Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information (2022)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

16 The state of practice presented in this chapter is based on the results of the national survey conducted as part of this synthesis report; a review of the state DOT manuals and documents that address load rating of bridges and culverts with missing or incomplete as-built information and are either available online or obtained as a result of the survey; the follow-up interviews; and some of the results of surveys conducted in earlier studies (as referenced in Chapter 2). The synthesis report authors developed a draft of the survey, which was sent to the topic panel for review and comment. The authors incorporated the panel’s feedback into a second revi- sion of the survey, which was implemented in Qualtrics, an online survey tool. The topic panel beta-tested the Qualtrics survey and provided final feedback to the research team. The survey is reproduced in Appendix A. The authors distributed the survey to members of the AASHTO Committee on Bridges and Structures (COBS). The committee members received an email from the University of Delaware that introduced the report team, explained the objectives of the survey and the report, and pro- vided a link to the survey (with a request that the member or an appointed designee complete the survey). If the COBS member decided to delegate the assignment, the email could be forwarded directly to the designee so that this person could complete it. The report team sent follow-up emails to those who had not started or completed the survey by the deadline. This follow-up prompted additional survey responses. Ultimately, the survey generated 46 responses, for a 92% response rate. General The first series of questions in the survey aimed to gather basic information about BCMI in the state inventories, general state procedures for load rating, and other related information. This section discusses these results. The responses from general Yes/No questions are detailed in Table 4. Distribution of BCMI The overwhelming majority of state DOTs (97.8%) report bridges in their inventory that have no plans or incomplete information (Q5). The survey asked states whether any bridges in their inventory were not state-owned (Q35): 93.2% of the responding states reported non–state-owned bridges in their inventory. The average percentage of non–state-owned bridges in agency inventories is 40.2%. As shown in Figure 1, 57.5% of respondents noted that they were responsible for load rating all of their non–state- owned BCMI, and 25% rated some of the non–state-owned BCMI (Q37). Some of the feedback C H A P T E R 3 State of Practice

State of Practice 17   during the follow-up interviews suggested that compared to state-owned bridges, non–state- owned bridges more commonly had missing information; as mentioned previously, Lequesne and Collins [6] recommended that FHWA notify local agencies to retain bridge plans for the life of the structure. e survey asked states to identify, from a list, the percentage of BCMI in their inventory (Q6): (1) less than 5%, (2) between and 5% and 10%, (3) between 10% and 20%, (4) between 20% and 40%, and (5) more than 40%. e results are summarized in Figure 2. Almost 32% of the respondents reported less than 5% of the bridges and culverts have missing or incomplete information, with just over half noting less than 10% and a modest 6.8% citing more than 40%. Using these percentages and the 2019 NBI, the total number of BCMI in the United States can be estimated. As reported by FHWA [57], the total number of bridges in the 2019 NBI was 617,084; the total for the 46 DOTs that responded to the survey question was 535,100. Using the low end of the range reported by each state and using 40% for the states that selected over 40%, 17.5% 57.5% 25.0% No Yes, all of them Yes, but only some of them Figure 1. Q37 Responses: Is the state responsible for the inspection/load rating of those bridges that are not state-owned? (n = 43). Question Yes (%) No (%) Q5: Does your state/agency have bridges in its inventory that have missing or incomplete as-built information? 97.8 2.2 Q35: Are there bridges in the state inventory that are not owned by the state? 93.2 6.8 Q9: Does your agency currently consider the condition of the deck in load rating bridges that have missing or incomplete as-built information? 54.5 45.5 Q10: Does your agency currently consider the condition of the substructure in load rating bridges that have missing or incomplete as-built information? 61.4 38.6 Q13: Does your agency have a formal procedure in place for load rating bridges that have missing or incomplete as-built information? 61.4 38.6 Q14: Is the procedure documented? 77.8 22.2 Q17: Does your agency use information about the current physical condition of the bridge to aid in the load rating of bridges that have missing or incomplete as- built information? 97.7 2.3 Q18: Does your agency use historical information and information about bridges with plans to aid in the load rating of bridges that have missing or incomplete as- built information? 84.1 15.9 Table 4. Survey results for general questions.

