Proposed Guideline for Reliability-Based Bridge Inspection Practices (2014)

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.

Proposed Guideline for Reliability-Based Bridge Inspection Practices P a r t I

3 This guideline describes a methodology for developing Risk-Based Inspection (RBI) prac- tices for highway bridges. The goal of the methodology is to improve the safety and reliabil- ity of bridges by focusing inspection efforts where most needed and optimizing the use of resources. The guideline provides the framework and procedures for conducting reliability assessments to develop suitable inspection strategies for bridges based on an engineering assessment of inspection needs. The methodology considers the structure type, age, con- dition, importance, environment, loading, prior problems, and other characteristics that contribute to the reliability and durability of highway bridges. RBI practices differ from traditional approaches that are generally calendar based, because the setting of inspection frequencies (or intervals) and scope are not fixed or uniform. Rather, reliability-based engineering analysis is conducted to assess the inspection needs for a particular bridge or family of bridges, and inspection requirements, i.e., frequency and scope, are aligned with those needs. This is achieved by analyzing the likelihood of anticipated damage modes and the associated outcomes or consequences. As a result, RBI practices can focus attention specifically on the damage and deterioration mechanisms that are most important for ensuring bridge safety. As such, they provide a better linkage between damage modes that affect bridges and the inspection approaches that will best reduce the associated risks, leading to improved bridge safety. This approach has been widely accepted in many industries with facilities that can be considered analogous to highway bridges: very large, expensive, and complex structural systems that are exposed to rugged environmental conditions and mechanical loading. The purpose of this document is to provide guidance for bridge owners for conduct- ing reliability-based assessments to determine inspection needs. The methodology requires bridge owners to perform a reliability assessment of bridges within their bridge inventory to identify those bridges that are most in need of inspection to ensure bridge safety, and those where inspection needs are less. An expert panel assembled at the owner level performs this assessment. The assessment considers the reliability and safety attributes of the bridges to assess the likelihood of damage and evaluate the potential outcomes or consequences in terms of safety and serviceability. Through this process, inspection needs are prioritized to improve the safety and reliability of the bridge inventory overall. S U M M A R Y Proposed Guideline for Reliability-Based Bridge Inspection Practices

4Definitions Attributes: Characteristics that affect the reliability of a bridge or bridge element. Condition Attributes: Characteristics that relate to the current condition of a bridge or bridge element. These may include element ratings, component ratings, and specific damage modes or mechanisms that have a significant effect on the reliability of an element. Consequence Factor: Factor describing the expected outcome or result of a failure. Damage Mode: Typical damage affecting the condition of a bridge element (e.g., spalling of concrete, cracking, etc.). Design Attributes: Characteristics of bridge or bridge element that are part of the element’s design. These attributes typically do not change over time except when renovation, rehabilitation, or preservation activities occur. Deterioration Mechanism: Process or phenomena resulting in damage to a bridge element (e.g., corrosion, fatigue, etc). Element: Identifiable portions of a bridge made of the same material, having similar role in the performance of the bridge, and expected to deteriorate in a similar fashion. Failure: Termination of the ability of a system, structure, or component to perform its intended function (1). For bridges, the condition at which a given bridge element is no longer performing its intended function to safely and reliably carry normal loads and maintain serviceability. Loading Attributes: Loading characteristics that affect the reliability of a bridge or bridge ele- ment, such as traffic or environment. Occurrence Factor: Factor describing the likelihood that an element will fail during a specified time period. Operational Environment: The operational environment is a combination of the circumstances surrounding and potentially affecting the in-service performance of bridges and bridge elements. These include typical loading patterns, ambient environmental conditions, construction quality and practices, maintenance and management practices, and other factors that may vary between different geographic regions and/or organizational boundaries. Probability: Extent to which an event is likely to occur during a given time interval (1). This may be based on the frequency of events, or on degree of belief or expectation. Degrees of belief about probability can be chosen using qualitative scales, ranks, or categories such as “remote/ low/moderate/high.” Reliability: Ability of an element or component to operate safely under designated operating conditions for a designated period of time.

Definitions 5 Risk: Combination of the probability of an event and its consequence. Risk Analysis: Systematic use of information to estimate the risk. Sources of information may include historical data, theoretical analysis and engineering judgment. Screening Attribute: Characteristics of a bridge or bridge element that: • Make the likelihood of serious damage unusually high, • Make the likelihood of serious damage unusually uncertain, and • Identify a bridge with different anticipated deterioration patterns than other bridges in a group or family.

6Introduction This guideline describes a methodology for developing Risk-Based Inspection (RBI) prac- tices for highway bridges. The goal of the methodology is to improve the safety and reli- ability of bridges by focusing inspection efforts where most needed and optimizing the use of resources. The guideline provides a framework and procedures for developing suitable inspection strategies based on a rational engineering assessment of inspection needs. The methodology considers the structure type, age, condition, importance, environment, loading, prior problems, and other characteristics that contribute to the reliability and durability of highway bridges. The methodology requires bridge owners to perform a reliability assessment of bridges within their bridge inventory to identify those bridges that are most in need of inspection to ensure bridge safety, and those where inspection needs are less. This assessment is conducted by con- sidering the reliability and safety attributes of bridges, assessing the likelihood of damage and associated deterioration mechanisms, and evaluating the potential outcomes or consequences in terms of safety and serviceability. Through this process, inspection needs are prioritized to improve the safety and reliability of the bridge inventory overall. This chapter of the document provides an introduction and overview of the process, as well as background information on the underlying theories and common practices for RBI and reli- ability assessments. Chapter 2 of the document describes the methodology for conducting a reli- ability assessment for bridges. This includes providing a definition of element failure suitable as an analysis tool, and a description of the key factors to be assessed in the typical reliability assess- ment conducted for inspection planning purposes. This chapter also describes the composition of the Reliability Assessment Panel (RAP) that will conduct the assessments. Chapter 3 describes the process for determining the appropriate maximum inspection interval and scope of inspection, based on analysis as described in Chapter 2. The underlying approaches for identifying inspection intervals and the techniques or methods to be used for the inspec- tions are discussed. Finally, Chapter 4 provides an overview of the overall process, guidance for bridge owners on beginning an RBI program, transitioning from traditional, calendar-based approaches, and general guidance on the training that may be required. There are six appendices in the document that describe in more detail the process and mechanics of the analysis. Guidance for determining the factors necessary to perform a reli- ability assessment are included in Appendices A, B, and C. Guidance on inspection methods and nondestructive evaluation (NDE) technologies that can be used for conducting RBIs is described in Appendix D. Appendix E contains commentary regarding specific, common attributes of bridges that influence damage modes and deterioration mechanisms, and relate to bridge reliability. Finally, Appendix F includes three example implementations of the methodology applied to bridges of common design: a multi-girder concrete bridge with C H A P T E R 1

Introduction 7 prestressed superstructure elements constructed in the past 5 years, a multi-girder steel bridge constructed more than 50 years ago, and a multi-girder reinforced concrete bridge constructed in 1963. 1.1 Process The process involves an owner (e.g., state) establishing an expert panel to define and assess the durability and reliability characteristics of bridges within the state. The expert group analyzes portions of the bridge inventory to assess inspection needs by using engineering rationale, expe- rience, and typical deterioration patterns to evaluate the reliability characteristics of bridges and the potential outcomes of damage. This is done through a relatively simple process that consists of three primary steps: Step 1: What can go wrong, and how likely is it? Identify possible damage modes for the ele- ments of a selected bridge type. Considering design, loading, and condition characteristics (attributes), categorize the likelihood of serious damage occurring into one of four occurrence factors (OFs) ranging from remote (very unlikely) to high (very likely). Step 2: What are the consequences? Assess the consequences, in terms of safety and serviceabil- ity, assuming the given damage modes occur. Categorize the potential consequences into one of four consequence factors (CFs) ranging from low (minor effect on serviceability) through severe (i.e., bridge collapse, loss of life). Step 3: Determine the inspection interval and scope. Use a simple 4 × 4 matrix to prioritize inspection needs and assign an inspection interval for the bridge based on the results of Steps 1 and 2. Damage modes that are likely to occur and have high consequences are priori- tized over damage modes that are unlikely to occur or are of little consequence in terms of safety. An RBI procedure is developed based on the assessment of typical damage modes for the bridges being assessed. Inspections are conducted according to the RBI procedure developed through this process. Results of the inspection are assessed to determine if the existing RBI procedure needs to be modified or updated as a result of findings from the inspection. Through this process, individual bridges, or groups of bridges of similar design characteristics, can be assessed to evaluate the inspection needs based on an engineering analysis of the likelihood of serious damage occurring and the effect of that damage on the safety and serviceability of the bridge. This approach considers the structure type, age, condition, and operational environment in a systematic manner to provide a rational assessment process for inspection planning. A docu- mented rationale for the inspection strategy utilized for a given bridge is developed. The damage modes most important to ensuring the safety and serviceability of the bridge are identified such that inspection efforts can be focused to improve the reliability of the inspection results. 1.1.1 Scope This guide is focused on the inspection of typical highway bridges of common design char- acteristics. Atypical structures, such as long-span truss bridges, cable-stayed bridges, suspen- sion bridges, and other unique or unusual bridge designs may require certain considerations not presently captured in this guideline; this guideline provides for inspection planning for the superstructure, substructure, and deck for typical highway bridges. Scour and underwater inspections have existing methodologies for evaluation, and, as such, are not included herein. Bridges assessed using this methodology are assumed to have a current load rating that indicates that the structural capacity is sufficient to carry allowable loads.

