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.

71 RESULTS OF SURVEY OF PRACTICE Introduction A survey of practitioners was distributed via e-mail to about 2,800 roadside safety researchers, bridge engineers, DOT engineers, and highway design consultants both within the US and abroad. This survey was conducted to determine the current policies for deciding which test level bridge railing to use in particular situations and to obtain data which could be used in this research. The distribution list was compiled from the AASHTO Subcommittee for Bridges and Structures mailing list, the TRB AFB20 and AFB20 subcommittees’ mailing lists, the ITE database, AASHTO-ARTBA-AGC TF13 mailing list, ATSSA training course participant list, and from a list of people who have purchased the Roadside Design Guide from AASHTO. The survey was assembled and made available using the on-line tool surveymonkey.com (i.e., www.surveymonkey.com). The survey had several purposes including: • What is the default bridge railing test level used in each State or country? • Are there specific warrants for using a different test level than the default? • If there are not specific warrants, are there informal guidelines or criteria for deciding when to use a different performance bridge railing? • Are construction or repair cost data available for bridge railings? • Can the bridge inventory be linked to the States crash data? Approximately 54 people started the survey and 45 people completed it resulting in a completion rate of 83 percent. The remaining recipients that did not respond in any way are presumed to know nothing about the selection of bridge railings or are not active in the roadside safety aspects of bridge design. The survey asked a variety of questions about bridge rail selection guidelines in use, bridge railing inventories and crash databases. The following sections discuss each question and summarize the responses. Survey Questions Question 1: Please provide the following optional information about yourself. Respondents were asked to provide contact information. Approximately 90 percent of the respondents provided this information. Respondents represent a variety of countries including the United States, England, China, and New Zealand. Representatives from 32 states responded (listed below), however, not all of these respondents represent a state Department of Transportation (DOT). Many are consultants or researchers who have exposure to local and regional design guidelines. Additionally, some respondents represent county level engineering departments responsible for their own bridge rails.

72 1. Alabama 2. Florida 3. Georgia 4. Hawaii 5. Illinois 6. Indiana 7. Iowa 8. Kansas 9. Kentucky 10. Louisiana 11. Maryland 12. Massachusetts 13. Michigan 14. Minnesota 15. Mississippi 16. Nebraska 17. Nevada 18. New Jersey 19. New York 20. North Dakota 21. Ohio 22. Oregon 23. Pennsylvania 24. Rhode Island 25. South Carolina 26. South Dakota 27. Texas 28. Utah 29. Virginia 30. Washington 31. Wisconsin 32. Wyoming In summary, this wide variety of respondents provided a good cross-section regarding the type of data available for this study as well the practices currently in use both nationally and internationally. Question 2: What are the approximate percentages of TL2, TL3, TL4, TL5, and TL6 bridge rails you have in your jurisdiction? Respondents were asked to provide specific information about the distribution of bridge rails within their jurisdictions. The test levels referred to in this question are Report 350 test levels. The approximate percentage, not mileage, of the rails was obtained from 35 of the people who took this survey. An additional five participants answered “unknown” while many others skipped the question altogether presumably because they could not provide an estimate. Figure 17 graphically displays the distribution of results. The horizontal axis is the range of possible percentages from the respondents. The vertical axis is a ratio of the responses per total respondents to the question which fall into each category. There appears to be a small percentage of TL5 bridge rails currently installed, while other test levels appear to vary widely. Only one respondent indicated that TL6 bridge rails are installed, therefore, the TL6 rails were not included in Figure 17. Another way to look at these results is to consider the average value for each category. Table 22 provides the average values of the responses. Again, there appear to be few TL5 bridge rails currently in-service, however, TL2 through TL4 barriers appear to have wide usage rates.