18 Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information a low estimate of the total number of BCMI for the states that responded to the survey is 51,000. Instead using the high end of the range reported by each state and applying 40% for the states that selected over 40%, a high estimate is 97,000. e low and high estimates yield an average range for number of BCMI, as a percentage, for the responding states of 9.5% to 18.1%. Project- ing to the full 2019 national inventory, the estimate is 58,600 to 111,700 BCMI nationwide. is estimate exhibits a high level of uncertainty, but it gives some idea of the scope of the current situation with BCMI in the country. Types of BCMI and Types of Missing Information e survey asked states to identify the types of bridges in their inventories that are missing plans or information (Q7). e list of options included (1) timber, (2) steel truss, (3) rolled steel beam/girder, (4) reinforced concrete slab, (5) reinforced concrete culvert, (6) reinforced con- crete beam/girder, (7) prestressed concrete beam/girder, (8) masonry, (9) concrete encased steel beam/girder, (10) built-up beam (riveted, welded, bolted), and (11) other. e results are detailed in Table 5. As a group, concrete bridges and culverts were the most common type of BCMI, followed by concrete slabs and concrete beam bridges. Just over 80% of the states 31.8% 20.5% 27.3% 13.6% 6.8% <5% 5-10% 10-20% 20-40% Over 40% Figure 2. Q6 Responses: What percentage of the agency’s bridges are missing as-built information? (n = 46). Bridge Type % Timber 43.2 Steel truss 38.6 Rolled steel beam/girder 40.9 Reinforced concrete slab 68.2 Reinforced concrete culvert 81.8 Reinforced concrete beam/girder 75.0 Prestressed concrete beam/girder 56.8 Masonry 36.4 Concrete encased steel beam/girder 29.5 Built-up (riveted, welded, bolted) steel 29.5 Other 25.0 Table 5. Q7 Responses: Percentage of agencies that reported bridge types with missing or incomplete as-built information (n = 44).

State of Practice 19   reported concrete culverts in their inventory with no plans or incomplete information. Respon- dents also could select “other” bridge types in their inventories with missing plans or infor- mation. In total, approximately 25% of the responding states reported other types of bridges beyond those listed (e.g., concrete arch, steel culvert, reinforced concrete channel, railroad car, Bailey, and pontoon). Asked to report the percentage of each type of BCMI in their inventory, few states did, so these results are not addressed here (although the raw data are included in Appendix B). To more completely understand the types of information described as missing, the survey asked respondents to select all applicable types from a list (Q8). The most common responses noted a lack of drawings (Table 6). Information on Rating Approaches Asked whether the agency considered the deck in load rating BCMI (Q9), 54.5% of respon- dents reported that they do (Table 4). When the question was whether the agency considered the substructure in load rating BCMI (Q10), 61.4% responded affirmatively (Table 4). Of the states that responded, 93% acknowledged using the AASHTO MBE [2] (Q11). The majority of states (59.1%) employ the LFR method, with 36.4% using LRFR and just under 5% ASR. (Note that this last question, Q12, was not well constructed because most states likely use more than one procedure, but the question did not allow more than one answer). The next two survey questions were particularly important in establishing the current state of practice for load rating BCMI and in differentiating approaches. The survey questioned states about whether they have a formal procedure for load rating BCMI (Q13): 61% do, but the rest do not (Table 4). States that answered Yes to this question were then asked whether the procedure was documented (Q14), and more than three-quarters (77.8%) answered Yes (Table 4). Lequesne and Collins [6] observed a very similar response, albeit for a smaller sample of states. The survey respondents were also asked to provide either the document or a link to the document containing the procedure. If the respondent answered No to the formal procedure question (Q13), they were asked to describe their procedure (Appendix B includes these responses). An overwhelming majority of the states (97.7%) reported using information about the current physical condition of the bridge to aid in load rating BCMI (Q17) (Table 4). A majority (84%) also refer to historical data and information about bridges with existing plans to support load rating BCMI (Q18) (Table 4)—for example, as Gearhart [18] noted, plans and specifications for similar bridges built in the same era, historical design guidelines, inspection reports, log books, mill certifications, and county records. Type of Missing Information % Lack design drawings 79.9 Lack as-built drawings 79.9 Lack shop drawings 59.9 Plans unreadable/incomplete 46.5 Missing material properties 31.0 Missing rebar size, quantity, and location 35.5 Missing geometric information 4.48 Other 6.54 Table 6. Q8 Responses: Percentage of states that reported each as a most common type of missing or incomplete information (n = 45).