8 Proposed Guideline for Reliability-Based Bridge Inspection Practices 1.1.2 Purpose The purpose of this document is to provide guidance for bridge owners for conducting reliability-based assessments for determining the frequency and scope of inspections for typical highway bridges. This document is intended to be used by bridge owners for assessing their bridge inventories in order to prioritize inspection needs based on an engineering analysis that considers the bridge type, age, loading, condition, and other characteristics of a bridge. This guideline is intended for application to typical bridges with common and ordinary forms of deterioration and damage. Advanced deterioration and/or specific defects such as fatigue cracks due to primary stresses or severe corrosion damage in concrete typically require more detailed engineering analysis than provided herein. 1.2 Background The periodic inspection of highway bridges in the United States plays a critical role in ensuring the safety, serviceability, and reliability of bridges. Inspection processes have developed over time to meet the requirements of the National Bridge Inspections Standards (NBIS)(2) and to meet the needs of individual bridge owners in terms of managing and maintaining their bridge inventory. The inspection frequency mandated by the NBIS requires the inspection interval (maximum time period between inspections) not to exceed 24 months. Based on certain criteria, that interval may be extended up to 48 months with approval from the Federal Highway Administration (FHWA) (3). Maximum inspection intervals of less than 24 months are utilized for certain bridges accord- ing to criteria developed by the bridge owner, typically based on age and known deficiencies. Most bridge owners utilize the uniform maximum inspection interval of 24 months, as mandated by the NBIS, for the majority of the bridges in their inventory, and the reduced intervals for bridges with known deficiencies. Only 15 states utilize the 48 month policy, often only for culverts. The uniform inspection interval of 24 months was specified at the origination of the National Bridge Inspection Program in 1971 based on experience, engineering judgment, and the best informa- tion available at the time. The uniform approach provides a single maximum inspection interval for most bridges, regardless of the bridge age, design, or environment. To date, this mandated inspection interval has provided an adequate level of safety and reliability for the bridge inven- tory nationwide. However, such a uniform inspection interval does not consider explicitly the likelihood of failure based on bridge condition, design, or operating environment, or the poten- tial consequences of a failure. A uniform inspection interval does not recognize that a newly constructed bridge with improved durability characteristics and a few years of exposure to the service environment may be much less likely to develop serious damage over a given time interval than an older bridge that has been exposed to the service environment for many years. Bridges that are in benign, arid operating environments are inspected at the same interval as bridges in aggressive marine environments, where significant damage from corrosion may develop much more rapidly. Current practices make it difficult to distinguish if the same or improved safety and reliability could be achieved by varying inspection methods or frequencies to meet the needs of a specific bridge based on its design and operational environment. The current approach also makes it difficult to analyze if a given inspection activity is excessive, or if it provides little or no measure of increased assurance of the safety and reliability of bridges. Given that any inspection activity carries with it a certain amount of risk to both the inspector and to the traveling public, inspections that are excessive or that provide little benefit may present added, unnecessary risks. Otherwise, inspections that are inadequate or fail to distinguish the importance of critical damage modes may also present certain added risks that require analysis. Recognizing the variability in the design, condition, and operating environments of bridges would provide for inspection requirements that better meet the needs of individual bridges to

Introduction 9 improve both bridge and inspection reliability. Other industries are increasingly recognizing the limitations of prescribed inspection frequencies and are developing methodologies for effi- ciently assessing inspection needs, ensuring the safety and reliability of systems, and focusing inspection resources most effectively (1, 4–6). Methodologies for assessing inspection needs based on the likelihood of a service failure, combined with the consequences of such a failure, is a common approach to inspection planning and to developing effective inspection strategies. These approaches are typically described as risk-based, where inspection planning is conducted considering the reliability of a component, i.e., how likely is it that the component or machine will fail during a certain time period, and the consequences of such an event. Damage modes and deterioration mechanisms are typically assessed explicitly to determine the likelihood of failure during a given time period, and to identify the appropriate inspection methods to detect critical damage prior to failure. A risk-based approach has been adopted in many industries as a tool for inspection plan- ning, to focus attention on the component or machine that represents the greatest “risk.” Risk is defined as the product of the probability of an event and the associated consequences: Risk Probability Consequence= × Probability in this equation is the likelihood of an adverse event or failure occurring during a given time period. This is sometimes expressed quantitatively as a probability of failure (POF) estimate for a given time interval, or as a qualitative assessment of the likelihood of an adverse event based on experience and engineering judgment. Consequence is a measure of the impact of the event occurring, which may be measured in terms of economic, social, safety, or environ- mental impacts. Risk can be expressed quantitatively using POF estimates or models and quantitative mea- sures of consequences, such as the cost of a certain event or the loss of service of a component. Risk can also be expressed qualitatively by estimating whether the likelihood of a certain event is high, medium, or low, and determining a qualitative estimate of the consequences. Present- ing risk qualitatively is a common and effective method for evaluating risk and for assessing relative risk efficiently. Figure 1 shows a qualitative risk matrix (1, 5). This matrix shows a good representation of the overall concept and basic principles of risk. A high likelihood (probability) of occurrence combined with a high consequence results in a high risk, located in the upper right corner of the figure. Low likelihood combined with a low consequence results in low risk, located in the lower left-hand corner of the figure. High risk and low risk elements typically do not create challenges in decision making; items that are “high risk” may not be acceptable and actions are required to lower the risk, either by reducing the likelihood of an event, or by reducing the consequences, or both. Items that are “low risk” are typically acceptable and may require little or no action. In the “medium risk” area, questions may arise about how much risk is acceptable, and what the appropriate decision-making strategies are for mitigating that risk. In terms of inspection strategies, items that are “high risk” are prioritized for more frequent and possibly more intense inspections to reduce uncertainty and to monitor the development of damage to ensure that safety is maintained. Items that are “low risk” may have longer inspection intervals and have less intense inspection protocols. An important concept in risk analysis is to understand that high likelihood does not necessar- ily mean high risk, if the consequences are small. Similarly, high consequence does not necessar- ily mean high risk, if the likelihood is small. The level of risk can only be determined once both of these variables are assessed. A risk-based planning approach focuses attention not on the items that are most likely to fail, but rather those items whose failure is most important, by considering both the likelihood of Figure 1. Risk matrix showing high-, medium-, and low-risk values.

10 Proposed Guideline for Reliability-Based Bridge Inspection Practices failure and the associated consequences. The setting of inspection frequencies or intervals is not a rigid process, such as is the case for uniform or calendar-based inspection frequencies. Rather, it is a process that evolves and changes over the life of a component such that inspection frequen- cies change as risk increases (or decreases). Therefore, the frequency of inspection is aligned with the needs and the associated risks, focusing attention on the most at-risk items. This approach has been widely accepted in many industries with facilities that can be considered analogous to highway bridges: very large, expensive, and complex structural systems that are exposed to rug- ged environmental conditions and mechanical loading (1, 4, 6). 1.2.1 Reliability and Probability Reliability is defined as the ability of an item to operate safely under designated operating conditions for a designated period of time or number of cycles. For bridges and bridge elements, reliability typically decreases as a function of time due to deterioration and the damage accumu- lated during the service life of a bridge. That is, the likelihood of failure typically increases with time as the element ages, due to deterioration mechanisms such as corrosion or fatigue. The reliability of a bridge or bridge element can be expressed as: PrR t T t( ) ( )= ≥ Where R(t) is the reliability, T is the time to failure for the item, and t is the designated period of time for the item’s operation. In other words, the reliability is the probability (Pr) or likeli- hood that the failure time exceeds the operation time. Sometimes, the probability is expressed as a probability density function (pdf) that expresses the time to failure of an item (T) as some generic distribution, such as normal, log normal, etc. This distribution can be used to calculate a POF function, F(t), to express the probability that the item will fail sometime up to time t. This time-varying function describes likelihood of failure up to some given time, or the unreliability of the item, and the reliability is then: 1R t F t( ) ( )= − In other words, the reliability is the probability that the item will not fail during the time period of interest. When a large population of test data of identical or near identical components exposed to the same operational environment are available, a probability function describing the failure characteristics of the component may be determined and verified based on the results. If test data are not available, a suitable distribution must be assumed based on the general char- acteristics of the population, typical failure behavior, and known deterioration mechanisms. These distributions are typically based on experience and assumptions regarding the anticipated performance of the system or component. This is challenging and can lead to unsubstantiated confidence in the model when the design characteristics, construction quality, condition, and operational environment of the components vary. Even if substantial data were readily avail- able, design and construction practices are constantly evolving such that past performance may not indicate future performance. Critical damage modes may have yet to manifest in observable damage, and as such may not be included in the data. Given the large variation in the design, construction, construction quality, and operational environments, the utility of probabilistic models to effectively predict the future performance of a specific bridge or bridge element is problematic. Under these circumstances, engineering judgment and experience is needed to estimate the expected reliability of a specific component, or set of components, of similar design and con- struction quality operating within a specific operational environment. Engineering judgment

Introduction 11 is required to estimate the reliability of bridge elements based on past experience, engineering knowledge, and a rational process to systematically assess bridges of common design and con- struction characteristics. The process involves engineers with experience and expertise in the performance of bridges within a particular operational environment using engineering judg- ment to assess the probability (likelihood) of failure during some future time period. When combined with an assessment of the consequences, an effective analysis can be conducted to identify inspection needs efficiently. 1.2.2 Consequences The primary purpose of bridge inspection is to ensure the safety and serviceability of highway bridges. As a result, the consequences to be assessed in prioritizing the importance of different damage modes are assessed in terms of bridge safety and serviceability. The consequence of fail- ure, or of serious damage developing in a bridge element, typically depends on the role of that element in the structural system of the bridge, and on the operating environment surrounding the bridge. For example, the consequence of an abutment having severe corrosion damage might be low, while the same damage in a main superstructure member may be high. The consequence of damage developing at the soffit of a bridge deck, such as concrete spalling, might be low if the bridge is over a flood plain, but high if the bridge is over an interstate highway. The consequence associated with a given damage mode can be assessed through engineering judgment, through common or related experience, or through theoretical analysis. The process developed and described herein requires the determination of two key param- eters: an estimate of the reliability of given bridge elements, based on the likelihood (probability) that the element would fail during a given time interval, and an assessment of the consequences of that failure. These data are then used to determine an appropriate inspection interval and scope (procedures and methods) for a bridge. As such, the methodology described is a reliability-based bridge inspection planning process for ensuring the safety and serviceability (i.e., reliability) of highway bridges.