73 Table 22. Question 2 Average Values. TL2 20% TL3 32% TL4 51% TL5 6% Figure 17. Distribution of Reponses to Question 2. With regards to TL4 bridge rails, respondents noted that “65% of new LRFD designed bridges” use TL4 bridge rails or that “all new construction since early 80's uses TL4.” One respondent also pointed out the disparity between State and Federal routes verses County and Town routes when noting that “95% State and Federal routes; 10% County/Township” are using TL4 bridge rails. It is important to also keep in mind, as a respondent pointed out, that some jurisdictions have “60% TL2 or less timber rail from 1960 or earlier.” In summary, there is a wide range of different bridge railing test levels in use. Not surprisingly, the most common bridge railings installed appear to be TL3 and TL4. TL4 bridge railings appear to be the most common at least on newer construction and there are a few higher-level bridge railings as well (i.e., TL5). 0.00 0.10 0.20 0.30 0.40 0.50 >1 0% 10 -2 0% 21 -3 0% 31 -4 0% 41 -5 0% 51 -6 0% 61 -7 0% 71 -8 0% 81 -9 0% 91 -1 00 % Un kn ow n Re sp on ce s/ To ta l Amount in-service TL-2 TL-3 TL-4 TL-5

74 Question 3: As part of this work, the research team is collecting in-service crash records for bridge railings. If you or your agency has data available to help with the development of these guidelines, please list the best way to contact you and the nature of the data in the box below. Four survey respondents listed contact information and suggested the research team contact them directly. The information gained from these individuals was included earlier in the Literature Review. Question 4: Are you aware of a bridge inventory which can be linked to State crash data? The respondents are not aware of any bridge inventories which are linked to State crash records. One respondent acknowledged that it may be possible but noted that “…it is a difficult process. It requires a tedious manual effort of overlapping two data bases of data. Unfortunately we don't have the personnel available to provide this information in a timely manner.” Question 5: Has your State sponsored any in-service performance studies of bridge rails or median barriers? While not a bridge rail study per se, the state of Washington conducted an in- service review of cable median barriers and concrete safety shape median barriers. The concrete safety shape results of that study were useful in postulating the effectiveness of similar concrete safety shaped bridge railings. A similar situation appeared for New Jersey where there is a study of concrete median barriers which provided some insight on similar concrete bridge railings. The survey respondent noted these documents can be found here: • Cable median barriers (http://www.wsdot.wa.gov/Projects/CableBarrier/Report2009.htm) • Jersey shape median barriers (http://www.wsdot.wa.gov/publications/fulltext/design/RoadsideSafety/T RB_Report.pdf) A survey respondent suggested the research team review documents found at http://www.highways.gov.uk/business/14008.aspx. These documents are research reports for a study of “Whole Life Cost-Benefit Analysis for Median Safety Barriers” where the costs for cable barrier and concrete barrier penetration crashes, the costs for the relocation of services when concrete barrier is installed, and the costs for traffic management were reviewed. Again, the study is not specifically about bridge railing but was used to obtain some useful information about similar roadside barriers extrapolated to bridge railings. One survey respondent suggested the research team review a Michigan study on the performance of concrete railings and barriers at http://www.michigan.gov/documents/mdot/MDOT_Research_Report_R1498_207581_7. pdf. Michigan conducted a study of the premature deterioration of Michigan’s concrete railings and barriers. The study found that material specifications, construction methods

75 and maintenance practices all contributed to the observed deterioration. In addition to the reports listed above, one respondent suggested that the research team make direct contact to obtain information. In summary, a few in-service studies of median barriers or bridge rails were reviewed in detail and have been summarized in the Literature Review section of this report. Question 6: Are you aware of construction or repair costs available for bridge rails? Many respondents suggested that bridge rail repair and retrofit costs range from “$50 to $300 per linear foot based on the extent of repair/retrofit” needed and the type of railing under repair. New construction costs in Massachusetts, for example, were reported as: • “S3-TL4 metal railing: $320/FT; • TL4 concrete barriers: $150/FT; and • TL5 concrete barrier: $200/FT.” These costs were “calculated from the price guidelines in the MassDOT Bridge Manual that can be downloaded from the following link: http://www.mhd.state.ma.us/default.asp?pgid=content/bridgeman_new02&sid=about.” Respondents provided these additional websites and directions for locating bridge rail construction costs: • http://www.txdot.gov/business/avgd.htm • Historical construction costs for some FDOT standard bridge traffic railings are available here under Items 0460-71-1 and 0521-5-x: ftp://ftp.dot.state.fl.us/LTS/CO/Estimates/12MonthsMoving.pdf • http://www.tdot.state.tn.us/construction/Average%20Unit%20Prices/aup_201 0.pdf • Once in the Indiana DOT website, in http://www.in.gov/dot/div/contracts/pay/ - click on English Unit Price Summary, then scroll to 706 in column A. Records identified as 706 are the bridge railing pay items. In summary, repair and retrofit costs range from $50 to $300/LF while new construction ranges from $150 to 320/LF. It is assumed there are large discrepancies by region and bridge rail test level, as well as bridge rail design. Construction costs are further reviewed and discussed later in this report. Question 7: In the conduct of your work, have you encountered guidelines for the placement of particular test level bridge rails or median barriers under certain situations? Reponses varied considerably to this question. One respondent stated that “PennDOT uses a TL5 barrier except in limited situations where sight distance is an issue. For these situations a 32 inch barrier is used.” Conversely, another respondent said that “WYDOT uses TL3 rails on all bridges, except on I-80 and in areas of high truck