20 Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information Testing The survey asked states to report on the various technologies and testing methods that aid in load rating BCMI. The survey addressed NDT, destructive testing, diagnostic load testing, and proof load testing. Just about half of the responding states noted previously using some type of NDT (Q19), destructive testing (Q21), and diagnostic load testing (Q23) to assist in load rating BCMI; however, less than 20% indicated that they conducted a proof load test for load rating BCMI (Q29). Of those DOTs that have not used diagnostic or proof load testing in the past, more than 80% reported they are not likely to use these technologies for load rating BCMI in the future (Q28 and Q34, respectively). These results are discussed in detail in this section. Nondestructive and Destructive Testing A little over half (56.8%) of the states employ NDT to support load rating BCMI (Q19) (Table 7). Those states answering Yes (n = 26) were asked to indicate the NDT technologies that they typically use, choosing from a list of 20 NDTs (Q20). Among the most commonly selected NDTs were chain drag, GPR, dye penetrant, infrared thermography, ultrasound, and profometer rebar locator and cover meter (Table 8). Next, the survey requested that states describe their use of destructive testing methods, which referred primarily to taking material samples from a bridge for laboratory testing; the list included concrete cores, metal samples, wood cores/samples, and other. Just under half (45.5%) reported the use of some form of destructive test (Q21) (Table 7). The most common methods included concrete cores and metal samples (Q22) (Table 9). Question Yes (%) No (%) Maybe (%) Q19: Does your agency use non-destructive testing technologies to gather information about your bridges that have missing or incomplete as-built information? 56.8 43.2 Q21: Does your agency use destructive testing technologies to gather information about your bridges that have missing or incomplete as-built information? 45.5 54.5 Q23: Are you aware of your agency ever using diagnostic load testing as a tool for gathering more information about your bridges that have missing or incomplete as-built information? 52.3 47.7 Q25: Do you expect to continue to use diagnostic load testing as a tool in load rating of bridges that have missing or incomplete as- built information? 39.1 8.7 52.2 Q26: Does your agency follow or have a documented procedure for diagnostic load testing? 26.1 73.9 Q28: Does your agency have plans to use diagnostic load testing as a tool for assisting in the load rating of bridges that have missing or incomplete as-built information? 14.3 85.7 Q29: Are you aware of your agency ever using proof load testing as a tool for gathering more information about your bridges that have missing or incomplete as-built information? 18.2 81.8 Q31: Do you expect to continue to use proof load testing as a tool for assisting in the load rating of bridges that have missing or incomplete as-built information? 50 25 25 Q34: Does your agency have plans or would like to use proof load testing as a tool for assisting in the load rating of bridges that have missing or incomplete as-built information? 19.4 80.6 Q32: Does your agency follow or have a documented procedure for proof load testing? 37.5 62.5 Table 7. Survey results for testing questions.