12 This section describes the methodology for reliability assessment of the bridge elements. Sec- tion 2.1 describes and defines failure as applied to typical bridge elements for the reliability assessment. Section 2.2 describes the methodology for evaluating the probability or likelihood that failure will occur (OF). Section 2.3 describes the methodology for evaluating the conse- quences of that occurrence (CFs). Finally, Section 2.4 discusses the panel that conducts the assessment, the RAP. 2.1 Definition of Failure It is critical that the conditions that constitute a failure be defined before beginning a reliabil- ity assessment. For bridges, catastrophic collapse would be one obvious condition that could be used to define failure. For most bridges, the probability of catastrophic failure is very remote. For bridge inspections, important concerns extend well beyond simply avoiding catastrophic failure. Ensuring the safety of the bridge, the safety of those traveling on or below the bridge, and the serviceability of the bridge are each critical. Maintenance and repair activities are needed to support the serviceability of the bridge and ensure the safety of motorists, even while the likeli- hood of a catastrophic failure remains remote. Therefore, failure requires a suitable definition that captures the need to ensure the struc- tural safety of the bridge, the safety of travelers on or below the bridge, and the serviceability of the bridge. Failure, utilized in this context, is defined as when an element is no longer per- forming its intended function to safely and reliably carry normal loads and maintain serviceability. For example, a bridge deck with severe spalling may represent a “failed” condition for the bridge deck even though the deck may have adequate load-carrying capacity, because the ability of the deck to reliably carry traffic is compromised. The condition rating of 3, “serious condi- tion” according to the NBIS rating system, is used in the analysis described herein as a general description of a “failed” condition. It is not envisioned that any bridges or bridge elements assessed using a risk-based approach are allowed to deteriorate to this condition. Rather, inspec- tion intervals are adjusted to ensure that the likelihood of failure in the time intervals between inspections always remains low. Bridge components that have deteriorated to this extent may no longer be performing their intended function, and remedial actions are typically planned to address such conditions. The subjective condition rating of 3 is defined within the Recording and Coding Guide (7) as follows: NBI Condition Rating 3: SERIOUS CONDITION: Loss of section, deterioration, spalling or scour have seriously affected primary structure components. Local Failures are possible. Fatigue cracks in steel or shear cracks in concrete may be present. C H A P T E R 2 Reliability Assessment of Bridge Elements

Reliability Assessment of Bridge Elements 13 This condition description is widely understood and there is significant past experience in the conditions warranting a rating of 3 throughout the bridge inventory. This condition descrip- tion is not absolute, but provides a frame of reference for the analyst considering the likelihood of damage occurring to a serious extent. In terms of the AASHTO Bridge Element Inspection Guide, this condition generally aligns with elements in condition state (CS) 4, “severe. ” (8) 2.2 Occurrence Factors What can go wrong, and how likely is it to occur? The first step in the reliability assessment is to address the question “What can go wrong, and how likely is it to occur?” The first part of this question, “what can go wrong” addresses the damage modes that affect typical bridge elements. In other words, what damage is likely to develop over the service life of the bridge, which may result in the failure of a given element? “Failure” used in this context is serious damage to the element such that its performance as intended cannot be assured, as described in Section 2.1 (e.g., condition rating = 3 or CS = 4). For concrete elements, spalling and cracking of the concrete is a typical damage mode. For steel elements, section loss or cracking are typical damage modes. The second part of this question, “how likely is it to occur,” describes the likelihood, or probability, of failure due to that damage mode occurring, given the design, materials, and current condition of a bridge element. The OF categorizes this likelihood on a qualitative scale that provides an assessment of the likelihood of serious damage, i.e., failure, occurring. For the assessment of bridge inspection needs, the OF is usually an assessment of the likeli- hood that a given damage mode will result in failure (i.e., serious condition), over a time period of 72 months (6 years). The deterioration mechanism resulting in the damage is considered in the assessment. In some cases the OF may be an estimate of the likelihood of a certain adverse event occurring, such as impact from an over-height vehicle or an overload. Each damage mode or adverse event must have a separate OF, based on the likelihood of the damage mode or the event resulting in failure of an element during the specified time interval. The OF describes the likelihood of failure of an element in one of four categories. The scale ranges from remote, when the likelihood is extremely small such that it would be unreasonable to expect failure, to high, where the likelihood of the event is increased, as shown in Table 1. To assess the appropriate OF for a given bridge element, key characteristics, or attributes, are considered. “Attributes” are characteristics of a bridge element that contribute to the ele- ment’s reliability, durability, or performance. These attributes are typically well-known param- eters affecting the performance of a bridge element during its service life. This includes relevant design, loading, and condition characteristics that are known or expected to affect the durability and reliability of the element. For example, consider the damage mode of spalling due to corro- sion damage in a concrete bridge deck. A bridge deck may have “good” attributes, such as being in very good condition, having adequate concrete cover, epoxy-coated steel reinforcing, and minimal application of de-icing chemicals. Given these attributes of the deck, it may be very Level Category Description 1 Remote Remote likelihood of occurrence, unreasonable to expect failure to occur 2 Low Low likelihood of occurrence 3 Moderate Moderate likelihood of occurrence 4 High High likelihood of occurrence Table 1. OF rating scale for RBI.

14 Proposed Guideline for Reliability-Based Bridge Inspection Practices unlikely that severe damage (i.e., failure) would occur in the next 72 months. This is based on the rationale that the deck is presently in good condition, and has attributes that are well-known to provide resistance to corrosion damage. As such, an OF of “Low” or “Remote” might be used to describe the likelihood of failure due to this damage mode. Alternatively, suppose the deck is in an environment where de-icing chemicals are frequently used, the reinforcement is uncoated, and the current rating for the deck is a 5, Fair Condition, indicating that there are signs of dis- tress in the deck. Based on this rationale, the likelihood of serious damage developing would be much greater, resulting in an OF rating of “Moderate” or “High.” Past experience with decks of a similar design, characteristics of the specific operating environment, and attributes of the deck are combined with engineering judgment and used to support the assessment of the specific OF for a given deck. Methodologies for determining credible damage modes and their associated attributes are included in Appendix A. Certain key attributes will ideally be identified as part of criteria for reassessment of bridge inspection requirements, following subsequent RBIs. These key attributes are typically associ- ated with condition, which may change over the service life of the bridge as deterioration occurs. When changes in these condition attributes cause a change in the likelihood of a given damage mode resulting in failure (i.e., the OF), reassessment of the inspection requirements is necessary. Deterioration rate data, trends, and theoretical models can be used to support the catego- rization of the OFs by providing insight regarding the average, typical, or expected behavior of elements of a similar design. Transition probabilities, Weibull statistics, or regression trends, developed based on past inspection results, can provide insight into the anticipated behavior of a group of similar bridge elements. Care should be taken to ensure that the bridge elements being assessed have similar or the same attributes as those represented by the data. Theoretical models may also be used to support these assessments. However, the complexity and variations in the operational environment, construction variability, and current condition can be difficult to capture in these models. Results need to be verified using engineering judgment. 2.3 Assessment of Consequences What are the consequences? The second factor to be assessed within an RBI process is the Consequence Factor, CF, a categorization of the likely outcome presuming a given damage mode were to result in failure of the element being considered. The assessment of consequence is geared toward assessing and differentiating elements in terms of the consequences of the assumed element failure. It should be noted that failure of an element is not an anticipated event when using an RBI approach, rather the process of assessing the consequences of a failure is merely a tool to rank the importance of a given element relative to other elements for the purpose of prioritizing inspection needs. The CF is used to categorize the consequences of the failure of an element into one of four cat- egories, based on the anticipated or the expected outcome. Failure scenarios are considered based on the physical environment of the bridge, typical or expected traffic patterns and loading, the structural characteristics of the bridge, and the materials involved. These scenarios are assessed either qualitatively, through necessary analysis and testing, or based on past experience with similar failure scenarios. The four-level scale used to assign the CF is shown in Table 2. The CF ranges from low, used to describe failure scenarios that are benign and very unlikely to have a significant effect on safety and serviceability, through catastrophic scenarios, where the threat to safety and life is significant. Thus, both short-term (generally safety related) and long-term (generally serviceability related) consequences can be considered.

Reliability Assessment of Bridge Elements 15 In assessing the consequences of a given damage mode for a given element, the RAP must estab- lish which outcome characterized by the CFs in Table 2 is the most likely. In other words, which scenario does it have the most confidence will result if the damage were to occur. Using the illustra- tion of brittle fracture in a girder, it is obvious that the most likely consequence scenario would (and should) be different for a 150-foot span two-girder bridge than for a 50-foot span multi- girder bridge. For the short-span, multi-girder bridge, an engineer may state with confidence that the most likely consequence scenario is “high” or “moderate” and that the likelihood of “severe” consequences is very remote for a multi-girder bridge, based on his or her experience and the observed behavior of multi-girder bridges. For the two-girder bridge, the consequence scenario is likely to be “Severe.” As this example illustrates, the CF simply ranks the importance of the damage mode as being higher for a two-girder bridge than for a multi-girder bridge. For many scenarios, qualitative assessments based on engineering judgment and documented experience are sufficient to assess the appropriate CF for a given scenario; for others, analysis may be necessary using suitable analytical models or other methods. A series of more detailed criteria for specific elements [i.e., decks, steel girders, prestressed (P/S) girders, etc.] are provided in the Appendix B that can be utilized during the assessment to determine the appropriate CF for a given element failure scenario. These criteria, combined with owner-specific requirements developed in the RAP or from other rational sources for assessing bridges and bridge redundancy, are then used to deter- mine the appropriate CF for a given scenario. 2.4 The Reliability Assessment Panel An important component of the analysis process is the elicitation of expert judgment regard- ing the likelihood of damage and the level of associated consequences. Because design features, construction specifications and practices, materials, environment, and bridge management strategies differ from state to state, or even within a particular state, the expert panel should be selected keeping in mind the need to have membership which is familiar with the operational environment of the inventory of bridges being evaluated. The RAP typically will consist of four to six experts from the bridge-owning agency. This panel should include an inspection team leader or program manager that is familiar with the inspection procedures and practices, as they are implemented for the inventory of bridges being analyzed. The team should include a structural engineer who is familiar with the common load paths and the overall structural behavior of bridges, and a materials engineer who is familiar with the behav- ior of materials in the particular environment of the state and has past experience with materials quality issues. A facilitator may be used to assist in the analysis process. The general characteristics of members of a RAP include the following: 1. Bridge Inspection Expert: Inspection team leader or program manager that oversaw the specific inspection process and the reports for the bridges being evaluated. This individual should be able to represent the inspection results reported in the bridge file, understand the Level Category Consequence on Safety Consequence on Serviceability Summary Description 1 Low None Minor Minor effect on serviceability, no effect on safety 2 Moderate Minor Moderate Moderate effect on serviceability, minor effect on safety 3 High Moderate Major Major effect on serviceability, moderate effect on safety 4 Severe Major Major Structural collapse/loss of life Table 2. CFs for RBI.

