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.
18 Overview A survey of state practices was conducted to document current state DOT practices and experience regarding collecting and ensuring the accuracy of element-level data. The syn- thesis also examined how the state DOTs are using bridge element data in asset management decision-making. The survey questionnaire was distributed to bridge management contacts in all 50 state DOTs and the District of Columbia DOT. All 51 invited DOTs participated in the survey (a 100% response rate). This chapter summarizes the findings from the survey of practices. The information is pre- sented in several formats, including both tables and figures as appropriate. A copy of the elec- tronically distributed survey questionnaire is provided in Appendix A, and individual survey responses are detailed in Appendix B. In the discussion, the percentage of agency responses and counts are reported. Because of the 100% response rate, the discussion refers to the state DOTs (i.e., for this survey, the 50 state DOTs and the District of Columbia DOT), instead of the respondent DOTs. Survey Content The survey questions were organized into the following five categories: ⢠Contact Information. The first two questions gathered basic information on the respondentâs DOT, name, email, and phone number. The answers were used to identify the responses by each DOT and to communicate with the respondents when necessary. ⢠State of the Practice in Bridge Element Data Collection. Six questions requested informa- tion on the type of element data collected (including defects, environment, ADEs) and the techniques used for collecting element data (including NDEs). ⢠Quality Control and Quality Assurance for Bridge Element Data. Two questions asked for information on the agencyâs self-assessment of its QC and QA processes and element data quality. ⢠Performance Measures and Models. Six questions addressed information on performance measures, cost and deterioration models, and decision rules or trees that are based on ele- ment data. ⢠Use of Bridge Element Data in Asset Management. Three questions requested information on how the state DOT utilized element data in asset management decision-making. C H A P T E R 3 State of the Practice
State of the Practice 19  State of the Practice in Bridge Element Data Collection Initial questions in the survey examined the current practice and DOT preferences in bridge element data collection. All state DOTs reported that they are collecting data for NBEs, aligned with the requirement to have begun collecting element-level data by October 1, 2014, per 23 U.S.C. 144(d)(2). In addition, 88% of the state DOTs reported that they are collecting data for BMEs, while 67% are also collecting data for ADEs (Figure 3-1). Actually, all state DOTs report BMEs (excluding approach slabs and protective systems for deck reinforcing steel) to FHWA, per its spe cification for NBI bridge elements (FHWA 2013). The discrepancy in the survey responses may be a result of confusion about BMEs included in, and excluded from, FHWA data submissions. Seven DOTs (in California, Delaware, Oklahoma, Louisiana, Kentucky, Michigan, and New York) shared their bridge element inspection manuals in response to a later survey question. Examples of ADEs from the Delaware, Oklahoma, and New York DOTs are listed in Table 3-1. Figure 3-1. Number of state DOTs that collect data for each element type. 51 (100%) 45 (88%) 34 (67%) N A T I O N A L B R I D G E E L E M E N T S ( N B E s ) B R I D G E M A N A G E M E N T E L E M E N T S ( B M E s ) A G E N C Y-D E V E L O P E D E L E M E N T S ( A D E s ) Table 3-1. Three examples of state DOT ADEs. Oklahoma DOT ADEs ⢠Soffit of concrete decks and slabs ⢠5-foot open girder ends ⢠Girder or beam under construction joint ⢠Stringer under construction joint, 5-foot stringer ends ⢠Steel truss (overhead) ⢠Steel truss (deck) ⢠Pier beam ⢠Post-tensioned concrete pier beam ⢠Curved girder diaphragm/cross-frame ⢠Wingwall ⢠Precast arch culvert New York DOT ADEs ⢠Erosion or scour ⢠Stream hydraulics ⢠Sidewalk ⢠Curb ⢠Secondary members ⢠Steel beam end ⢠Backwall ⢠Abutment pedestal ⢠Pier pedestal ⢠Wingwall ⢠Headwall Delaware DOT ADEs ⢠Sidewalks ⢠Curbs ⢠Medians ⢠Drains, downspouts, scuppers ⢠Deck/slab under fill ⢠Diaphragms ⢠Steel live load anchor assembly ⢠Filled arch ⢠Headwall ⢠Sheeting ⢠Backwall ⢠Strut ⢠Timber pier slab ⢠Wingwall/retaining wall ⢠Wingwall/retaining wall cap ⢠MSE wall ⢠Steel protective coating: galvanizing system ⢠Soffit ⢠Erosion ⢠Electrical system ⢠Mechanical system