76 traffic, then TL4 railing is used.” Other respondents indicated that Alabama specifies a minimum of TL4 railing; WSDOT requires “TL5 at T-intersections on a bridge or structure or when the barrier is on the outside curve of a structure with radius of curvature less than 500 ft.” WSDOT also requires TL5 “where approach speeds are 50 mph or greater.” “Some standards call for the use of TL5 bridge rails for horizontal curved bridge with high speeds while allowing for TL3 bridge rails to accommodate historic bridge needs.” Many States rely on the 1989 AASHTO Guide Specifications for Bridge Railing in combination with the AASHTO LRFD Bridge Design Specifications, AASHTO Roadside Design Guide, and the AASHTO Task Force 13 barrier guide. Some States, including Utah and New Mexico, are in the process of updating design standards to include guidelines for the selection of bridge rails and median barriers. Maryland has a draft Policy and Procedure Memorandum (D-78-16(4) Barrier Railing Systems on New or Rehabilitated Structures) for selecting bridge railings. Internationally, the Chinese standard is issued by Ministry of Transport. In the UK, the specification is TD19/06, and New Zealand follows NCHRP 350 guidelines and at times references EN1317. In summary, there does not appear to be any consensus about when to use what type of bridge railing, however, there are some existing guidelines. Question 8: Does your organization have a published set of guidelines for selecting bridge rails? Several survey respondents provided references to specific design guidelines which were reviewed and summarized in the Literature Review. These references are shown here: • FDOT Structures Design Guidelines, Section 6.7: http://www.dot.state.fl.us/Structures/StructuresManual/CurrentRelease/Structu resManual.shtm • MNDOT Bridge Design Manual: http://www.dot.state.mn.us/bridge/manuals/LRFD/index.html • NJDOT Design Manual for Bridges and Structures, 5th Edition, Section 23.: http://www.state.nj.us/transportation/eng/documents/BSDM • Rhode Island Bridge Engineering Design Guides http://www.dot.ri.gov/engineering/guides/index.asp • Michigan Bridge Design Manual: http://www.michigan.gov/mdot/0,1607,7- 151-9622---,00.html • Chapter 13 of Design Manual Part 4. ftp://ftp.dot.state.pa.us/public/PubsForms/Publications/PUB%2015M.pdf • Chapter 3 of Part I of the MassDOT Bridge Manual that you can download from the following link:

77 http://www.mhd.state.ma.us/default.asp?pgid=content/bridgeman_new02&sid =about • NYSDOT Bridge Manual: https://www.nysdot.gov/divisions/engineering/structures/repository/manuals/b rman-usc/Section_6_US_2010.pdf • http://www.dot.state.oh.us/Divisions/HighwayOps/Structures/standard/Bridge s/test/railing%20selection%20procedure.pdf • NCDOT Design Manual, Section 6.2.4: http://www.ncdot.org/doh/preconstruct/highway/structur/designmanual/lrfd/L RFDManual(December2010).pdf • http://www.dot.nd.gov/manuals/design/designmanual/designmanual.htm • http://www.dot.state.mn.us/bridge/manuals/LRFD/index.html • SCDOT Bridge Design Manual: http://www.scdot.org/doing/bridge/06design_manual.shtml • Illinois Bridge Manual Section 2.3.6.1.7.: http://www.dot.il.gov/bridges/brmanuals.html • INDOT Design Manual: http://www.in.gov/dot/div/contracts/standards/dm/english/index.html, click on "Structural Design", then BRIDGE DECKS. Scroll down the bookmark to 61- 6.0 BRIDGE RAILING, then click on it. The Manual copy then appears on the right. • NDOT Structures Manual, Chapter 16, Section 16.5.1.2.: http://www.nevadadot.com/divisions/011/ • WSDOT BDM, Chapter 10. Search the WSDOT website for the most recent electronic version. • Virginia DOT Structure and Bridge Division Manuals/Manual of the Structure and Bridge Division, Volume V-Part 2, Chapter 25. Question 9: Which guidelines does your organization use for selecting/specifying bridge rail crash test performance? Approximately two-thirds of the respondents indicated that NCHRP Report 350 is used for specifying crash test performance of bridge railings, however, many respondents indicated that the organizations they represent are transitioning to MASH. Participants also noted the Chinese standard of transport JT/T F83-2004 and EN1317 are used for specifying bridge rail crash test performance internationally. Question 10: Which criteria do you use in the rail selection process? Survey respondents were asked which criteria apply when selecting bridge rails. Respondents were given several choices and allowed to select as many choices as applicable. Choices included: • Engineering Judgment,

78 • ADT, • Percent Trucks, • Posted Speed Limit, • Design Speed, • Number of Lanes, • Accident data, • Cost/benefit, • Roadway type, • Horizontal alignment, • Vertical alignment, • Land use around the structure, and • Other. Design speed, traffic characteristics (i.e., volume and truck percentage), and engineering judgment have the most impact on the selection of bridge rails while the roadway type, horizontal alignment, and land use also appear to significantly impact selection. All of the responses are displayed in Figure 18. Figure 18. Distribution of Question 10 Results. Survey respondents noted that accommodations for pedestrians, bicycles, flood waters, snow removal and scenic views also play rolls in the selection of bridge rails. Other considerations range from shoulder width and sight distance to funding source (i.e., local, state, federal, etc.) Often times the weight and maintenance costs of the rail are

79 concerns. “Aluminum rails are being stolen at an alarmingly high rate, so the state of Iowa is proactive about replacing them with the lightest functional rail available.” One survey respondent noted “all of the above criteria are considered, however, there are not any published guidelines that cover all of the possible combinations …. A formula would be nice for determining what test level to use.” Question 11: Does your organization have a policy for identifying and/or retrofitting substandard bridge rails? Approximately 40 percent of the respondents indicated the organization they represent does not have retrofitting policies. Many of the 60 percent of respondents, who do have policies, have informal or unwritten policies. Many States inspect bridge railing during bridge inspections. Substandard rails are identified based on age and height in some cases or by type and visual inspection. Respondents noted that when a highway safety or resurfacing project is under construction in the area, “any structures within the limits of work are evaluated for the parapet/railing type and condition, as well as the transition from the roadway approach barrier to the bridge parapet. Deficient transitions are typically brought up to standards within the safety/resurfacing contract. Deficient bridge barriers are noted and included in a regionalized list of deficient bridge barriers. Periodic area-wide contracts are procured to upgrade deficient bridge barriers within a region.” Other respondents note that rails are only reviewed “per federal funding criteria, when an existing bridge is being rehabilitated.” Respondents suggested a review of policies at these locations: • Minnesota Bridge Preservation, Improvement and Replacement Guidelines: http://www.dot.state.mn.us/bridge/documentsformslinks/techmemos/06-10-b- 01.pdf. A more current version should be available soon. • ODOT Bridge Design Manual, Section 304.1: http://www.dot.state.oh.us/Divisions/HighwayOps/Structures/standard/Bridge s/BDM/BDM2004/BDM2004_10-15-10.pdf • Indiana Design Manual: http://www.in.gov/dot/div/contracts/standards/dm/english/index.html - click on "BRIDGE REHABILIATION". 72-3.01(03) Rehabilitation Techniques. Table BD-7. Also, 72-7.02(04) Bridge Railing. • Illinois Bridge Design Manual: http://www.dot.il.gov/desenv/demanuals.html • WSDOT Bridge Design, Section 10.4. Search for "BDM" on the WSDOT home page. It appears very few organizations have formal policies for identifying substandard bridge rail.