State of Practice 21   Diagnostic and Proof Load Testing The survey polled the states on their use of diagnostic testing and proof load testing, first asking whether they were aware of the agency ever using diagnostic testing as a tool for gathering more information about BCMI (Q23), with the following responses: • 52.3% answered Yes (Table 7). Those states then were asked to select from a list the approx- imate number of bridges in their inventory that had been tested (Q24), with choices including 1 to 3, 4 to 6, 7 to 10, 11 to 20, and more than 20. Just over half reported that they had tested 1 to 3 bridges; however, approximately 22% tested more than 20 bridges (Figure 3). Of those responding states that used diagnostic load testing, 39% expected to continue to use load testing, 52% may continue, and almost 9% did not expect to continue (Q25) (Table 7). In addition, the vast majority of these states did not have a documented procedure for diag- nostic load testing (Q26) (Table 7). • 47.7% answered No (Table 7). These states were asked whether they had plans to use diag- nostic testing in the future (Q28): the overwhelming majority answered No (Table 7). In follow-up interviews, one state noted that it had conducted diagnostic load tests of a handful of bridges—but not for load rating a BCMI. This state admitted that it would not look to use diagnostic testing for this purpose. Another state had conducted between 15 and 20 diagnostic tests over the years for bridges with available plans, but under special circum- stances that justified the load test. This state was in the process of transferring the technology Destructive Testing Type % Concrete cores 95 Metal samples 80 Wood cores/samples 35 Other 15 Table 9. Q22 Responses: Percentages of states that reported using these destructive technologies (n = 20). Nondestructive Testing Type % Sounding (chain drag, hammer tap) 68 Schmidt hammer 44 Dye penetrant 56 Pullout test 24 Profometer (rebar locator and cover meter) 52 Ground-penetrating radar (GPR) 68 Impact-echo 44 Infrared thermography 52 Impulse response (including crosshole logging) 32 Ultrasound 48 Half-cell potential 24 Electricity resistivity 24 Linear polarization resistance 4 Magnetic flux leakage 4 Eddy current 12 Magnetic particles 36 Radiography (including backscatter and tomography) 12 Air permeability 8 Penetration resistance 20 Moisture content 16 Other 12 Table 8. Q20 Responses: Percentage of states that reported using these nondestructive technologies (n = 25).

22 Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information to state forces, which would lower the cost, and hoped to employ diagnostic and proof load testing on BCMI more oen in the future. A similar sequence of survey questions addressed the use of proof load testing. Respondents were asked whether they were aware of the agency ever using proof load testing as a tool for gathering more information about BCMI (Q29), with the following responses: • 18.2% answered Yes (Table 7). Of the eight states that used proof load testing, six reported conducting 1 to 3 proof load tests, with Maryland at between 11 and 20 proof load tests and Florida at over 20 proof load tests (Q30) (Figure 4). Four of the eight states expected to continue performing proof load tests; two said they might; and two did not expect to continue (Q31) (Table 7). Of the eight states that conducted proof load tests, ve did not have a documented procedure for conducting the test, but the rest did (Q32) (Table 7). • 81.8% answered No (Table 7). ese states were asked whether they had plans to use proof load testing in the future (Q34), and 80.6% responded No (Table 7). Follow-up interviews with a few states indicated hindrances to using proof load testing, including cost, lack of familiarity with the process, lack of experience, and logistics of the needed higher loads. 52.2% 17.4% 8.7% 21.7% 1 to 3 4 to 6 7 to 10 11 to 20 More than 20 Figure 3. Q24 Responses: Approximately how many bridges that have missing or incomplete as-built information have been diagnostically load tested for the purpose of load rating? (n = 23). 75.0% 12.5% 12.5% 1 to 3 4 to 6 7 to 10 11 to 20 More than 20 Figure 4. Q30 Responses: Approximately how many bridges that have missing or incomplete as-built information have been proof load tested for the purpose of load rating? (n = 8).

