levels of service are on the encroachment models so the predictions from RSAPv3 may not be reliable. Further research is needed to determine how the encroachment relationships change at high traffic volumes and levels of service of D or E. If the traffic volumes for a particular highway type result in a level of service of D or greater, the resulting selection may plot above the risk boundary. To prevent this from happening, Figure 30 and Table 68 were constructed such that each line representing a highway

163 type ends at the AADT corresponding to the transition from level of service C to D for 40 PT. The engineer is also told explicitly in Step 2b not to extrapolate to the right of the highway type lines because doing so will violate the basic assumptions used to develop the tables. If a particular AADT plots to the right of the highway type line, the user is directed to Step 6 for advice. A selection above the risk boundary should not be the result of poor level of service as long as the user has followed the instructions in Step 2. More important is the fact that the risk line indicates that there are probably other issues related to the site or traffic conditions which a simple choice of bridge railing test level may not adequately address. The selection figures use the expected number of lifetime encroachments on a 1,000-ft section, the percent of trucks and the hazard environment as input to select a bridge railing. If a particular bridge situation plots above the risk line or outside the boundary of the figure then the combination of these three input values have resulted in a situation where one of the other inputs should be considered for change. The hazard environment determines which figure is used in the selection process. The hazard environment is determined based on the character of the area beneath and around the bridge. Changing the hazard environment is generally not a feasible alternative since it involves land use outside the typical DOT’s right of way and control or very expensive changes to the transportation infrastructure. For example, a gasoline tank farm may be located beneath a bridge causing the hazard environment to be categorized as a high. While moving the tank farm is a theoretical possibility it is probably not practical due to the expense involved as well as the need to coordinate and collaborate with private property owners and a variety of local agencies. For these types of reasons, changing the hazard environment is generally not an option. Sometimes selections below the risk boundary could be made if the PT were reduced thereby moving the plot point to the left. If a particular bridge plots above the hazard line the engineer may want to consider if the high percentage of truck traffic is desirable at the site and if there are ways to reduce the truck traffic in the long term. Returning to the tank farm example, if the site experienced 30 PT and was located in a heavily urban area it might make sense to consider redirecting truck through-traffic to another route to avoid the high-hazard areas (e.g., loop route around urban area). Reducing the PT from 30 to 10 may be enough to reduce the risk below the risk line. Obviously, this alternative has consequences far afield from the consideration of the one bridge under consideration since it would involve a change in the operation of the highway network and how traffic is managed. Probably the most important consideration, however, is to examine why so many encroachments are predicted at the site. There are several possible reasons for predicting a large number of encroachments including (1) very high traffic volumes and/or (2) geometric characteristics that result in large adjustments. The adjustment factors can increase the expected number of encroachments dramatically. For example, if a one-way ramp has a radius of horizontal curvature of 950 ft, a 6 percent downhill grade and a speed limit of 45 mi/hr, Table 67 indicates an encroachment adjustment of 9.44 is needed. This is a very high adjustment indicating that the geometry of the roadway may be very challenging. This adjustment applied to encroachments for a moderate AADT could well place the site above the risk boundary. The engineer should seriously consider addressing the curvature and grade of the site to reduce the number of encroachments expected and bring the site conditions below the risk boundary.

164 While a MASH TL6 bridge railing might be appropriate for some conditions above the dotted risk line, there is at present only one crash tested TL6 bridge railing, the 90-inch tall TX T80TT bridge railing. [TXDOT13] This bridge railing requires the use of non-standard deck details since the dead-load of the bridge railing and overturning resistance are so large. Because it cannot be used on a conventional deck it is not generally a viable option for most bridge construction projects. Before selecting a TL6 bridge railing, the engineer will need to carefully examine the structural characteristics of the deck and bridge structure as well as traffic and site conditions to determine if a TL6 bridge railing is a realistic alternative. There is no guaranteed that even if a TL6 bridge railing is used it will satisfy the risk criteria since sometimes the added risk is a result of very high traffic volumes and passenger vehicle redirection rollovers which are likely not improved with a TL6 bridge railing. As these examples illustrate, when a point plots above the risk boundary in Figure 32 reducing the risk will almost always involve a more comprehensive approach than simply selecting a bridge railing. Reducing the risk may involve highway geometric design, traffic operations and management, structural design of the bridge deck and superstructure as well as the selection of a bridge railing. Recommended Selection Guidelines Verification The 1989 AASHTO Guide Specification to Bridge Rails, NCHRP Report 22-08 and a series of example problems were reviewed and compared to the recommended procedure as a verification exercise. Comparisons to the 1989 AASHTO Guide Specification and NCHRP 22-08 The 1989 AASHTO Guide Specification for Bridge Railings and the selection guidelines that were prepared in NCHRP 22-08 have been converted to the same format as the proposed selection guidelines for comparison purposes. These converted tables are shown in Figure 33 on the left side. While the purpose of this project is not to mimic the 1989 AASHTO GSBR guidelines, those earlier guidelines do provide some insight into what roadside safety engineers in the past have considered “reasonable” selection guidelines. The proposed selection guidelines from this project are compared to the 1989 GSBR and the NCHRP 22-08 selection guidelines to gauge how these new guidelines compare to what was accepted to some degree in the past. Figure 33 shows the recommended risk figure for a medium hazard in the center overlaid on the 1989 GSBR guidelines. It appears the recommendations from this research are slightly more conservative than the 1989 GSBR for a lifetime risk of 0.01. Figure 33 also shows the risk of 0.03, which more closely matches the 1989 GSBR guidelines. Figure 34 provides a comparison of the GSBR with the recommended BCR approach. Figure 35 directly compares the GSBR selection guidelines, the recommended risk-based approach and the corresponding cost- benefit approach. Each of these tables have underlying assumptions and adjustments which cannot be captured in a straight-forward comparison of the charts. Furthermore, the GSBR was developed using bridge rails designed to a different performance specification. While no official crash test equivalencies have been released to compare Report 350 and MASH test level, Table 70 was originally released by the FHWA to compare the GSBR performance levels with Report 350 and