16 Proposed Guideline for Reliability-Based Bridge Inspection Practices notes and sketches included in the file, and have an understanding of the scope and the meth- ods of the inspections used for the bridges under consideration. 2. State Program Manager or Bridge Management Engineer: Individual familiar with the characteristics and the behavior of the bridge inventory throughout the state. 3. Bridge Maintenance Engineer: An individual familiar with the standard methods and tech- niques used for bridge maintenance, the level of maintenance typical for the bridges under consideration, and the outcomes of bridge maintenance. 4. Materials Engineer: A materials engineers who is familiar with the history of materials perfor- mance within the state. This individual should be experienced with the materials historically used within the state, be knowledgeable of any prior problems with the quality or with the performance of the materials used, and be knowledgeable of typical deterioration patterns. 5. Structural Engineer: An engineer with sufficient training and experience to understand the consequences, in a structural sense, of bridge element failures. For example, the structural engi- neer should be able to recognize the load paths in a structure and to understand the importance of elements in the overall structural system of the bridge. 6. Independent Experts: The RAP may include independent experts, academics, or consultants to address specific or complex damage modes, provide independent review, and/or supple- ment the knowledge of the panel as needed. 7. Facilitator: A RAP facilitator may be used to assist in the RAP analysis, to lead expert elicita- tions, and help build consensus during the analysis process. The expert panel may also include representatives from the FHWA to monitor the process, to fulfill oversight responsibilities, and to assist with the implementation of the methodology used for inspection planning.

17 This section describes the process of determining the inspection interval and scope based on the assessment completed as described in Chapter 2. This process leads to a prioritization of inspection needs, highlights critical damage modes for bridges, and results in an RBI practice. 3.1 Inspection Interval The inspection interval is selected based on the RAP assessment of the OFs and CFs. Once these factors have been determined, their numerical values are used to place a given damage mode in the appropriate location on a reliability matrix. A typical reliability matrix is shown schematically in Figure 2. In this figure, the horizontal axis represents the CF as determined for a particular damage mode for a given bridge element. The vertical axis represents the outcome of the OF assessment for a given damage mode for a given element. Elements that tend toward the upper right corner of the reliability matrix require shorter inspection intervals, and possibly more intense inspections, than elements that fall in the lower left corner. The matrix is utilized to determine the appropriate maximum inspection interval for a given bridge or bridge type. These inspection intervals are determined to ensure that the probability, or likelihood, of failure remains low during the inspection interval. The maximum inspection inter- val is established in order to be consistent with the assessment of the OF, as determined over the predefined assessment interval of 72 months, as described in Section 2.2. Keeping this in mind, the actual maximum inspection interval is determined such that the likelihood of occurrence within the time between inspections (i.e., the inspection interval) always remains low. For example, if the OF is “low” over a 72-month period, than it may be reasonable to assign the inspection interval of 72 months (ignoring the influence of consequence for the time being). However, if it were found that the OF were high, the analysis is really indicating a failure is relatively likely to occur before the end of the 72-month interval. Since the goal is to ensure that the possibility of failure occurring before the end of the interval is always low, one would shorten the inspection interval, for example to 24 months. In other words, by inspecting every 24 months, the possibility of failure occurring before the end of the interval (now reduced to 24 months) remains low. Obviously, the OF is not the only parameter that should be evaluated when setting the inter- val. The consequence of the failure must also be incorporated into the process of selecting the appropriate interval. Using the example above, where the OF were high and the interval was reduced to 24 months; if the consequence of that same damage was determined to be severe, it would be appropriate to assign a shorter interval of, for example, 12 months. This provides an extra measure of confidence and safety (i.e., a reduction from 24 months to 12 months due to the severe nature of the consequences). Although there are many permutations of the OF and the CF, the above illustrates the concept. C H A P T E R 3 Determination of Inspection Interval and Scope

18 Proposed Guideline for Reliability-Based Bridge Inspection Practices This is a relatively easy task for elements where the OF is high and the CF is severe, and hence an interval of 12 months or less is needed. However, if the OF is remote and the CF is low, then it would also seem reasonable and justifiable that the inspection interval should be greater than the longest interval assumed in the OF assessment (72 months). (If the OF is remote, this indicates the members of the RAP concluded that there is a “remote likelihood of occurrence, unreasonable to expect failure to occur” in the next 72 months for this element and damage mode.) This information, coupled with the observation that failure, should it occur, is a low consequence, may justify the use of an inspection interval longer than 72 months. The actual inspection interval selected is based on the shortest inspection interval determined from the analysis. In other words, whichever element has the shortest maximum inspection interval, based on the likelihood of failure and associated consequence. In certain circumstances, there may be one element of the bridge that results in a much shorter inspection interval than the other elements of the bridge. In such a case, a different inspection interval may be identified for that particular element, based on engineering judgment and the discretion of the bridge owner. For most cases, multiple elements would be expected to have the same or very similar intervals, with the shortest interval being selected for practical reasons. 3.1.1 Inspection Scope Under an RBI practice, the inspection scope is determined from the damage modes identi- fied through the reliability analysis. In other words, the inspection methods used are selected based on their effectiveness and reliability for detecting the specific damage mode(s) that are most important. Guidelines for the selection of inspection methods to be used are included in Appendix D. In many cases, visual inspections supplemented with sounding are well-proven approaches to detecting typical damage in highway bridges. However, in a risk-based process, these inspections would include hands-on access to key portions of a bridge, such that damage is effectively identified to support the RBI assessment. For example, when assessing the likeli- hood of fatigue cracking in a bridge, it would be necessary to know if there were currently fatigue cracks in the bridge. Therefore, the inspection scope used to support the assessment must utilize an approach that is capable of making that determination. This would require hands-on access to certain locations where fatigue cracking is likely to occur. In some cases, NDE techniques are required, often based on a limited access for visual inspection (e.g., for detecting a crack in a bridge pin). Based on the assessment of the OF and the CF, damage modes for a bridge are prioritized based on the product: IPN OF CF= × Where IPN = Inspection Priority Number. For example, if the fatigue cracking has a moder- ate likelihood of occurring and the consequence is severe, then the IPN would be 3 × 4 = 12. If fatigue cracking were moderately likely, but the consequence were only moderate (minor service disruption), for example, if the bridge in question is a short-span, multi-girder bridge with known redundancy, the IPN for that damage mode would only be 3 × 2 = 6. This process highlights the damage modes that are most important, that is, most likely to occur, and have the greater associated consequences, if they did occur. It should be noted that the calculation of the IPN for each damage mode identified in the process does not limit the scope of the inspection to only those damage modes. However, it does provide a simple method to prioritize damage modes that are most important, based on a rational engineer- ing assessment that incorporates bridge type, age, design details, condition, etc., as well as the con- sequences of failure. Figure 2. Risk matrix for determining maximum inspection intervals for bridges.

Determination of Inspection Interval and Scope 19 3.1.2 Sampling When using the RBI approach, it may be appropriate to inspect a representative sample of a bridge element, using the inspection method identified. This can be used to reduce or limit inspection activities that provide little or no measure of increased benefit or that introduce risks that are unjustified. The sampling population size (number of locations or area, for example) should reflect the nature and type of damage to be assessed through the inspection. When dam- age modes are expected to be widespread and relatively uniform, such as spalling in a bridge deck, an appropriate sampling based on area may be justified. For example, inspecting 25% of the bridge deck to assess if delaminations are present. When damage modes are isolated or non-uniform, such as fatigue cracks, sufficient sampling must be based on analysis to identify the location and number of inspections. Criteria and analysis supporting the sampling should be documented. 3.1.3 Maintenance Inspections RBIs are typically more focused and intense than calendar-based, general-condition inspec- tions, and the maximum interval between inspections may be increased. For bridges with extended inspection intervals, maintenance inspections may be specified periodically to ensure the maintenance of traffic safety and to address general maintenance needs. These inspections are typically conducted by maintenance personnel with responsibility for the maintenance of the roadways and the bridges in the district or region where the bridge is located. The purpose of a maintenance inspection is to: • Identify bridge maintenance needs (minor patching, clearance of debris, vegetation control, etc.). • Confirm general conditions have not significantly changed. • Monitor unreported vehicular damage to a structure. • Evaluate traffic safety issues (maintaining signage, roadway delineations, etc.). Intervals for maintenance inspection would typically not exceed 2 years. Such maintenance inspections may be integrated into the business practice of a district or region. 3.1.4 Initial Inspections Initial inspections, the first inspection of a bridge following construction or reconfiguration of a structure (e.g., widening, lengthening, supplemental bents, etc.) are required according to AASHTO’s The Manual for Bridge Evaluation (9). In addition to this initial inspection, at least one RBI should be conducted at the interval of 24 months prior to initiating an RBI practice utilizing an interval greater than 24 months. Newly rehabilitated bridges should also have at least one RBI at the interval of 24 months following rehabilitation. The purpose of these inspections is to ensure that construction errors or deficiencies have not significantly altered the anticipated performance, and that a thorough inspection based on the RAP analysis has been conducted. 3.1.5 Start-Up Inspections When initiating an RBI practice for a bridge, the first RBI should be conducted at the regular interval for the bridge, typically 24 months under the current NBIS. This start-up RBI will imple- ment the practice as determined through the RAP analysis. Following the start-up inspection, the inspection results should be assessed for conformance with criteria and attributes identified by the RAP to determine if reassessment is necessary before implementing any modifications to the inspection interval.

20 Proposed Guideline for Reliability-Based Bridge Inspection Practices 3.1.6 Quality Control/Quality Assurance Quality control (QC) and quality assurance (QA) processes should be employed to ensure quality in the implementation of RBI practices. Procedures for QC could include data model reviews, scoring and reliability factor reviews, RAP procedures, and application of inspection intervals based on the RBI analysis. Procedures for QA could include analysis of historical bridge performance, consistency in data models developed from the RAP analysis, and field reviews of bridge performance under the RBI process. Additional methods for QC and QA for bridge inspection programs are available in the literature (10).