State of Practice 23   The final survey questions (Q38 and Q39) asked respondents to provide examples of load ratings performed in their state for one or more BCMI. In total, 13 different states described 24 examples, and seven of these evolved into the case examples in Chapter 4. National Bridge Inventory Coding Before discussing procedures, it is important to note how load rating procedures are coded in the NBI and how load ratings would typically be coded for bridges and culverts with missing or incomplete as-built information. Items 63 and 65 of the NBI record denote the methods for determining operating rating and inventory rating, respectively, as FHWA observed in 1995 [58]. The various methods are coded by a number between 0 and 8 or a letter between A and F. Relevant to the discussion here are codes 0, 4, and A through F. Item 63 or 65 coded as 0 denotes “engineering judgment.” As specified by FHWA in 2011 [59]: Code 0 is to be used when the load rating is determined by field evaluation and documented engineering judgment, typically done when plans are not available or in cases of severe deterioration. Field evaluation and engineering judgment ratings must be documented. A code of 4 for Item 63 or 65 indicates “load testing.” The A through F assigned ratings can be utilized when all of the following conditions are met, as cited by FHWA in 2011 [60]: (1) The bridge was designed and checked using either the AASHTO Load and Resistance Factor Design (LRFD) or Load Factor Design (LFD) methods to at least HL-93 or HS-20 live loads, respectively; and (2) The bridge was built in accordance with the design plans; and (3) No changes to the loading conditions or the structure condition have occurred that could reduce the inventory rating below the design load level; and (4) An evaluation has been completed and documented, determining that the force effects from State legal loads or permit loads do not exceed those from the design load; and (5) The checked design calculations, and relevant computer input and output information, must be acces- sible and referenced or included in the individual bridge records. The A through F codings are assigned to an MS18 loading and depend on the design proce- dure (ASD, LFD, LRFD) and whether the loading is reported as a metric ton or a rating factor. Per the five criteria that must be met, an A through F coding could be applied to a BCMI in some instances, but they are also employed in assigning ratings for many bridges that are not missing information. Procedures The survey asked states to describe their procedures for load rating BCMI, either directly in the survey or indirectly through a reference to the location of the documented procedure. The report team obtained and reviewed all of these procedures. As a starting point, the AASHTO MBE [2] is nearly always referenced by a responding state as a foundation of its procedures. As MBE Section 6.1.4 (Bridges of Unknown Structural Components) states: For bridges where necessary details, such as reinforcement in a concrete bridge, are not available from plans or field measurements, a physical inspection of the bridge by a qualified inspector and evaluation by a qualified engineer may be sufficient to establish an approximate load rating based on rational criteria. Load tests may be helpful in establishing the safe load capacity for structures. A concrete bridge or concrete bridge length culvert with unknown details need not be posted for restricted loading if it has been carrying normal traffic for an appreciable period and shows no distress. The bridge shall be inspected regularly to verify satisfactory performance. A bridge may also be load tested to determine its capacity.

24 Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information The commentary to this section adds: Knowledge of the live load in the original design, the current condition of the structure, and live load history may be used to provide a basis for assigning a safe load capacity. Bridge Owners may consider non- destructive proof load tests to establish a safe load capacity for such bridges. On the basis of the information in the surveys, documents obtained from the states, and follow-up interviews, the procedures for load rating BCMI seemingly fall into one of the following five categories: 1. Experience-based rating (EB) 2. Field measurement and historical data (FH) 3. Rational evaluation rating (RE) 4. Design load rating (DL) 5. Field testing (FT) This report specifically avoids using “engineering judgment” and “assigned” in the names of these five procedure categories to prevent confusion with the terms used in coding NBI Items 63 and 65. When finally reported, a load rating derived by using any of these five types of procedures would receive one of the NBI codes previously discussed. The boundaries between these categories are not strict, and many states utilize aspects of more than one approach. Table 10 summarizes the procedures employed by the state DOTs. If the state DOT’s procedure is documented and publicly available, the source is cited in the column next to the state in Table 10. The five types of procedures are described in the fol- lowing sections. State Doc1 EB FH RE DL FT State Doc1 EB FH RE DL FT Arizona X Missouri X X X Alabama X X Montana X X Arkansas NP X X Nebraska X X California NP X Nevada X X Colorado [61] X X X New Jersey [62] X Connecticut X New Mexico X X Washington, DC X New York NP X Delaware NP X X North Carolina [63] X Florida [64] X X North Dakota [65] X X Georgia X Ohio X X Hawaiʻi X Oklahoma X X Idaho [66] X X Oregon [67] X X Illinois [68] X X Pennsylvania [69] X Indiana X Rhode Island [70] X X Iowa X X South Carolina [71] X Kentucky [72] X X South Dakota X X Louisiana [73] X X Tennessee X X Maine X X Texas [74] X Maryland NP X Utah [75] X Massachusetts [76] X Virginia [77] X X Michigan [78] X X Washington [79] X Minnesota [80] X X X X West Virginia NP X Mississippi X Wyoming X X X 1 Shaded cell: procedure is undocumented. NP: procedure is documented but is not publicly available. Bracketed number: procedure is documented in the numbered citation in the References list. Table 10. Agency procedures.