80 Question 12: What testing has been done to support the retrofitting of substandard bridge rail policy? Respondents indicate that retrofitting policy is based on materials found in NCHRP Report 350, the 2005 FHWA/CALTRANS Bridge Rail Guide, and historic crash tests. [FHWA05] Many respondents indicated that no additional testing has been done in support of the policies, which is not surprising given the lack of formal policies. Question 13: Does your organization have a default bridge rail crash performance Test Level installed/specified in the absence of other controlling criteria? Approximately 65 percent of the respondents indicate that the organization they represent have minimum requirements for bridge rail crash performance. Two respondents went on to explain this minimum is TL3 while 12 respondents explained the minimum is TL4. Three respondents indicated that TL5 railings are used exclusively on interstate highways. Question 14: What is the default bridge rail? Respondents indicated a variety of default rail choices. Many defaults including 32” concrete barriers such as the New Jersey shape, the F-Shape and the Single Sloped shape. Some respondents report the default use of 42” single sloped concrete on the Interstate and US Route systems. Other default rails are shown here: • TL3 steel two tube railing, • Kansas corral rail, • Oregon Rail, • 34" Single Slope, • Side mounted R-34 railing, • S3-TL4 where a see-through railing is needed, • Usually the 4-rail system, or • EN1317, N2 or H4a. In summary, responses suggest a variety of default rails currently in use. Over 65 percent of respondents, in fact, indicate that default rails are suggested in their region. Question 15: Are there any reasons for not installing the default bridge rail? Not surprisingly, over 75 percent of the respondents indicated that there are often reasons for not installing the default bridge rail. In some cases, the standard rails do not meet the needs of the situation. For example, the rails may not attach to some types of bridges with adequate height, rails may not combine with sound barriers or rails may not transition well to existing rails. Respondents noted that aesthetics, the need for a higher or lower test level, vertical and/or horizontal alignment and sight distance issues may all lead to installing bridge rails other than the default. Respondents also noted that often times a super-elevated horizontal curve, a high design speed, significant truck traffic or high crash rates may lead to the installation of rails other than the default. Sometimes it is more cost effective to use different

81 construction methods (e.g., extrude the railing). Dead load may be an issue as well, especially for re-decking projects. In historic areas or areas using context sensitive design principles, some lower test level rails are installed for aesthetic reasons. In summary, many highway design elements are factors when choosing a bridge rail other than the default. Summary It appears there are few in-service studies of median barriers or bridge rails. The respondents were not aware of any bridge inventories which are linked to state crash records. Only four crash databases were located as part of this survey. Survey respondents noted the difficulty in linking crash records with state databases. There does not appear to be any consensus about when to use what type of bridge railing, however there are some existing guidelines both formal and informal. Several survey respondents provided references to specific design guidelines. Approximately 65 percent of the respondents indicate that the organizations they represent have minimum requirements for bridge rail crash performance and/or default bridge railings. Many highway design elements are factors when choosing a bridge rail other than the default, however these design decisions and polices appear to lack formality. One survey respondent noted “A formula would be nice for determining what test level to use.” Few organizations have formal policies for identifying substandard bridge railings, however, retrofitting policies are based on materials found in NCHRP Report 350, the 2005 Bridge Rail Guide, and historic crash tests. Many respondents indicated that no additional testing has been done in support of the policies. The wide variety of respondents represented a good cross-section of the current practices both nationally and internationally. With bridge rail repair and retrofit costs ranging from $50 to $300/LF while new construction ranging from $150 to 320/LF, the time for formal national guidelines which examine the appropriate installation of different test level barriers under different scenarios has come.