21 4.1 Overview of Process The overall process for implementing an RBI is shown schematically in Figure 3. The process begins with the selection of a bridge or family of similar bridges to be analyzed. For the selected bridge or bridges, the RAP identifies credible damage modes for elements of the bridge, given the design, materials, and operational environment. Key attributes are identified and ranked to determine the OFs, and the appropriate CFs associated with the damage modes are analyzed. Based on the assessment of the OFs and CFs for the bridge, the inspection practice is established including the interval and scope (procedures) for the inspection, and criteria for reassessment of the inspection practice. The criteria for reassessment are typically based on conditions that may change as a result of deterioration or damage, and may affect the OFs for the bridge. The RBI practice is then implemented in the subsequent inspection of the bridge. Following the inspection, inspection results are assessed to determine if any established criteria have been vio- lated, or if conditions have changed that may require a reassessment of the OF. If such changes exist, a reassessment of the OF is completed and the inspection practice modified accordingly. If no such changes or conditions exist, the inspection practice can remain unchanged for the subsequent inspection interval. Using the overall process described above, bridge owners can initiate an RBI practice for bridges in their inventory. However, the process and inspection requirements under an RBI practice may diverge significantly from traditional, calendar-based, and uniform inspection strategies. There- fore, consideration is needed regarding the scope of initial RBI assessments, training for inspec- tors and RAP members, and integration with existing software and databases. The sections that follow discuss these considerations. 4.2 Setting the Scope of the Analysis RBI requires increased planning resources relative to calendar-based or uniform inspection processes. An effective strategy for transitioning from a calendar-based inspection practice to RBI is needed to facilitate the process and ensure adequate resources are available to conduct the necessary assessments. A suitable approach for transitioning an inspection program from a calendar-based, uniform inspection strategy to RBI is to identify those bridges where a reli- ability analysis can most readily be conducted and begin the process by assessing those bridges first. These bridges may be identified by conducting a simple qualitative risk assessment of the overall bridge inventory. This assessment should identify those bridges or family of bridges that are of very common design characteristics, and where significant experience exists regarding the anticipated damage and deterioration patterns. Such an assessment can be rapidly conducted based on general bridge characteristics such as span length, bridge type, number of spans, and C H A P T E R 4 Establishing an RBI Program

22 Proposed Guideline for Reliability-Based Bridge Inspection Practices current condition. For those bridges where past experience is greatest, uncertainty regarding both the development of the damage and the associated consequences is reduced. Bridges that are more complex, suffer from advanced forms of deterioration, or have unique design attributes require a higher level of assessment, as shown schematically in Figure 4. More data and a more sophisticated or more specialized assessment may be required. Therefore, to initiate an RBI practice, bridge owners can conduct a general, fully qualitative assessment of their inventory and assign or determine the scope of the initial assessment to be conducted. Bridges that are of common and simple design, and are in good condition, are identified for analysis first. These bridges can be considered to be in a low risk category because they are of Figure 3. Flow chart showing RBI program activities. Figure 4. Schematic diagram of qualitative risk assessment for a bridge inventory.

Establishing an RBI Program 23 simple design and there is significant experience and confidence in their performance. For exam- ple, bridge owners conduct a simple analysis of their inventory to determine bridges that are multi-girder, short span, and in generally good condition for assessment first. Reliability assess- ment for these bridges may be relatively simple. Conducting the reliability assessment of these bridges first helps develop the RBI practice and develops the knowledge and experience of the RAP members. After this analysis is completed, the assessment moves on to bridges that are more complex, require more data for assessment, or require more sophisticated analysis to determine the factors necessary for a reliability assessment. 4.3 Training Requirements As noted, the inspection planning process is more involved and complex under an RBI scheme relative to a calendar-based inspection planning process. The approach to inspection planning is more focused on inspection needs for the individual bridge. Further, the assessment of reliability characteristics requires an understanding of the approach and the assessment needs. Therefore, training for both members of the RAP and for inspectors that will implement the results of the RBI planning process will be necessary. 4.3.1 Training for RAP Members Participants in the RAP process require training to understand the underlying philosophy and processes involved in conducting RBI planning. This training should provide sufficient knowledge in the theory and underlying approach to RBI planning, address methodologies for expert elicitation, and processes for determining the OFs and CFs required for the analysis. A full understanding of the underlying concepts and reliability theories utilized in the process is necessary to conduct effective assessments. Facilitators that may be used to assist in the expert elicitations and overall reliability assessment should be similarly trained. 4.3.2 Training of Inspectors RBI assessments for inspection planning provide a prioritization of inspection needs for a bridge based on the anticipated or expected damage modes, and the importance of that damage in terms of safety of the bridge. Criteria developed through the RAP process identify key condi- tion attributes used to determine the reliability of individual elements of the bridge. Inspections are necessary that are capable of determining these conditions, and, as such, these inspections are typically more intense than traditional inspections that are intended to report on the general condition of bridge components. Training is therefore required in conducting an element-level inspection to meet the needs of an RBI assessment. Assessments for detecting specific damage modes may be more thorough than under traditional calendar-based practices. For example, training in the detection of fatigue cracks in steel or reliable use of sounding to detect subsurface damage in concrete may be needed. In certain cases, training for inspectors in the application of advanced NDE technologies may be required. Training on the use of NDE technologies is specialized in nature, and certification and training for specific NDE technologies is typically available from commercial sources. Training for bridge inspectors in the underlying philosophy of the RBI approach is also needed. Appropriate implementation of the inspection prioritization developed through the process, and an understanding of the importance of the quality of bridge inspection out- comes, is needed to implement the process and to transition from traditional inspection approaches.

24 Proposed Guideline for Reliability-Based Bridge Inspection Practices 4.4 Software Development and Integration The processes for assessing the OFs, such as identifying and scoring key attributes of bridge elements, can be repetitive once established, and therefore lends itself to software implementa- tions. Many of the attributes identified by the RAP may already be stored in existing databases and bridge management systems. Condition attributes and screening criteria for RBI could be implemented through existing software developed for bridge inspection and storing bridge inspection data, or appropriate software may be developed. Therefore, the process of imple- menting an RBI practice can be simplified by the development of software to more rapidly implement the methodology. Integration with existing software and databases that store relevant information is beneficial.

25 1. American Petroleum Institute (API), API Recommended Practice 580, Risk-Based Inspection. 2002: Washington, D.C., p. 45. 2. National Bridge Inspection Standards, in 23 CFR part 650. 2004: USA., p. 74419–74439. 3. FHWA, Revisions to the National Bridge Inspection Standards (NBIS). 1988: p. 21. 4. American Bureau of Shipping (ABS), Surveys Using Risk-Based Inspection for the Offshore Industry. 2003: Houston, TX. 5. ASME, Risk-Based Inspection: Development of Guidelines. General Document. 1992: p. 155. 6. ASME, Inspection Planning Using Risk-Based Methods. 2007: p. 92. 7. FHWA, Recording and Coding Guide for the Structural Inventory and Appraisal of the Nation’s Bridges, U.S.DOT., Editor: 1995:. 8. AASHTO, AASHTO Bridge Element Inspection Manual. 2010, AASHTO Publications: Washington, D.C. p. 170. 9. AASHTO, The Manual For Bridge Evaluation. 2008, AASHTO Publications: Washington, D.C. 10. Washer, G. A., and Chang, C. A., Guideline for Implementing Quality Control and Quality Assurance For Bridge Inspection. 2009, Transportation Research Board: Washington, D.C. p. 65. References

26 27 A 1 Introduction 28 A 2 Damage Modes 28 A 2.1 Expert Elicitation for Credible Damage Modes 29 A 2.2 Example of Soliciting Expert Judgment for Damage Modes 30 A 3 Element Attributes 31 A 3.1 Screening Attributes 32 A 3.1.1 Qualitative Assessment of Elements and Details 32 A 3.2 Identifying Key Attributes 33 A 3.3 Ranking Attributes 34 A 4 Occurrence Factor Assessment 34 A 4.1 Estimating the Occurrence Factor 34 A 4.2 Calibration of Scoring Regime 35 A 4.3 Occurrence Factor Scale Numerical Estimates 36 A 4.4 Use of Deterioration Rate Data 37 A 4.5 Use of Surrogate Data 38 A 4.6 Rationale and Criteria Based on RAP Assessments A P P E N D I X A Guideline for Evaluating the Occurrence Factor

Guideline for Evaluating the Occurrence Factor 27 A 1 Introduction The Occurrence Factor (OF) is used within an RBI to estimate the likelihood of serious dam- age (i.e., failure) developing in a bridge element during a specified time interval, based on engi- neering rationale. This rationale is developed through a systematic process that considers and documents the anticipated damage modes for bridge elements. The characteristics, or attributes, of bridge elements that contribute to their reliability, considering the expected damage modes, are identified. The damage modes and attributes are identified through an expert panel process described herein, and subsequently used in a rational process that identifies those bridges with elements that are highly reliable and durable, and those bridges with elements that are more likely to suffer from deterioration and damage. The overall process for estimating the OFs is as follows: 1. Identify the likely damage modes that will affect a bridge element from commonly known damage modes, past experience, and engineering judgment. 2. Identify attributes that contribute to the reliability and the durability of the element consider- ing the damage modes identified. 3. Rank the importance of each attribute’s influence on the reliability and the durability of the bridge element. 4. Develop rationale based on the damage modes and attributes of the bridge element to esti- mate the likelihood of serious damage (i.e., failure) occurring during the specified interval. An empirical scoring procedure based on the key attributes identified for a given element is used to provide a rational method of estimating the OF. The analysis can be used to construct criteria that can be applied to individual bridges, or groups of very similar bridges, to categorize the likelihood of serious damage (i.e., “failure”) occurring in the next 72-month time frame into one of four categories, ranging from “remote” to “high,” i.e., the OF. The OF represents a probability of failure (POF) estimate over a time interval of 72 months. This time period was selected based on engineering factors that included prior research, analysis of data from the National Bridge Inventory (NBI), expert judgment, and data from corrosion and damage models. It was also selected as a time interval for which an engineer could reason- ably be expected to estimate future performance within four fairly broad categories, ranging from “remote” to “high,” based on key attributes that describe the design, loading, and condition of a bridge or bridge element. In addition, this time interval was selected to provide a suitable balance between shorter intervals, when the POF could be unrealistically low due to the typi- cally slow progression of damage in bridges, or longer intervals, where uncertainty would be increasingly high. The analysis provides the rationale for categorizing the OF on a rating scale from “remote,” when the likelihood is extremely small such that it would be unreasonable to expect failures, to “high,” where the likelihood is increased. This rating scale is shown in Table A1. In some cases, the OF may be an estimate of the likelihood of a certain adverse event occurring that results in a failure, such as impact from an over-height vehicle or an overload. Table A1. Occurrence factor rating scale for RBI. Level Category Description 1 Remote Remote likelihood of occurrence, unreasonable to expect failure to occur 2 Low Low likelihood of occurrence 3 Moderate Moderate likelihood of occurrence 4 High High likelihood of occurrence