20 Bridge Element Data Collection and Use Examination of the ADEs from these shared manuals shows five general trends in the selec- tion of these elements. (1) NBEs (e.g., steel girders) are commonly divided into sub-elements to better define sub-element locations, cause and effects on bridge, priority of response, and feasible actions. These sub-element quantities are then rolled up into one element when reported to FHWA. Sub-elements such as girder ends may require more frequent repairs, and associated ADEs make it easier to track their conditions. (2) Multiple ADEs are commonly used for elements with different materials or design types (e.g., box culvert vs. pipe culvert), such as wearing surfaces (e.g., thin overlay and asphaltic overlay) and still protective coating systems (e.g., galvanizing system and weathering steel). (3) ADEs may be created for custom elements that are not neces- sarily covered in NBEs or BMEs, such as headwalls, wingwalls, and movable bridge components. (4) ADEs may be assigned for roadway elements (e.g., curbs, medians, or sidewalks) around bridges. (5) ADEs may be used for specifying additional agency-defined defects (e.g., wearing surface rutting) and for separating a defect into multiple categories (e.g., patch, spall, and delamina- tion noted as three defects). These ADEs are reminiscent of CoRe smart flags (AASHTO 1997) that are now mostly incorporated into the defects in the MBEI (AASHTO 2019). Most state DOTs also collect data on element defects, but such collection is optional. Spe- cifically, 39 state DOTs (76%) reported collecting element defect data (Figure 3-2) while less than half (22) of the state DOTs (43%) indicated that they collect data on environments (Fig- ure 3-3). Defect-level data collection includes the assignment to CSs of defect type and quantity Yes 22 (43%)No 29 (57%) Figure 3-3. Data collection on environments, by percentage of state DOTs. Yes 39 (76%) No 12 (24%) Figure 3-2. Data collection on element defects, by percentage of state DOTs.
State of the Practice 21  (e.g., cracking, delamination, exposed rebar, efflorescence, and damage). Data on environ- ments identify exposure to varying environmental factors (e.g., traffic volume, exposure to road salt, or changing climate) that may lead to different rates of deterioration of bridge elements. The survey included a question on whether the state DOT utilizes NDE techniques for bridge element inspections. Of the state DOTs, 84% reported that they use NDE techniques in bridge element inspections while 16% do not (Figure 3-4). As a follow-up, the state DOTs were then asked to note particular NDE techniques that they use and how often they employ these techniques in bridge element inspections. The following NDE techniques were specified in the survey: ⢠Chain Drag. This method involves dragging a chain across the concrete bridge deck surface while listening for changes in the acoustic response to locate and measure the extent of spalls and delaminations (Scott et al. 2003, Hooks and Weidner 2016). This method provides a qualitative assessment but is rapid and inexpensive. ⢠Ground Penetrating Radar (GPR). This method is commonly used for infrastructure inspec- tions involving concrete structures (Senin and Hamid 2016). GPR can be used to assess sub- surface conditions and to monitor concrete infrastructures such as bridge decks and building components (Barrile and Pucinotti 2005; Alani, Aboutalebi, and Kilic 2013; Dinh and Zayed 2016). Common uses of this technique for bridge inspections include detection and charac- terization of concrete deterioration in bridge decks; presence, pattern, and depth of structural steel reinforcement in the deck; estimation of deck thickness; and identification of anomalies (Hooks and Weidner 2016). ⢠Infrared Thermography (IRT). This method is used to identify the delaminated areas and depth of delamination by detecting and analyzing differences in surface temperatures (Maser and Roddis 1990; Rens, Nogueira, and Transue 2005). ⢠Impact-Echo. This method is based on the analysis of longitudinal stress waves generated by the impact of ball bearings on the concrete surface (Rens, Nogueira, and Transue 2005). The impact-echo method can be employed to detect delamination in concrete slabs (with or without overlays), characterize surface-opening cracks, measure deck thickness, and analyze concrete overlay debonding (Lin, Sansalone, and Poston 1996; Sansalone 1997; Sansalone and Streett 1997; Tawhed 2002). ⢠Electromagnetic. These methods are typically used to detect flaws or stresses and to test ferro- magnetic materials (e.g., steel bridges) to determine structural parameters (e.g., stress, strain, microstructure) and related flaws. Other magnetic methods (e.g., magnetic particle testing, magnetic flux examination, and eddy current testing) can be applied to identify flaws such as cracks, voids, corrosion, and section loss (Rens, Wipf, and Klaiber 1997). Yes 43 (84%) No 8 (16%) Figure 3-4. Use of NDEs for bridge element inspection, by percentage of state DOTs.
