Criteria for Restoration of Longitudinal Barriers (2010)

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.

32.1 Objective The purpose of this chapter is to synthesize current U.S. and Canadian criteria for repair of damaged flexible or semi- rigid longitudinal barriers. 2.2 Methodology The general methodology for this study was to both examine the available literature and conduct a survey of transportation agencies to ascertain current damaged barrier repair thresholds among transportation agencies in the U.S. and Canada. The literature review focused on available national guardrail repair guidance and individual agency guidelines for the repair and maintenance of semi-rigid and flexible longitudinal barriers. These individual agency guidelines generally fell into two cat- egories: (1) maintenance manuals that describe conditions that warrant repairs on a particular barrier and (2) mainte- nance assessment criteria that are used to assess the barrier condition against a reference condition. Maintenance assess- ment criteria typically evaluate barrier functionality but can also include other factors such as aesthetics. Although mainte- nance assessment criteria may not be directly linked to barrier repair, they have been included as they are a gauge of barrier condition. Using the findings from the literature survey, a survey in- strument was developed for distribution to U.S. and Canadian transportation agencies. The 22 question survey was organized into the following five sections: • Inventory of Guardrail and Median Barriers; • Repair Policies; • Non-Crash Related Damage/Deterioration; • Notification and Repair Responsibilities; and • Inspection Policies and Procedures. The purpose of the barrier inventory section was to under- stand the types of barriers most used within a particular agency’s jurisdiction. The repair policies section, the crux of the survey, was intended to provide insight into what thresholds are currently used to determine barrier repair needs, how dam- aged sites are prioritized, timelines for repairs, documented cases of impacts into damaged barriers, and whether the agency would benefit from more quantitative barrier repair guidelines. This chapter presents a summary of the survey results on the guardrail inventory and repair policies sections. 2.3 Results 2.3.1 National Guardrail Repair Guidance National guidance regarding the repair of w-beam barri- ers is provided by the FHWA’s “W-Beam Guardrail Repair: A Guide for Highway and Street Maintenance” (2008). This document provides highway maintenance personnel with a comprehensive overview of the importance and logistics of w-beam barrier repair. Guidance is provided on determining whether repair is necessary, based on a site visit and a classifi- cation of the damage severity. A damaged barrier is classified into one of three categories, as summarized in Table 1. According to the FHWA guidelines, each transportation agency should develop guidelines for the timing of repair for each damage category. The FHWA report recommends that timing be based upon the expected frequency with which the damaged section will be struck, the severity of impact to the damaged section, and agency resources. Despite the relatively quantitative description of the damage categories shown in Table 1, the guidelines appear to have been developed based on previous state experience with w-beam barrier and engi- neering judgment. The report does not reference either tests or quantitative analysis as the basis for the guidelines. The American Association of State Highway and Trans- portation Officials (AASHTO) also provide guidelines on longitudinal barrier maintenance in their Maintenance Man- ual (AASHTO, 2007). Although comprehensive in terms of what types of damage requires repair, little is provided in terms C H A P T E R 2 Synthesis of Current Repair Criteria for Longitudinal Barriers with Crash Damage