28 Proposed Guideline for Reliability-Based Bridge Inspection Practices The following sections describe how a Reliability Assessment Panel (RAP) identifies the dam- age modes to be assessed, determines important attributes for each damage mode, and ranks and scores those attributes to support assessment of an individual bridge or families of bridges of nearly identical attributes, damage modes, and design. The RAP is an expert panel assembled by the bridge owner as described in section 2.4 of the main report. A 2 Damage Modes The first step in the process is to answer the question “What can go wrong?” For most common bridges, the damage modes that affect the bridge are well known. Spalling and cracking of the concrete as a result of corrosion, or section loss and fatigue cracking in steel elements, are typi- cal examples. The RAP, through a consensus process, develops a listing of the credible damage modes for the elements of a bridge or a family of bridges being assessed. A credible damage mode is one that could reasonably or typically be expected to occur during the service life of the bridge element. Current and past research and experience should be considered in developing the listing. An expert elicitation process described in section A 2.1 may be used to identify the typical damage modes for consideration. This process may also be used to identify unusual or uncommon damage modes that may be relevant for a particular bridge inventory. Table A2 lists damage modes that may be identified by the RAP, as examples to illustrate typical damage modes for several common bridge elements. A 2.1 Expert Elicitation for Credible Damage Modes In many cases, the credible damage modes for a given bridge element may be readily identi- fied from past experience and engineering knowledge. In other cases, it may be necessary for the RAP to form a consensus on the credible damage modes for a given element. To identify damage modes that are specific to the type of bridge and elements being considered, the RAP can utilize a process to elicit the expert judgment of the panel based on their experience and knowledge. The process is an expert elicitation of judgments from the panel that consists of the following: 1. Identify the element scenario: The first step in the process is to frame the problem for the panel. This includes describing the element under consideration, including the material and known design parameters. The operational environment for the element should also be described, such as the environment and loading, especially if the operational environment is atypical or unique. For example, if the element under consideration is a concrete beam located in an aggressive coastal environment. Table A2. Typical damage modes for common bridge elements. Element Damage Modes Steel Girder Corrosion damage/section loss Fatigue cracking Fracture Impact damage Prestressed Girder Corrosion damage (spalling/cracking) Strand fracture Shear cracking Flexural cracking Impact damage Piers and Abutments Corrosion damage (spalling/cracking) Damage to bearing areas Unexpected settlement/rotation

Guideline for Evaluating the Occurrence Factor 29 2. Identify damage modes: The facilitator poses a question to the RAP such as: “The inspection report indicates that the element is rated in serious condition. In your expert judgment, what is the most likely cause (i.e., damage mode) that has produced/resulted in this condition?” This question is intended to elicit from the panel a listing of damage modes that are likely to occur for the element. Each expert is asked to independently list the damage modes he/she judges are most likely to have caused the element to be rated in serious condition. The expert records each damage mode he/she identifies, and provides an estimate of the relative likelihood that the damage mode was the cause. This is done by assigning relative probabilities to each damage mode, typically with a minimum precision of 10% (the sum of the ratings should be 100%). The expert notes any supporting rationale for their estimate. The individual results from each member of the RAP are then aggregated to evaluate consensus among the panel on the most likely damage modes for the element. An iterative process may be necessary to develop consensus on the credible dam- age modes for a given bridge element. However, for most elements, the damage modes are well known and consensus can be reached quickly. A 2.2 Example of Soliciting Expert Judgment for Damage Modes This section provides an example of the process for eliciting expert judgment from the RAP for a typical bridge element. In this example, the RAP is provided with the following descrip- tion for a steel bridge member: The element under consideration is a painted, rolled steel girder in a simply supported, multi-girder bridge with a typical span length, in a moderate environment. If you were told this girder is rated in serious condition, what would be the most likely cause of this condition? Each member of the RAP is then asked to list the damage modes that they identify as the most likely causes (e.g., cracking, section loss) for the member condition, and estimate its relative likelihood of being the cause, relative to other damage modes they identify. The results of this independent exercise are then aggregated as shown in Table A3, showing illustrative results from a six member RAP team assessing the given element scenario. Following the independent elicitation, the panel discusses the results of the assessments. Any damage mode with an average score of less than 10% may be assessed to determine if that dam- age mode is credible for the given scenario. Rationale for inclusion or exclusion of the particular damage mode should be recorded. Any damage modes with variance of >20% from the average are also discussed, and RAP members are provided an opportunity to revise their individual ratings based on the discussions. In this steel girder example, the panel considers the damage mode of corrosion damage/section loss to be most likely to have resulted in severe damage to the steel girder. Less likely damage modes include fatigue, overload, and impact damage. Each credible damage mode identified will be assessed by the RAP to determine its OF. Table A3. Expert elicitation of damage modes for steel girders. Damage Expert 1 Expert 2 Expert 3 Expert 4 Expert 5 Expert 6 Average Corrosion/ section loss 60% 60% 50% 50% 70% 50% 57% Fatigue 30% 30% 30% 20% 10% 20% 23% Overload 10% 10% 10% 20% 20% 20% 15% Impact 0% 0% 10% 10% 0% 10% 5% Sum 100% 100% 100% 100% 100% 100% 100%

30 Proposed Guideline for Reliability-Based Bridge Inspection Practices The elicitation process is repeated for each key element of the bridge to develop a listing of damage modes to be considered in the analysis. For example, considering a typical steel girder bridge with a bare concrete deck and concrete piers and abutments, damage modes for each element of the bridge that might be identified by a RAP are shown in Table A4. For the deck in this illustration, the most common damage mode is identified as spalling of the deck due to corrosion damage of the reinforcing steel; widespread cracking, and damage due to alkali-silica reactivity (ASR) and/or freeze-thaw cycles. For the steel girder, corrosion damage (section loss) is identified as the most likely damage mode; fatigue cracking, fracture, and impact are also identified by the RAP. For the piers and abutments, damage modes included corrosion damage that results in spalling, damage to the bearing areas (beams seats, for example), and unexpected settlement or rotation. Such a listing is developed through a consensus process by the RAP for a specific bridge and element types under consideration, as previously discussed. Once this listing of damage modes has been identified, the next step in the process is to iden- tify key attributes that contribute to the reliability and durability of the element, considering these damage modes. A 3 Element Attributes “Attributes” are characteristics of a bridge element that affect is reliability. These attributes are typically well-known parameters affecting the performance of bridge elements during their service lives. For example, bridge elements can have “good attributes” that are known to provide good service-life performance. A bridge deck can have “good” qualities such as having adequate concrete cover and use of epoxy-coated reinforcing steel for corrosion resistance. Alternatively, bridges may have qualities or attributes that contribute to more rapid deterioration or increased likelihood of damage. Using the concrete deck example, heavy use of de-icing chemicals, mini- mal concrete cover, and unprotected reinforcement would be examples of attributes that con- tribute to more rapid deterioration. For a steel girder, fatigue-prone details may be an attribute indicating increased likelihood of damage. The identification of key attributes is simply a listing of these attributes and a relative ranking of their importance in terms of the reliability and the durability of the element. These attributes can be generally grouped into three categories: Design, Loading, and Condi- tion attributes. Design attributes are characteristics of a bridge element that are part of the ele- ment’s design. Design attributes are usually intrinsic characteristics of the element that do not change over time, such as the amount of concrete cover or material of construction [concrete, high performance concrete (HPC), etc.]. In some cases, preservation or maintenance activities Table A4. Example damage modes for a steel girder bridge. Element Damage Modes Bare Concrete Bridge Deck Spalling resulting from steel corrosion Widespread cracking Rubblization of concrete due to freeze/thaw damage or ASR Steel Girder Corrosion damage Fatigue cracking Fracture Impact damage Piers and Abutments Spalling resulting from corrosion Damage to bearing areas Unexpected settlement/rotation