22 Bridge Element Data Collection and Use ⢠Ultrasonic. This method consists of measuring the time needed for ultrasonic pulses to travel through concrete members and can be used to detect cracks and voids, determine the quality of concrete compressive strength, measure the variation of elastic modulus over the concrete deck, and locate and measure cracks or discontinuities in steel members (Rens, Nogueira, and Transue 2005; Hooks and Weidner 2016; Kaczmarek, Piwakowski, and Drelich 2017). Figure 3-5 shows the number of state DOTs that use an NDE technique, by the frequency of use. Among the selected NDE techniques, state DOTs most often employed the chain drag, followed by ultrasonic and other testing methods. The state DOTs that selected other methods were directed to a follow-up question requesting details. Seven DOTs mentioned dye-penetrant testing, which is used to detect discontinuities that are open to the surface, such as cracks, seams, laps, cold shuts, laminations, through leaks, or lack of fusion (Hooks and Weidner 2016). Two DOTs listed D-Meters, an ultrasonic technology to measure thickness. Chain drag, GPR, electro magnetic detection, and ultrasonic testing were the most common methods that state DOTs sometimes use. The number of total selections for frequency categories (never, rarely, sometimes, and often) indicate that the rare use of NDE techniques for bridge element inspec- tions is most typical. Quality Control and Quality Assurance for Bridge Element Data In answer to a question on agency confidence in the quality of the bridge element data, the majority of state DOTs (55%) reported moderate confidence, followed by 39% with high 1 (2%) 7 (14%) 11 (22%) 18 (35%) 16 (31%) 5 (10%) 14 (27%) 5 (10%) 26 (51%) 25 (49%) 18 (35%) 15 (29%) 13 (26%) 4 (8%) 21 (41%) 11 (22%) 8 (16%) 6 (12%) 14 (27%) 23 (45%) 4 (8%) 21 (41%) 0 1 (2%) 3 (6%) 0 7 (14%) 3 (6%) 0 5 10 15 20 25 30 Chain drag Ground penetrating radar Infrared thermography Impact echo Electromagnetic Ultrasonic Other Often Sometimes Rarely Never Figure 3-5. Use frequency of NDE techniques, by percentage of all state DOTs.
State of the Practice 23  confidence and 6% with low confidence (Figure 3-6). Not one agency indicated no confidence in the quality of its bridge element data. Most state DOTs also noted that they have QC and QA processes in place that improve the quality of bridge element inspections (Figure 3-7). Performance Measures and Models The majority of state DOTs did not report performance measures, project decision rules, or decision trees that are based on bridge element data. However, 45% of the state DOTs noted that their performance measures are based on element data while 39% maintain decision trees based on element data that facilitate decisions for projects (Figure 3-8). Responses to a survey question on the existence of element cost models showed that 18 state DOTs (35%) do not have such models (Figure 3-9). Agencies could select multiple responses to this question because an agency may fit into two categories. The state DOTs either developed cost models or used default cost models available in their BMSs. Only six state DOTs (12%) expressed confidence in their existing element cost models; 26 state DOTs (more than half, 51%) noted that their cost models need further improvement. 20 (39%) 28 (55%) 3 (6%) 0 0 5 10 15 20 25 30 High confidence Moderate confidence Low confidence No confidence Figure 3-6. Agency confidence in the quality of bridge element data. Yes 47 (92%) No 4 (8%) Figure 3-7. Existence of QC and QA programs that improve quality.