criteria and 3 maintenance manual criteria). For the purpose of this study, “quantitative” was defined as both objective and measurable. A guideline indicating that posts out of alignment more than 305 mm (12 inches) horizontally require repair, for instance, would be considered “quantitative.” However, a guideline indicating that barriers need to be repaired if 5% of the barrier is not functional would not be classified as “quantitative” as there is no measurable definition of “not functional.” For transportation agencies, quantitative barrier repair criteria are important for consistently and objectively identifying barrier damage that requires repair. As additional quantitative barrier repair criteria were iden- tified via the survey responses, all quantitative criteria were combined and discussed further in the survey results section. Table 2 summarizes selected agency barrier repair thresholds 4 Damage Category Damage Attributes (1) Non-Functional Rail element is no longer continuous 3 or more posts broken off or no longer attached to rail Deflection of rail element more than 18 in. Rail element torn Top of rail less than 24 in. (2) Damaged but should function adequately under majority of impacts Rail element is continuous (can be bent or crushed significantly) 2 or fewer posts are broken or separated from the rail element Deflection of the rail element is less than 12 in. (3) Damaged but should not impair the guardrail’s ability to perform Rail element is continuous (can be crushed or flattened) No posts are broken off or separated from the rail element Deflection of the rail element is less than 6 in. Table 1. Guardrail damage classification details (FHWA, 2008). Agency Type* Criteria Description/Excerpt (Reference) Alabama DOT MM Repair or replacement of guardrail sections, posts and hardware due to crash damage or normal deterioration. (AL DOT, 2005) Idaho Transportation Department MM Any guardrail that is damaged. Most guidance is with respect to upgrading non-standard guardrail to standard hardware if it is damaged. (ID TD, 2008) Indiana DOT MM Maintain guardrail to assure that it will function as designed. Repairs of non-functional barrier should be performed within 5 working days. (IN DOT, 2001) Kentucky Transportation Cabinet (TC) MA Measure and record the total linear feet of guardrail that is damaged to the extent that structural integrity or functionality is lost. (KY TC, 2000) Michigan DOT MM Only a description of how repair work should be completed. No criteria for when guardrail is considered deficient or should be repaired. (MI DOT, 2004) Montana DOT MM “Guardrails are repaired and replaced in order to maintain its structural integrity” (MT DOT, 2002a) North Carolina DOT MA Threshold condition is “Guardrail damaged or not functioning as designed.” (NC DOT, 1998; NC DOT, 2004) Oregon DOT MM Description only of the work involved. Maintain, repair, realign, or replace guardrail to preserve or restore the installation to its designed condition. (OR DOT, 2004) South Carolina DOT MA Threshold condition: “Guardrail damaged or not functioning as designed.” (SC DOT, 2004) Utah DOT MA Each guardrail run should function as intended - all posts, blockouts, panels, and connection hardware shall be in place. (UT DOT, 2004) * MM denotes criteria present in a maintenance manual; MA denotes maintenance assessment criteria. Table 2. Selected state transportation agencies with non-quantitative guardrail repair guidelines. of quantitative guidelines. For instance, w-beam guardrail repair is recommended when a “deep pocket in the rail line” exists, with no mention of a length or depth threshold. Other examples of guardrail damage requiring repair include “sec- tions torn loose from posts,” “rail section flattened,” or an “anchor at either end of a run broken loose.” 2.3.2 Published State Transportation Agency Guidelines for Damaged Barrier Repair The literature review included published guidelines from 26 U.S. state transportation agencies relating to the mainte- nance and/or performance assessment of longitudinal barrier. Of these 26 agencies, only 9 were found to have quantitative longitudinal barrier repair criteria (6 maintenance assessment

that were not classified as quantitative. The prevailing mainte- nance manual and maintenance assessment damage threshold is stated as “damage that affects the structural integrity of the barrier.” For maintenance assessment criteria, several agen- cies even rate barriers in terms of a percentage that is “func- tional” without specifically defining damage that impairs barrier functionality. Without an objective definition of the damage that affects barrier integrity, maintenance personnel tasked with evaluating barrier repair need may have signifi- cantly different interpretations of what damage impairs bar- rier functionality. The fact that the majority of state agencies employ this blanket statement without accompanying quan- titative guidelines underscores the importance of developing a better understanding of how quantifiable barrier damage correlates to subsequent impact barrier performance. Also evident from this literature review is the variation between