165 was expanded to add the first line to represent the approximately equivalencies for MASH test levels. Table 70. Approximate Crash Test Acceptance Equivalencies. [after Horne97] Bridge Railing Testing Criteria Acceptance Equivalencies MASH TL1 TL2 TL3 TL4 TL5 TL6 Report 350 TL1 TL2 TL3 TL4 TL5 TL6 Report 230 MSL-1 MSL-2† 1989 AASHTO Guide Spec PL1 PL2 PL3 AASHTO LRFD Bridge Spec PL1 PL2 PL3 † This is the performance level usually cited when describing a barrier tested under NCHRP Report 230. It is close to TL3 but adequate TL3 performance cannot be assured without a pickup truck test. Each of these guidelines has been used to evaluate a series of example problems where the bridge rail is already in place as a verification exercise. Example Bridge Railing Selection Table 71 shows several example bridge railings along with the characteristics of the site, traffic conditions and the type of bridge railings currently installed at the bridge. These examples are meant to compare what some agencies currently have installed and compare the current installation to the bridge railings selected by the recommended procedure and the 1989 GSBR. Table 71 also serves as a verification exercise since the bridge railing selection determined from the recommended procedure is compared to the result recommended by an independent RSAPv3 analysis. While the recommended selection guidelines are based on numerous RSAPv3 simulations, the results were re-formulated, rearranged and simplified to develop the recommended procedure so it was important to verify that the final procedure was consistent with individually generated results from RSAPv3. As shown in Table 71, the recommended selection guidelines produce the same or slightly more conservative results than performing an RSAPv3 analysis. This verification exercise was completed for each risk level and BCR presented in the alternate figures shown in Appendix A.

166 1989 GSBR and 22-08 Findings Risk≤0.01 overlaid on 89GSBR Risk≤.03 overlaid on 89GSBR Figure 33. Comparison of 89GSBR, NCHRP22-08 with Risk values. MASH TL5 MASH TL4 GSBR PL3 MASH TL5 MASH TL4 MASH TL2 GSBR PL2 MASH TL2 GSBR PL1 22-08 PL1/PL2 breakline 22-08 PL2/PL3 breakline 22-08 GSBR 89GSBR PL1/PL2 breakline 89GSBR PL2/PL3 breakline 89GSBR PL2/PL3 breakline 89GSBR PL1/PL2 breakline

167 1989 GSBR and 22-08 Findings BCR=2 overlaid on 89GSBR BCR=5 overlaid on 89GSBR Figure 34. 89GSBR, NCHRP22-08 with BCR values. MASH TL5 GSBR PL3 GSBR PL2 MASH TL2 GSBR PL1 22-08 PL1/PL2 breakline 22-08 PL2/PL3 breakline 22-08 GSBR MASH TL5 MASH TL2 89GSBR PL1/PL2 breakline 89GSBR PL2/PL3 breakline 89GSBR PL2/PL3 breakline 89GSBR PL1/PL2 breakline

168 A. MASH TL2 satisfies risk and cost- benefit criteria. B. MASH TL2 satisfies risk criteria, MASH TL3 is cost-beneficial. C. MASH TL2 satisfies risk criteria, MASH TL4 is cost-beneficial. D. MASH TL4 satisfies risk criteria, MASH TL2 is cost-beneficial. E. MASH TL4 satisfies risk criteria, MASH TL3 is cost-beneficial. F. MASH TL4 satisfies risk criteria, MASH TL4 is cost-beneficial. G. MASH TL4 satisfies the risk criteria, MASH TL5 is cost- beneficial. H. MASH TL5 satisfies the risk criteria, MASH TL5 is cost- beneficial. A-F comparable to 89GSBR PL1 G comparable to 89 GSBR PL2 H is comparable to 89 GSBR PL3 1989 GSBR overlaid on findings Risk≤0.01 and BCR=2 Comparison of Risk and CBR tables Figure 35. 89GSBR, Risk and BCR comparison. G A D C B E F H A D C B E F H G

169 Table 71. Selected Examples of Existing Bridge Railings Compared to the Recommended Selection Guidelines.

170 Table 71. Selected Examples of Existing Bridge Railings Compared to the Recommended Selection Guidelines. (continued)

171 Table 71. Selected Examples of Existing Bridge Railings Compared to the Recommended Selection Guidelines. (continued)

172 Table 71. Selected Examples of Existing Bridge Railings Compared to the Recommended Selection Guidelines. (continued)