24 Bridge Element Data Collection and Use The status of element deterioration models is very similar to that of the element cost models (Figure 3-10). A total of 36 state DOTs reported deterioration models, but 15 did not. Of the 36 state DOTs with such models, 5 used default models, and 31 developed their own. More than half of the state DOTs (51%) indicated that their deterioration models need further improvement. Depending on the decision-making tools utilized by a DOT, the state may prefer to compare element condition forecasts to NBI GCRs because federal performance measures are based on NBI GCRs. A total of 26 state DOTs (51%) responded that they do not compare element condition data and NBI GCRs (Figure 3-11). Six state DOTs (12%) reported conversion profiles in which they are confident while another six state DOTs use the default conversion profiles available 18 (35%) 6 (12%) 6 (12%) 26 (51%) 0 5 10 15 20 25 30 We do not have element cost models. We use default cost models that were available in the BMS. We developed element cost models that we are confident in. We developed element cost models, but they need further improvement. Figure 3-9. Status of element cost models, by DOT count. Yes 23 (45%)No 28 (55%) Use of performance measures based on element data Yes 20 (39%) No 31 (61%) Project decision rules or decision trees based on bridge element data Figure 3-8. Use of performance measures and decision trees based on bridge element data.
State of the Practice 25  in their BMSs. More than 25% of the state DOTs (13 agencies) developed a conversion profile that they believe needs further improvement. Use of Bridge Element Data in Asset Management State DOTs most commonly used bridge element data in asset management decision-making (Figure 3-12) for the selection of bridge preservation projects, bridge-level decision-making and analysis, and selection of bridge rehabilitation and replacement projects (as indicated by 33â34 state DOTs, or roughly 65%). More than half of the state DOTs employed bridge element data to select bridge maintenance projects or to perform network-level decision-making analysis, such as project prioritization or strategy assessment within a BMS (about 52% to 55%). Only four DOTs (8%) reported not using element data in asset decision-making while seven DOTs (14%) noted that they utilize bridge element data for asset decision-making purposes other than the listed options. Of the seven state DOTs that recorded other uses for bridge element data, five briefly explained these applications. The Arizona DOT observed that it employs bridge element data to control NBI GCRs. The NBI GCRs impact sufficiency ratings that are used for scores that in turn are inputs for network-level decision-making. The Connecticut DOT reported that although bridge element data are not a factor in asset management decisions, the agency is working toward 15 (29%) 5 (10%) 8 (16%) 25 (51%) 0 5 10 15 20 25 30 We do not have element deterioration models. We use default deterioration models that were available in the BMS. We developed element deterioration models that we are confident in. We developed element deterioration models, but they need further improvement. Figure 3-10. Status of element deterioration models, by DOT count. 26 (51%) 6 (12%) 6 (12%) 13 (25%) 0 5 10 15 20 25 30 We do not compare them. We use a default conversion profile available in the BMS. We developed a conversion profile/model that we are confident in. We developed a conversion profile/model, but it needs further improvement. Figure 3-11. Status of element condition data and NBI GCR comparison, by DOT count.
26 Bridge Element Data Collection and Use applying bridge element data for this purpose. The Connecticut DOT is currently transitioning from a component-based BMS to an element-based BMS. The Michigan DOT indicated that its elements are used to identify the items that exist on a bridge (with multiple ADEs on wearing surfaces and protective systems) and consequently determine the potential project candidates. The New York DOT indicated that bridge element data function as inputs for its legacy model- ing that does performance calculations. The Ohio DOT also reported working toward utilizing element data in asset decision-making. Ohio recently started element inspections, and the DOT is planning to use a BMS for network-level decision-making. Two questions in the survey investigated state DOT confidence in element- and component- based asset management decisions. As a comparison, state DOTs hold relatively more confi- dence in component-based decisions than in element-based decisions (Figure 3-13 and Fig- ure 3-14). A total of 43 state DOTs (roughly 85%) reported moderate or high confidence in 7 (14%) 4 (8%) 27 (53%) 34 (67%) 33 (65%) 34 (67%) 28 (55%) 0 10 20 30 40 Other We do not use bridge element data to support asset management decisions. Network-level decision making/analysis (e.g. project prioritization and strategy assessment with a BMS that uses element models). Bridge-level decision making/analysis (e.g., work type or scope for individual structures). Selection of bridge rehabilitation/replacement projects. Selection of bridge preservation projects. Selection of bridge maintenance projects. Figure 3-12. State DOT use of element data in asset decision-making. 2 (4%) 21 (41%) 15 (29%) 0 13 (25%) 0 5 10 15 20 25 High confidence Moderate confidence Low confidence No confidence We do not use element data or models Figure 3-13. Confidence in decisions based on element data or models.