State of Practice 25   Experience-Based Rating (EB) The actual process of rating based on “engineering judgment” is not well defined. The authors have not found any details anywhere on exactly what this means; however, the term usually connotes using one’s experience and knowledge of engineered systems as the basis for some decision or recommendation. More specifically, a rating is based on the rating engineer’s experience and knowledge of the condition of the bridge, the traffic, and the ratings of other similar structures. In some cases, a team of engineers collectively determines the ratings. For example, the Maine DOT relies on a posting committee to recommend load rating and posting of BCMI. The com- mittee consists of two members from bridge maintenance, two members from bridge manage- ment, two members from bridge design, one member from traffic, and one junior engineer. All of the members (except the junior engineer) have more than 20 years of experience. This wealth of experience and varied perspectives are brought to the table and feed into the final decisions about rating and posting of BCMIs. In another state, a bridge inspector and the load rating engineer work together to arrive at the judgment-based ratings. Field Measurement and Historical Data (FH) All indications are that load rating is straightforward for steel and timber bridges that are missing information, and all states using this procedure handle such bridges similarly. Field measurements establish the overall dimensions of the bridge and the member cross-sections. Material properties are assumed based on historical data or on NDT or destructive coupon tests of the materials. With this information, a state can construct a model of the bridge, and the load rating then is completed as it would be for a bridge with plans. For other BCMI, simple field measurements and even NDT cannot generate all of the infor- mation needed to compute a theoretical load rating. Examples of this type of BCMI include reinforced concrete and prestressed concrete bridges and culverts with missing reinforcement details and strand patterns, masonry bridges, and earth-infilled concrete arches. For these bridges, field measurements can help determine the overall size and geometry of the struc- ture. Historical records and data can be used to estimate material properties, and knowledge of the design procedures in place when the bridge or culvert was constructed can be applied to guesstimate or bound reinforcement details. With this information, a state can construct a model and compute a load rating. Furthermore, a comparison is possible between the rating for the BCMI and the ratings from a similar era for other similar structures with available calculated ratings. In follow-up interviews, Georgia and Louisiana each referred to using field measurements and “standard plans” as a starting point for load rating a BCMI. If standard plans can be identi- fied that closely match the BCMI in question, they are used to conduct the load rating. Rational Evaluation Rating (RE) A rational evaluation rating is based on the condition of the bridge and on an assumed load. The inventory and operating load ratings are based on the bridge’s NBI condition ratings (CRs). The details of the procedure vary from state to state. The load ratings may decline with any decrease in superstructure, substructure, deck, or culvert CR, or they may drop only when the CR falls below a certain level, usually 5 or 4. These cases are categorized as REs. Idaho uses an RE procedure; the relevant section from the Idaho inspection manual is shown in Figure 5. The load ratings are based on an HS-20 design load, and the manual describes the

26 Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information Figure 5. Rating procedure from Idaho Manual for Bridge Evaluation (Section 6.1.4) [66].

State of Practice 27   Figure 6. Rating procedure from Illinois Structural Services Manual 2017 (Section 4.4.3.3) [68]. steps to be taken if the NBI condition rating is 4 or lower. e Illinois procedure (Figure 6) is based on the premise that the HS-20 inventory and operating load ratings are proportional and can be assigned as a function of the condition of the bridge. e North Carolina procedure (Figure 7) load rates BCMI using a table that takes into account the year the bridge was built (assuming an H15 design load for bridges built before 1950 and an H20 design load for bridges built during or aer 1950) and the span length of the bridge. Design Load Rating (DL) e DL procedure involves either knowing the design load for the bridge or estimating it based on the known (or assumed) age and design loads in use when the bridge was built. If IR = 1.0, then the capacity minus dead load (the numerator in Equation 1.1) eect is assumed to equal the live load moment caused by the design load. e rating for other vehicles is then computed as the ratio of the design live load moment divided by the live load moment of

28 Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information Figure 7. Rating procedure from North Carolina for load rating reinforced concrete bridges with missing information (from a memorandum dated April 1, 2016, from T.K. Koch to Project Engineers and Project Design Engineers; North Carolina DOT [63]). the vehicle of interest. e operating rating is then proportioned from the inventory rating; depending on the condition of the bridge, factors may be applied to reduce the ratings. All of these cases are categorized as DL. Utah uses a DL procedure, breaking down the process into four categories: (1) the design load is known from original documents; (2) the design load is assumed based on the average daily truck trac (ADTT) and the era when the bridge was built; (3) load testing and hand calcu- lations are used; and (4) all other cases. Except for ratings determined by a load test, ratings are multiplied by a condition factor that depends on the NBI condition rating of the bridge. e Utah procedure is described in Figure 8.