Guideline for Evaluating the Occurrence Factor 31 that contribute to the durability of the bridge element may be a design attribute, such as the use of penetrating sealers as a preservation strategy. Loading attributes are characteristics that describe the loads applied to the bridge element that affect its reliability. This may include structural loading, traffic loading, or environmental load- ing. Environmental loading may be described in macro terms, such as the general environment in which the bridge is located, or on a local basis, such as the rate of de-icing chemical application on a bridge deck. Loading attributes describe key loading characteristics that contribute to the damage modes and deterioration processes under consideration. Condition attributes are characteristics that relate to the current condition of a bridge or a bridge element. These can include the current element or component level rating, or a specific condition that will affect the reliability of the element. For example, if the damage mode under consideration is concrete damage at the bearing, the condition of the bridge joint may be a key attribute in determining the likelihood that severe corrosion will occur in the bearing area. Relevant attributes are identified for the damage modes and underlying deterioration mecha- nisms determined by the RAP. In many cases, attributes are well-known characteristics of bridges and bridge elements that contribute to the reliability and durability of the elements. However, because bridge designs, environments, and management policies differ, attributes and their relative importance may also differ between bridge owners. Therefore, it is necessary that the RAP identify those attributes that contribute most significantly, including any special or unique attributes that might contribute significantly (either positively or negatively) to the likelihood of damage for bridges in their inventory. Attributes that are not relevant or do not have significant impact on durability and reliability should not be included in the analysis. A 3.1 Screening Attributes Screening attributes can be used to quickly identify bridges or elements that should not be included in a particular analysis, either because they already have significant damage or they have attributes that are outside the scope of the analysis being developed. Screening attributes are typically attributes that: • Make the likelihood of serious damage occurring very high. • Make the likelihood of serious damage occurring unusually uncertain. • Identify a bridge with different anticipated deterioration patterns than other bridges in a group. Once the attribute listing has been completed, attributes that match these criteria can be identified. The RAP should identify the appropriate value or condition for the attribute to use as a screening tool. In any scoring scheme there is the possibility, and hence a concern, that the value of key attributes can be diminished when the scoring for all of the relevant attri- butes are combined. Screening attributes are useful to ensure key conditions are identified, to address this concern. For example, if considering the likelihood that the steel bridge will suffer corrosion dam- age that reduces its rating to a 3, and the current rating is 4, the RAP may consider that such condition indicates that there is a high likelihood of further damage developing over the next 72-month period, regardless of other attributes. In such a case, the analysis can move forward to an assessment of the consequences without assessing the specific attributes of the element, since the likelihood has already been assessed to be high. Design features may be useful as screening criteria, particularly if the features result in the likelihood of serious damage being unusually uncertain. For example, for bridges that possess details susceptible to Constraint-Induced Fracture (CIF), there is a high potential for sudden

32 Proposed Guideline for Reliability-Based Bridge Inspection Practices brittle fracture. For fracture-critical bridges in particular, inspection will provide no protection as the CIF occurs without any warning and before any detectable cracks are observed. Hence, it would be prudent to screen these bridges from the analysis, because the likelihood of serious damage is unusually uncertain. Another strategy, such as retrofitting the critical details, should be performed to ensure safety. Another example would be to screen steel beam elements in bridges that have open decking. Since the open decking allows drainage directly onto the steel beams, the deterioration of these bridges would not be similar to steel beams with typical concrete decks. Therefore, it would be prudent for these bridges to be screened from the analysis of steel beam bridges, as they may require separate analysis. It may be appropriate to treat these bridges as a separate group, devel- oping the analysis to consider key attributes of those bridges with open decking. In some cases, it may be more practical to screen bridges from the analysis entirely through a qualitative reliability assessment of the overall inventory, as described in the following section. A 3.1.1 Qualitative Assessment of Elements and Details A simple qualitative assessment can also be used early in the RAP process to identify appro- priate families or groups of bridges to be analyzed. This tool can be used to separate potentially problematic details or elements that may require more in-depth analysis. These elements may include, for example, rocker bearings in long-span bridges, modular expansion joints, or other details that have the potential to affect the reliability of a bridge uniquely. The qualitative assess- ment uses a simple three-level scale, as shown in Table A5. This tool can be used to perform an assessment of a bridge inventory and sort bridges that include attributes that are perceived to have low reliability or require special analysis. The assessment is useful for identifying bridges that can be easily assessed from those for which more detailed or individual assessments may be required. For example, assume the RAP is going to assess multi-girder rolled beams, but it considers those beams with rocker bearing to require special analysis and to potentially have low reliability (relative to bridges with other bearing types); these bridges are simply screened from the process using the qualitative assessment, such that the balance of the bridges in that family can be assessed appropriately. A separate analysis that addresses this specific attribute can then be developed, if necessary. This qualitative screening process would typically be used early in the reliability assessment process to identify an appropriate family or group of bridges and make assessments more efficient. A 3.2 Identifying Key Attributes Attributes can be identified generally through a variety of means such as past performance, experience with the given bridge element, previous and contemporary research, analysis of his- torical performance, etc. While there are potentially many attributes that contribute, in some way, to the durability and reliability of a bridge element, it is necessary to identify those attributes that have the greatest influence on the future performance of an element. Key attributes for a Table A5. Qualitative reliability scale for screening details. Relative Reliability High Moderate Low

Guideline for Evaluating the Occurrence Factor 33 given damage mode can be identified through expert elicitation of the RAP. For example, the facilitator could ask the following question pertaining to a particular damage mode, X: • Consider damage mode X for the subject bridge element. If you were asked to assess the likeli- hood of serious damage occurring in the next 72 months, what information would you need to know to make that judgment? The resulting input from the RAP can be categorized appropriately and ranked according to the relative importance of the attribute for predicting future damage for the identified damage mode and element. Rationale for each attribute should be documented. Many of the most com- mon attributes are described in Appendix E, and can be documented by reference. For attributes not included in Appendix E, a brief summary of the rationale for the attribute should be devel- oped and recorded by the RAP. As an example, Table A6 illustrates typical attributes identified by a RAP for corrosion damage on a steel girder element. Based on an expert elicitation, the primary attributes that contribute to the likelihood of serious corrosion damage developing for a steel girder bridge element include design attributes, loading attributes, and condition attributes, as shown in the table. The ratio- nale for these attributes is relatively simple and straightforward. For example, the presence of deck joints and the quality of the drainage system may indicate whether or not the bridge has deck drainage that is likely to spill de-icing chemicals directly onto the steel girder, thereby result- ing in an increased likelihood of corrosion occurring. Built-up members are more likely to suffer crevice corrosion and would therefore be more likely to suffer serious corrosion damage than a rolled or welded section. The attribute of deck type considers if there is open decking that allows de-icing chemicals to drain directly onto the girder, thereby increasing the likelihood of corro- sion damage, etc. These attributes are identified by the RAP by applying common engineering knowledge to develop criteria from which a steel bridge element can be assessed to determine if it is likely to suffer serious corrosion damage, or if corrosion damage is unlikely. Elements that have little exposure to de-icing chemicals, are in mild environments, and are currently in good condition may be unlikely to develop serious corrosion in the near future. Conversely, a steel element with active corrosion present, which is in an aggressive environment, and/or is exposed frequently to de-icing chemicals, is more likely to develop serious corrosion damage. A 3.3 Ranking Attributes Once the key attributes have been identified, the attributes are ranked on a simple three-level scale according to their importance in assessing the reliability of a bridge element. The ranking is based on the consensus of the RAP. This scale, shown in Table A7, is used to rank a particular attribute’s importance as high, moderate, or low. Once ranked, the attributes are assigned a point value corresponding to their importance, to be used in the attribute scoring methodology that Table A6. Attributes related to the damage mode of corrosion for a steel girder. Design Attributes Loading Attributes Condition Attributes Deck Joints/Drainage Macro- Environment Existing Condition Built-Up Members Micro-Environment Joint Condition Deck Type Maintenance Cycle Age/Yr of Construction Condition History/ Trend Debris Accum.

34 Proposed Guideline for Reliability-Based Bridge Inspection Practices supports the RAP assessment of the OF. For attributes that are ranked with high importance, a scale of 20 points can be assigned, 15 points for an attribute that has a moderate importance, and 10 points for an attribute that plays a minor role, but is still an important indicator. For example, for the corrosion of a steel beam, a leaking joint which results in drainage of de-icing chemicals directly onto the superstructure is highly important in assessing the likelihood of serious cor- rosion damage occurring. Therefore, this attribute would be assigned a 20 point scale. Age of the structure contributes to the likelihood of corrosion damage, but to a much lesser extent, relatively, such that it would have 10 points allocated. Maintenance cycle, built-up members, and debris build-up are moderate indicators; these may be assigned a point scale of 15 points. Once the importance of the attribute is identified, different conditions or situations may be described to distribute points appropriately based on the engineering judgment of the RAP. Again, a simple high-, moderate- and low-ranking model should be used to distribute scores among different conditions or situations that are appropriate for a given attribute. Depending on the number of different conditions or situations, scoring may be distributed over two, three, or four different levels for a given attribute. Using a joint as an illustration, if the joint is leaking or can reasonably be expected to be leaking, it will have the highest effect and might be scored the full 20 points. If the joint is debris-filled or exhibiting moderate leakage, a score of, for example, 15 points may be appropriate; if there is a joint, but it is not leaking, a score of 5 points may be assigned. If the subject bridge is jointless, a score of 0 points may be used. The distribution of scoring for a particular attribute is determined by the RAP. Numerous examples for scoring vari- ous attributes are included in the Attribute Index and Commentary located in Appendix E. The RAP should assess if the suggested scores in Appendix E are appropriate, based on the character- istics of the bridges being assessed, and assign appropriate scoring regimes for attributes selected. A 4 Occurrence Factor Assessment A 4.1 Estimating the Occurrence Factor Once the appropriate attributes have been identified and ranked for a given element, the attributes are used to estimate the appropriate OF for a that element. A simple scoring procedure is developed to evaluate the reliability characteristics of a given element based on the attributes and their relative ranking, as described above. The developed scoring procedure provides a data model that is used to assess the OF. Attributes scoring sheets may be used to record the relative scoring of the attributes for a given element, or the data model may be implemented through suitable software. Illustrative examples are included in Appendix F. A 4.2 Calibration of Scoring Regime Once the appropriate attributes are selected and ranked, the overall outcome of the scoring procedure (i.e., data model) should be tested to ensure results are adequate for categorizing the subject elements. In some cases, the weighting of particular attributes may need to be increased Table A7. Ranking scales for key attributes. Ranking Descriptor Total Points High 20 Moderate 15 Low 10