maintenance manuals and maintenance assessment criteria even within the same jurisdiction. For instance, North Carolina had quantitative barrier repair guidelines in the main- tenance manual but no quantitative guidelines for mainte- nance assessment (see Table 2). It should be noted that these criteria for a given agency are not required to coincide as these manuals are typically developed independently. In addi- tion, maintenance assessment criteria are not necessarily used by maintenance personnel to justify barrier repair and may include factors other than the safety performance of the barrier in their scope. For all the published maintenance assess- ment manuals found in this study, however, functionality was a main component of barrier condition. Another observation from these published guidelines was that there was little dis- tinction between the repair thresholds based on barrier appli- cation, e.g., on the roadside or in the median. 2.3.3 Analysis of Survey Responses A total of 39 transportation agencies responded to the survey. From the United States, there were responses from 29 trans- portation agencies from the continental states as well as Hawaii and Puerto Rico. From Canada, there were responses from a total of 8 Canadian Provinces: Alberta, British Colum- bia, Manitoba, New Brunswick, Nova Scotia, Ontario, Prince Edward Island, and Quebec. Approximately 38 percent of the respondents (15 agencies: 11 U.S. States, 3 Provinces, and Puerto Rico) provided detailed information for guardrails within their respective jurisdictions. In total, these agencies provided an inventory in excess of 37,000 miles of longitudinal barrier (no distinction was made between roadside and median barriers). The strong-post w-beam barrier was the most fre- quent barrier type, accounting for roughly 60 percent of total barrier length by the responding state agencies. Excluding the two agencies that reported no use of strong-post w-beam (South Carolina and British Columbia), the average use of strong-post w-beam barriers was approximately 75 percent. Concrete, cable barrier, strong post thrie beam, and the weak post w-beam were ranked second through fifth, respectively, based on the responding agencies providing detailed barrier information. The distribution of barriers identified in this survey appears similar to that reported by Ray and McGinnis (1997). Note, however, that the Ray and McGinnis study did not request agencies to report barrier mileage. Approximately 60 percent of responding agencies (23 of 39) indicated the presence of specific guidelines for determining when a guardrail needs to be repaired. Of these 23 agencies, however, only 7 were classified as “quantitative” with 2 of these agencies previously identified through the literature review. In general, the quantitative guidelines resulting from the survey were similar to those found through the literature review. For the purpose of this study, the quantitative criteria found via the survey and literature review have been combined and shown in Tables 3–6. Tables 3–5 summarize the metal beam barrier cri- teria while Table 6 summarizes the criteria for cable barriers. Each criterion was classified based on the barrier component to which it refers, i.e., the rail element, the posts/blockouts, or the connections. For the rail element and post/blockout cat- egories, the criteria have been further classified into 3 general damage types: (1) deflection, (2) tearing/breaks and/or punc- tures, or (3) deterioration. The transportation agencies using each of these criteria are listed on the right hand side of the table and grouped into one of two categories: maintenance or maintenance assessment. Again, note that for the same agency, maintenance manual-based criteria and maintenance assess- ment criteria are not necessarily the same. The Ohio DOT, for instance, has quantitative criteria for both barrier mainte- nance and maintenance assessment; however, as indicated in the table, these criteria are not the same. Another exam- ple is the Indiana DOT that has quantitative maintenance assessment criteria, but the maintenance manual uses only a non-quantitative “functional/non-functional” criterion and thus was not included in the tables. Note that references for each agency’s barrier repair criteria appear next to the agency name. Current FHWA guidelines for metal beam barriers have been provided for reference and are the thresholds to distin- guish between the “minor damage” and “damaged but should function adequately under majority of impacts” categories. No FHWA guidelines were found for cable barriers. The majority of the criteria listed in the table are those used to distinguish between minor damage and damage that needs to be repaired (or results in a “deficient” rating in terms of maintenance assessments). Some agencies also have (or only have) criteria for severely damaged barriers; these criterion are marked with an asterisk. For metal beam barrier rail elements, the most prevalent quantitative criterion for repair was barrier deflection with a majority of agencies using the FHWA-endorsed 152 mm (6 inches) threshold. Maintenance assessment procedures 5