State of Practice 29   California also uses a DL procedure, with the major steps noted in Figure 9. If RF < 1 based on a capacity derived from the design load, then an alternate calculation can be performed where the capacity at the operating level is determined by the maximum moment for AASHTO legal loads. Inventory ratings are then proportioned downward from the operating rating. e proce- dure does not specify precisely how to handle cases if the bridge shows distress or deterioration but does instruct the rater to consult with the load rating branch. Field Testing (FT) A handful of states in the survey reported using a full suite of NDT, destructive material test- ing, diagnostic load testing, and proof load testing as tools to assist in load rating BCMI. ese states seemingly adopted the technologies as part of their standard practice, as documented in their policies and procedures. Without question, eld testing comes at the expense of additional time, sta-hours, and cost. In some cases, DOTs are more familiar with some of the newer technologies and therefore are more comfortable with them and more inclined to use them. Some states own their testing equipment and maintain in-house forces to conduct the tests while other states rely solely on consultants to perform the testing. All of these factors weigh into the decision of whether testing is used to assist in the load rating. Ultimately, the additional expense and eort must be balanced against the benet (positive or negative) that may be derived from the more detailed informa- tion obtained through testing and the eventually resulting load rating. As documented in its survey, Florida conducted more than 20 diagnostic load tests and more than 20 proof load tests. Such testing is performed by in-house forces, namely the Structures Research Center, following the Florida Department of Transportation Bridge Load Rating Manual [64]: It is anticipated that the Structures Research Center will perform a minimum of three (3) load tests each scal year. Within 60 days of completion of the load test, the Structures Research Center will send the load test report to the District Structures Maintenance Engineer and the State Load Rating Engineer. Within 14 days of receipt, the District Structures Maintenance Engineer will update the BrM/BMS database with the results of the load test report. Figure 7. (Continued).

30 Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information Figure 8. Utah DL procedure (from Utah Bridge Management Manual, Section 4.3.2.3.7 [75]).

State of Practice 31   Figure 9. California DL-based procedure for concrete bridges with unknown reinforcement (from survey response).

32 Load Rating of Bridges and Culverts with Missing or Incomplete As-Built Information New York reported conducting more than 20 diagnostic load tests and one to three proof load tests. A section on field load testing in the state’s internal procedure document discusses diagnostic and proof load testing. This internal procedure also includes a section on documen- tation of results and a list of relevant references. Oregon reported that it performed one to three diagnostic load tests and one to three proof load tests. A section in the ODOT LRFR Manual [67] addresses improving low rating factors. Among the approaches explicitly described as “complex refinements” are the following: Use actual material properties in the load rating analysis, which requires a minimum of 3 samples (concrete cores or steel coupons) of the bridge materials to be taken from the existing structure and tested in a materials lab. . . . Perform load testing on the bridge to calibrate actual live load distributions, determine actual dynamic impact, verify composite action up to a given load, or proof test the bridge to establish a target live load capacity. Summary For this synthesis report, a survey of state DOTs was conducted to determine the state of practice for load rating bridges and culverts with missing or incomplete as-built information. The report team received a total of 46 responses. The survey results showed that nearly all DOTs have BCMI and that the majority of those bridges and culverts are constructed of con- crete. Approximately half of the responding DOTs reported using some type of nondestructive or destructive technology to assist in load rating BCMI, and about half cited diagnostic load testing while just under 20% noted the use of proof load testing. The report team categorized the procedures for load rating BCMI into five categories. The most common procedures are experience-based rating (EB) and field measurement and historical data (FH), followed by rational evaluation rating (RE) and design load rating (DL), which require some additional work and calculation beyond the EB and FH methods. Only a few DOTs reporting using the final category of field testing (FT), which can lead to a more accurate and refined load rating, but at additional cost to the bridge owner.