State of the Practice 27  the asset management decisions that they make when relying on component data and models; however, only 23 state DOTs (about 45%) acknowledged moderate or high confidence in their decisions based on element data and models. Zero DOTs indicated no confidence in their deci- sions based on either data type, but more than 20% of the state DOTs responded that they do not utilize element data or models in the first place. Among the state DOTs that confirmed using element data or models in decision-making, 23 of the 38 state DOTs (60%) reported moderate to high confidence in the decisions based on element data or models. In comparison, 43 of the 47 state DOTs (91%) claimed moderate to high confidence in their decisions based on component data or models. Responses to other survey questions (Figure 3-8 through Figure 3-11) suggest room for improvement and plans to enhance element performance measures and models. The relatively higher level of confidence in decisions relying on component data and models may stem from the lengthier history of state DOTs applying and developing models for component data. Survey Takeaways Survey responses indicate that all DOTs are collecting NBE and BME data aligned with federal guidelines while 67% of the state DOTs are also gathering data for ADEs. Agencies are also col- lecting data on element defects (76%), but data gathering on environments is less common (43%). State DOT NDE methods often include chain drags for bridge deck inspections and some- times involve electromagnetic or ultrasonic testing (or both) for element inspections. NDE tools such as impact-echo tests, IRT, GPR, dye-penetrant testing, and D-Meters are also employed, but less frequently. Compared to NBI GCRs, bridge element data are more quantitative and detailed. However, state DOTs report more confidence in the results based on component data. The majority of state DOTs (55%) express moderate confidence in the quality of their bridge element data, and most of the remaining DOTs (39%) indicate high confidence in the quality of their bridge element data. The state DOTs also believe that their existing QC and QA programs improve the quality of bridge element inspections. Less than half of the state DOTs have established project decision rules, decision trees, or performance measures based on bridge element data. One-fourth of the state DOTs express confidence in their element cost and deterioration models. Most typically, the state DOTs have developed element or cost deterioration models but observe that these models need further improvement. In addition, 26 state DOTs do not compare element condition data and NBI GCRs. Meanwhile, the remaining 25 state DOTs have developed a conversion profile that needs further improvement, have generated a profile in which they are confident, or employ a default conversion profile available in their BMSs. 9 (18%) 34 (67%) 4 (8%) 0 4 (8%) 0 5 10 15 20 25 30 35 40 High confidence Moderate confidence Low confidence No confidence We do not use component data or models Figure 3-14. Confidence in decisions based on component data or models.
28 Bridge Element Data Collection and Use One-fourth of the state DOTs do not integrate element data or models into asset management decisions. Confidence in models and decision-making based on component data is relatively high compared to the same measure of decisions based on element data or models. Responses to survey questions indicate that state DOTs do have plans to improve element performance measures and models. The relatively more robust confidence in decisions based on component data and models may stem from the lengthier history of state DOTs applying and developing models for component data. At the end of the process, state DOTs rely on a combination of data resources, engineering judgment, and experience when making final decisions. The most common uses of element data in asset decision-making involve the selection of bridge preservation projects, bridge-level decision-making (e.g., choice of work type or scoping for individual structures), and selection of bridge rehabilitation or replacement projects. State DOTs also commonly apply element data in choosing bridge maintenance projects and making network-level decisions. Aside from four state DOTs, all of the rest report some form of use for bridge element data.