in Missouri, however, allow only a 76 mm (3 inches) deflec- tion threshold for guardrails. Even with severe metal beam barrier damage, there are variations; the California mainte- nance manual specifies 305 mm (12 inches) of rail deflec- tion while the North Carolina maintenance manual specifies 457 mm (18 inches). With respect to rail flattening, two states (Montana and Washington State) specify guardrail deficient if rail flattening is present even if the barrier was not deflected more than 152 mm (6 inches). The maintenance assessment procedures in Iowa were the only guidelines that prescribed specific thresholds for rail flattening: 50 and 30 percent of the cross-section thickness and height, respectively. For damage to posts, a majority of the agencies used a threshold of one or more broken or cracked posts. Two exceptions were Ohio and Indiana maintenance assessment procedures which pre- scribed two or more broken or cracked posts. For post deflec- tion, a majority of the agencies used horizontal distance out of alignment; a notable exception was Pennsylvania and Nova Scotia which use post angle. For metal beam barrier connec- tions, most maintenance assessment criteria rate a barrier as deficient if one or more bolts are missing while maintenance assessment in Wyoming specified 4 or more missing bolts. Interestingly, none of the quantitative maintenance criteria used a threshold for missing bolts. Similar variations can be found with respect to cable bar- rier repair/assessment criteria. The overall number of criteria pertaining to cable barriers, however, was substantially less than that of metal beam barriers. Notable differences include criteria for cable sag which varies from 38 mm (1.5 inches, Iowa maintenance assessment) to 51 mm (2 inches, Ontario maintenance manual) up to 152 mm (6 inches, Pennsylvania maintenance assessment). For broken posts, a majority of agencies used a threshold of one or more (Ohio, Quebec, and Montana) while Ontario uses 3 or more consecutive posts. In general, maintenance assessment criteria employed by Iowa were found to be the most quantitative and comprehensive with respect to both flexible and semi-rigid longitudinal bar- rier assessment. 6 Maintenance Maintenance Assessment Category Type Criteria Description FH W A (2 00 8) C al ifo rn ia (2 00 6) O hi o (20 05 ) N or th C ar ol in a (20 00 ) Qu eb ec (2 00 4) Io w a (20 04 ) M on ta na (2 00 2b ) O hi o (20 04 ) W as hi ng to n St at e ( 20 06 ) W isc on sin (2 00 4) Pe nn sy lv an ia (2 00 6) M iss ou ri (2 00 3) In di an a (20 06 ) W yo m in g (20 06 ) N ov a Sc ot ia (2 00 6) Fl or id a (20 07 ) Deflection > 76 mm (3 in.) X Deflection > 152 mm (6 in.) X X X X Deflection > 152 mm at any point in 3.6 m section X X X X * Deflection > 305 mm (12 in.) X * Deflection > 457 mm (18 in.) X Rail flattening > 50% thickness X X Rail flattening > 30% height X > 50% crushed X X > 50% torn X X Rail distortion > 25% of rail section length X Any rail flattening (even if <152 mm deflection) X X X X Rail height varies > +/- 51 mm (2 in.) from 706 mm (27 in.) standard height X Rail height varies > +/- 76 mm (3 in.) from 706 mm (27 in.) standard height X Rail height < 610 mm (ground to top of rail) X X Deflection Rail height > 762 mm (ground to top of rail) X Horizontal tear > 25 mm wide and 305 mm long X Any length vertical tear X * Any splits or tearing X X > 50% torn X Non-manufacturer hole in rail > 25 mm diameter X Tearing/Breaks & Punctures > 3 Non-manufacturer holes in rail X Rail Element Deterioration Any structural corrosion X X X * Maintenance criteria is used to indicate a threshold for severe barrier damage (e.g., immediate repair). X Agency uses the criteria to determine barrier repair need (maintenance column only) or barrier deficiency (maintenance assessment column only). Table 3. Summary of quantitative damaged barrier criteria: metal beam barrier rail elements.

7Maintenance Maintenance Assessment Category Type Criteria Description FH W A (2 00 8) C al ifo rn ia (2 00 6) O hi o (20 05 ) N or th C ar ol in a (20 00 ) Qu eb ec (2 00 4) Io w a (20 04 ) M on ta na (2 00 2b ) O hi o (20 04 ) W as hi ng to n St at e ( 20 06 ) W isc on sin (2 00 4) Pe nn sy lv an ia (2 00 6) M iss ou ri (2 00 3) In di an a (20 06 ) W yo m in g (20 06 ) N ov a Sc ot ia (2 00 6) Fl or id a (20 07 ) Deflection > 76 mm (3 in.) X Deflection > 152 mm (6 in.) X X X X Post angle > 15° angle from vertical X Post angle > 20° angle from vertical X * Deflection > 305 mm (12 in.) X * Deflection > 457 mm (18 in.) X 1 or more twisted/misaligned blockouts X X 3 or more continuous twisted/misaligned blockouts X X Deflection > 10% of blockouts twisted X 1 or more broken/cracked posts X X X X X X X X X 2 or more broken/cracked posts X X *3 or more broken posts X 1 or more missing blockouts X X X X Tearing/Breaks 3 or more continuous missing blockouts X X X 1 or more rotten posts X 2 or more continuous rotten posts X X X Rotten post ( > 50% cross section) X > 10% of posts/blockouts deteriorated or rotten X Posts & Blockouts Deterioration Any structural corrosion X X * Maintenance criteria is used to indicate a threshold for severe barrier damage (e.g., immediate repair). X Agency uses the criteria to determine barrier repair need (maintenance column only) or barrier deficiency (maintenance assessment column only). Table 4. Summary of quantitative damaged barrier criteria: metal beam barrier post and blockouts. Maintenance Maintenance Assessment Category Type Criteria Description FH W A (2 00 8) C al ifo rn ia (2 00 6) O hi o (20 05 ) N or th C ar ol in a (20 00 ) Qu eb ec (2 00 4) Io w a (20 04 ) M on ta na (2 00 2b ) O hi o (20 04 ) W as hi ng to n St at e (20 06 ) W isc on sin (2 00 4) Pe nn sy lv an ia (2 00 6) M iss ou ri (2 00 3) In di an a (20 06 ) W yo m in g (20 06 ) N ov a Sc ot ia (2 00 6) Fl or id a (20 07 ) Splice damage (< 32 mm of rail material left at any point around the bolt) X 1 or more missing/loose/damaged splice bolts X Loose/missing or damaged hardware X 1 or more missing bolts X X X X X X 1 or more posts separated from rail X X 4 or more missing/loose bolts in single section X Connections Integrity Loss *Bolts are missing or torn through rail element X * Maintenance criteria is used to indicate a threshold for severe barrier damage (e.g., immediate repair). X Agency uses the criteria to determine barrier repair need (maintenance column only) or barrier deficiency (maintenance assessment column only). Table 5. Summary of quantitative damaged barrier criteria: metal beam barrier connections.

For 27 different minor barrier damage types, respondents were asked to indicate whether the damage type would be repaired and the corresponding repair priority. A total of 33 respondents filled in this information in whole or in part; the remaining 6 agencies did not provide any information. Table 7 summarizes the responses by indicating the percent- age of agencies that would repair the particular guardrail damage. For each damage type, the number of respondents for which it is based has also been listed. Note that not every agency provided a repair indication for each damage type; in most cases, the agency did not provide a response or, in fewer instances, provided alternate responses (other than the yes/no specified by the survey instructions). There appears to be a consensus among respondents that post/rail deflection in excess of 152 mm (6 inches) and vertical rail tears need to be repaired. Splice damage, cable tension loss, damage to cables, soil erosion around posts, and bent or missing cable hooks had repair percentages in excess of 90 percent. There appears to be no particular consensus on what damage type does not need to be repaired. Rail deflection only and post/ rail defection less than 6 inches appear to be the least likely to be repaired with 50 and 27 percent repair percentages, respectively. A total of 34 agencies provided repair priority information for each damage type. Respondents were asked to categorize repair priority into one of 4 categories: (1) repair immediately, (2) repair as part of scheduled maintenance, (3) do not repair, and (4) at the discretion of maintenance personnel. Again, not all 34 agencies indicated repair priority for all damage types. On average, however, there were 27 respondents for each damage type. Figure 2 is a summary of the top 10 damage cat- egories based on the percentage of respondents indicating the damage should be repaired as soon as possible. Not surpris- ingly, post and rail deflections in excess of 152 mm (6 inches), rail tears, and damage to cables ranked as high-priority repairs. With the exception of erosion of soil around posts, there is very good agreement between these top 10 and the top 10 presented in Table 7. 8 Maintenance Maintenance Assessment Category Type Criteria Description C al ifo rn ia (2 00 6) O hi o (20 05 ) N or th C ar ol in a (20 00 ) Qu eb ec (2 00 4) O nt ar io (2 00 3) Io w a (20 04 ) M on ta na (2 00 2b ) O hi o (20 04 ) W as hi ng to n (20 06 ) W isc on sin (2 00 4) Pe nn sy lv an ia (2 00 6) M iss ou ri (2 00 3) In di an a (20 06 ) *Cable is on the ground X X X Top cable height varies > +/- 51 mm (2 in.) from 762 mm (30 in.) standard height X X Spacing between cables > 76 mm (3 in.) X Horizontal deflection > 76 mm (roadside cable barrier) X Horizontal deflection > 25 mm (median cable barrier) X Deflection Horizontal deflection > 152 mm (6 in.) X Any broken cable strands X Frayed cable X Tearing/Breaks * Broken cable X X X Any structural rust X Cable sag > 38 mm (1.5 in.) between posts X Cable sag > 51 mm (2 in.) X Rail Element Deterioration Cable sag > 152 mm (6 in.) X Deflection Post angle > 15° angle from vertical X 1 or more broken posts X X X 3 or more consecutive posts missing/broken X Missing first 2 posts adjacent to anchor(s) X Tearing/Breaks * 4 or more posts knocked down X Posts Deterioration Any structural rust X X Missing cable hooks (unsecured cables) X Damaged cable hooks X Connections Integrity Loss Corroded cable hooks (unsecured cables) X * Maintenance criteria is used to indicate a threshold for severe barrier damage (e.g., immediate repair). X Agency uses the criteria to determine barrier repair need (maintenance column only) or barrier deficiency (maintenance assessment column only). Table 6. Summary of quantitative damaged barrier criteria: cable barrier.

9Damage Type / Description % Agencies that would Repair # of Respondents Post/rail deflection > 6 in. (152 mm) 100 30 Rail tear (vertical) 100 28 Loss of tension (cable barrier) 96 25 Damage to cable 96 24 Erosion of soil around posts 96 23 Bent or missing hooks (cable) 95 22 Snowplow damage 95 19 Splice damage 92 26 Missing bolts/hardware 92 25 Cable sag 91 22 Rail tear (horizontal) 89 28 Missing blockout 89 28 Loose bolts/hardware 87 23 Mowing damage 83 18 Rail flattening 81 27 Post wood rot 81 21 Slope-related barrier lean 79 24 Tear in steel post 78 27 Bolt pulled-through rail 77 26 Twisted blockout 77 26 Insect damage 68 19 Rail/post corrosion or rust 67 18 Cracked wood post 64 22 Holes > 1 in. (25 mm) in rail 58 24 Rail deflection only 50 22 Post/rail deflection < 6 in. (152 mm) 27 22 Table 7. Agency guardrail repair priorities by damage type. 17 19 23 30 30 36 37 39 41 50 0 10 20 30 40 50 60 Missing bolts/hardware Missing Blockout Snowplow damage Cable Sag Loss of tension (cable barrier) Splice Damage Rail Tear (vertical) Rail Tear (horizontal) Damage to Cable Post/rail deflection > 152 mm Percentage of Respondents Indicating To Repair ASAP Figure 2. Damage type ranked by ASAP repair priority. With respect to known cases of a vehicle impacting a pre- viously damaged barrier, 32 of 39 respondents indicated no documented cases. Three other responding agencies did not provide an answer to the question while two agencies an- swered “unknown.” Only two agencies (Oklahoma and New Hampshire) indicated documented cases of a vehicle impact- ing a damaged barrier. In Oklahoma, the single case identi- fied a vehicle impacting a Truck Mounted Attenuator (TMA) that was in place (presumably in front of the damage section). In New Hampshire, the only details provided were that sec- ond impacts do not happen often. Two-thirds of responding agencies (26 of 39) indicated that more quantitative guidelines for the repair of guardrail would be beneficial. Eleven agencies (28 percent) indicated that more quantitative guidelines would not be beneficial to their organi- zation; however, only two (California DOT and Florida DOT) of these agencies reported quantitative barrier repair guide- lines. Of the remaining two agencies, one indicated that more

quantitative guidelines may be beneficial while the other indi- cated only if sufficient resources were available to comply with the more quantitative guidelines. The second agency expressed concern about the increased liability the agency would incur if unable to comply completely with the quantitative guidelines. 2.4 Discussion A review of the available literature and a survey of U.S. and Canadian transportation agencies support several impor- tant notions regarding the current longitudinal barrier repair practices and priorities amongst transportation agencies. First is the general lack of quantitative guidelines to assess the longi- tudinal barrier damage level and the associated need for repair. Combining the literature review and survey results, data was obtained from a total of 40 of 50 U.S. states and 8 of 10 Canadian Provinces (approximately 80 percent of the U.S. and Canadian transportation agencies). Only 13 States and 2 Canadian Provinces, less than one-third of the 48 transporta- tion agencies, had either quantitative barrier repair criteria or quantitative maintenance assessment guidelines for longitu- dinal barrier. For the remaining two-thirds of agencies, bar- rier repair and barrier assessment criteria usually required a determination of whether the barrier was “functional,” with no specific guidelines for making that assessment. The cur- rent FHWA guidelines, published in 2008, do provide some loosely quantitative guidelines for barrier repair; however, the guidelines appear to be founded on engineering judgment in- stead of a strong analytic foundation. In addition, the survey responses suggest that transportation agencies would see a benefit in more quantitative barrier repair guidelines. Second is the apparent variation between barrier assessment criteria, as present in maintenance assessment procedures, and those criteria used to determine the need for barrier repair, as prescribed in the maintenance manual. For thirteen agen- cies, information from both maintenance assessment proce- dures and corresponding agency maintenance manuals was available. Six agencies (Indiana, Iowa, Montana, Pennsylvania, Florida, and Washington State) had quantitative maintenance assessment criteria but lacked quantitative barrier repair crite- ria in the maintenance manual. Two agencies (California and North Carolina) had quantitative barrier repair criteria in the maintenance manual but lacked quantitative barrier assess- ment criteria. Ohio was the only agency that had both quan- titative barrier repair criteria and quantitative maintenance assessment criteria while the remaining four agencies (Texas, Tennessee, Virginia, and Kansas) had neither quantitative bar- rier repair nor quantitative maintenance assessment criteria. Although these criteria are not required to coincide, all of the maintenance assessment criteria found in this study were either largely or solely based on barrier functionality. At a minimum, the variations noted in maintenance criteria and maintenance assessment criteria warrant further investigation. Third, failure to promptly repair a damaged barrier may increase a transportation agencies legal liability. Crashes involv- ing vehicles impacting previously damaged barriers are found to occur in the field. A review of the available tort liability cases in the U.S. revealed that impacts into previously damaged bar- riers have occurred and have been litigated (Keller v. State of Illinois, 1982; Leonard Paxton v. Department of Highways, 1999; McDonald v. State of New York, 2002; Rosemary F. Woody v. Department of Highways, 1989; Volpe v. State of New York, 2000). Many of these cases, such as McDonald v. State of New York, were dismissed. However, it would seem advanta- geous, at least from a legal perspective, to have more quantita- tive guidelines for when to repair damaged barrier and pri- oritize damaged barrier sections. Interestingly, the survey respondents could provide almost no documented cases of vehicles impacting previously damaged barriers. All of these notions seem to point to the need for a better understanding of the effects of barrier damage on barrier performance. 2.5 Conclusions Based on the findings of the literature review and analysis of the survey responses, the following conclusions are drawn: 1. A majority of the current U.S. and Canadian transportation agency guidelines for longitudinal barrier repair lack quan- titative measures to evaluate the need for barrier repair. In most of these cases, the practice is to repair barriers if it is “non-functional” with no specific guidance on making that assessment. 2. There is a need for the development of more quantitative guidelines for longitudinal barrier repair that are based on a strong analytical foundation. This analytical foundation should include full-scale crash testing of damaged barrier, pendulum testing of damaged barrier sections, and finite element modeling of damaged barrier impacts. 3. Several state transportation agencies, including California, Iowa, Montana, Ohio, Washington State, North Carolina, Pennsylvania, Missouri, and Wisconsin, were found to have quantitative measures to rate or provide guidance on the repair of flexible and semi-rigid barriers. Even in these cases, however, there appears to be little connection between the criteria used to evaluate the condition of longitudinal barriers for the purpose of maintenance assessment and the criteria used by maintenance personnel to determine the need for barrier repair. As both criteria are based heav- ily on barrier functionality, these variations warrant fur- ther investigation. 10