Guideline for Evaluating the Occurrence Factor 35 or decreased to provide suitable results. Since operational environments and design and con- struction practices vary, rankings for attributes and associated values may need to be adjusted. When a large number of attributes are identified, the relative weight of the most important attributes becomes diminished relative to the overall scoring, and may need to be adjusted to appropriately characterize the anticipated reliability of the element. Screening attributes can also be used for this purpose. Sensitivity studies and Monte Carlo simulation may also be used to assess the relative weights designated for attributes and calibrate the scoring regime developed. The effectiveness and accuracy of the scoring regime developed can also be evaluated using back-casting, a process for analyzing historical inspection records to verify the effectiveness of the data model (i.e., attributes and scoring) developed. In a back-casting assessment, the attri- butes and scoring regime are applied to historical inspection records to assess their effectiveness for identifying the likelihood of serious damage occurring. Regardless of the method(s) used to calibrate the data model, engineering judgment should be used to verify the adequacy of the data model developed. A 4.3 Occurrence Factor Scale Numerical Estimates The OF is a qualitative assessment of the likelihood of failure occurring during the next 72 months. Four categories are used to characterize the likelihood considering a particular ele- ment and damage mode. Table A8 includes numerical ranges that could be used to describe the OF scale quantitatively. Such numerical values provide ranges or target values for the qualitative rankings that could be used to map quantitative data, if these were available, to the qualitative rating scales. Failure of a bridge element is a relatively rare event, and design and construc- tion details vary widely. As a result, relevant and verifiable frequency-based probability data are scarce. In some cases, modeling may be used to provide an estimate of a particular failure fre- quency or probability. Probabilistic models or assessments may also be developed for a particu- lar bridge element or elements. The numerical values shown in Table A8 are target values that could be used to map these data or models to the qualitative scales used for analysis. Providing a quantitative estimate of the OFs allows for the data from the probabilistic analysis to be incor- porated directly in the reliability-based bridge inspection practice. These numerical categories can also provide a framework for future development of models or data derived from analysis of the deterioration patterns in a particular bridge inventory. An estimate of a particular damage mode having a “low” likelihood is somewhere between 1/1,000 and 1/10,000. Although the quantitative probability is not necessarily known, engi- neering judgment supported with an evaluation of the reliability characteristics of the ele- ments is adequate to differentiate between different categories: for example, a likelihood in the “low” category, where the chances are 1/1,000 or less, versus something in the “moderate” Table A8. Occurrence Factor categories and associated interval estimates. Level Category Description Likelihood Remote Remote probability of occurrence, unreasonable to expect failure to occur 1/10,000 Low Low likelihood of occurrence 1/1,000-1/10,000 Moderate Moderate likelihood of occurrence 1/100- 1/1,000 High 1 2 3 4 High likelihood of occurrence >1/100

36 Proposed Guideline for Reliability-Based Bridge Inspection Practices category, where the likelihood in less than 1/100 but greater than 1/1,000. Estimates from deterioration rate information or from statistical modeling can also be used to support the categorization of the OF. The quantitative description can be also be used as a vehicle for expert elicitation by using common language equivalents for engineering estimates. For example, if you asked an expert to estimate the probability of serious corrosion damage (widespread spalling, for example) for a particular bridge deck given its current condition, a common engineering response might include a percentage estimate, for example, less than 0.1% chance or less than 1 in a thousand. This estimate can then be mapped to the qualitative scale as being “low.” Such estimates are typically very conservative, particularly for lower, less likely events. For engineering estimates of the likelihood of a failure occurring for a given bridge element, the qualitative scale can be interpreted as shown in Table A9. A 4.4 Use of Deterioration Rate Data Data on the previous performance of bridge elements can provide some insight into the likeli- hood of damage occurring for a bridge element. Such data can provide supporting information for decision making regarding the appropriate OFs for a family of similar bridge element types. The user is cautioned that deterioration rate data records only historical events that may not reflect the rate or likelihood of future events. For example, a state may have never had corro- sion damage occur in prestressing tendons; however, this provides little insight into how likely it is that tendon damage will occur in the future. It may be that the population of bridges from which the data is obtained has simply not reached the age where tendon damage would become apparent. Further, deterioration rate data based on condition states or condition ratings may provide little insight into the deterioration mechanisms that caused the condition states or ratings to change. Caution and careful judgment should be used in determining the relevance of the deterioration data to the particular bridge under consideration. Considerations for utilization of deterioration rate data include: • Similarity of Operational Environment: The RAP should consider if the particular bridge under consideration shares the same operational environment as the elements from which data were obtained. Key elements of the operational environment include the average daily traffic (ADT), average daily truck taffic (ADTT), macro-environment of the bridges (severe environment vs. benign environment), micro-environment (salt application, joint and drain- age conditions, exposure to overspray), and typical maintenance and management. • Similarity of Key Attributes: Key attributes that affect the damage modes and mechanisms for the bridge element should be similar for the bridge under consideration to those from which deterioration rate data were obtained. This may include materials of construction, design attributes, and condition attributes. Quality of construction and years in service may also be a factor. Table A9. Percentage estimates for Occurrence Factor ratings. Qualitative Description Expressed as a Percentage Remote 0.01% or less Low 0.1% or less Moderate 1% or less High >1%

Guideline for Evaluating the Occurrence Factor 37 Deterioration rate data typically describe the mean or average behavior of the bridge element based on the observed behavior of a population of similar elements. Statistical descriptors of the dispersion of the data, such as the standard deviation, may be provided and then used as indica- tors of the variability of the data. Applying such data to a specific bridge assumes that the specific bridge has the same design, operational environment, and attributes as those in the larger popu- lation from which the statistics were derived. Attributes identified through the RAP process may be used to judge if a particular bridge or family of bridges could be expected to perform above the average or mean, or below. Statistical data from a bridge management system or other databases can also be used to inform this process if it is available. This data can be useful in determining the damage modes and the overall deterioration behavior of similar bridge elements in the past. However, this data should not be used exclusively because past experience does not necessarily indicate what would occur in the future. Therefore, it is important that the RAP utilize their collective engineering judgment, expe- rience, and rationale for identifying and assessing damage modes that can affect bridge elements. Lastly, when using such data, one would have to decide which data to use: the mean, or say, two standard deviations below the mean. If the mean is used, there may be a 50-50 chance that the bridge being assessed will deteriorate more quickly than predicted by using the mean deteriora- tion data. However, using some confidence limit, say 2 standard deviations from the mean, may be overly conservative and result in all bridges, good or bad, having unrealistically high estimates of likelihood. Thus, using such data without the ability to also consider or incorporate specific information (condition, design data, details, etc.) from the bridge under consideration must be done with caution, and with a full understanding of the ramifications of such an approach. A 4.5 Use of Surrogate Data For many bridges, the use of “surrogates” for the attributes identified in the reliability analysis may be considered to improve the efficiency of the analysis for larger families of bridges. As used herein, “surrogate” refers to specific data that can be used to either infer or determine another piece of information that is required for the reliability assessment. For example, assume a frac- ture critical bridge was designed and built in the year 2000, which is well after the implementa- tion of the AASHTO/AWS Fracture Control Plan. This information can be used to determine that the steel must at least meet certain minimum toughness requirements, and the bridge meets modern fatigue design requirements. Note that this was determined only from the date of con- struction and with no detailed review of the design calculations or specifications. As stated, the use of surrogates is particularly attractive when identifying and assessing a fam- ily of bridges. Design and loading attributes identified by the RAP are typically static in nature, that is, they do not change over time. The condition attributes will typically change over time, as damage accumulates and deterioration mechanisms manifest. However, when elements are in generally good condition, specific condition attributes identified by the RAP may not require individual assessment for each bridge or family of bridges; the previous inspection results can simply be used as a surrogate for the individual attributes. This will typically allow for larger groups of bridges of similar design to be grouped into a particular inspection interval, based on the criteria developed by the RAP. For example, again considering steel bridges built to modern design standards, it is known that the design attributes that would increase the likelihood of fatigue cracking and fracture have been mitigated through improvements in the design, fabrica- tion, and construction process. The condition attributes that are required to assess the reliability of the element would include the presence of fatigue cracks due to out-of-plane distortions, fatigue cracking due to primary stresses, and corrosion damage. However, if the component rating is 7, in good condition according the NBIS scale, or CS 1 in an element-level scheme, the existing ratings can be used as a surrogate for the condition attributes. Note: This assumes the

38 Proposed Guideline for Reliability-Based Bridge Inspection Practices inspection result is from an RBI procedure, i.e., the inspection was capable of identifying the neces- sary condition attributes. This allows all bridges that are of this same rating (and similar design, loading, and condition attributes) to be treated collectively in a process that is data-driven and does not require much detailed analysis of individual bridges. If the condition rating or condi- tion state changes, then the bridges can be reevaluated, according to the RAP criteria. If the condition does not change between periodic inspections, reassessment may not be necessary. It is important to note that this process is significantly different than assigning an inspection interval based simply on the current condition of the bridge, for example, deciding to inspect all steel bridges with a rating of 7 on a longer interval than all of those rated a 6. The RAP analysis forms a rationale that identifies not only the current condition attributes that affect the reliability of the element, but also the design and loading attributes of the bridge or bridge element that affect the potential for damage to occur. This RAP evaluation forms an engineering rationale for the decision-making process that considers not only the condition of the element, but also the damage modes and the potential for that damage to occur. For element-level inspection schemes, the attributes identified by the RAP may map directly to an element and element condition state. For example, consider that the RAP identifies leak- ing joints as an attribute driving the likelihood of section loss in the bearing area of a steel beam. The element condition state (joint leaking) is recorded in the inspection process and can be used as a criterion for that attribute score. In some cases, all of the attributes identified by the RAP as being critical to the likelihood of failure of an element may be included in a comprehensive element-level inspection process, in other cases, they may not. For NBI-based inspection schemes, attributes identified by the RAP may map to sub-element data collected in addition to the required condition ratings for the primary components of the bridge. These data could be used if it is collected under a standardized scheme for rating and data collection for the sub-elements. For the primary components, the generalized nature of the component rating makes this more difficult for specific attributes. Mapping of the attributes from the RAP analysis to the elements, sub-elements, or element condition states should not be performed until the RAP analysis has been completed indepen- dently. In some cases, the RAP analysis may identify attributes or factors not presently included in the available data, and these data may need to be obtained from other sources. For example, for the case of fatigue cracking in a steel beam, element condition states would indicate fatigue crack- ing, but not the presence of fatigue sensitive details, i.e., the potential for cracking may be high, even though no cracking is currently present. This is an important consideration in the assessment of appropriate scope and interval of inspection. This data may be readily available in the bridge file, or may need to be ascertained from design plans, records, or other data on the bridge design. In any case, the RAP analysis shall not be constrained by the data presently available; the RAP should identify what data is needed and then assess if that data is readily available. In some cases, addi- tional data may need to be collected to support the analysis. A 4.6 Rationale and Criteria Based on RAP Assessments The RAP assessment for a given bridge or a family of bridges provides an engineering ratio- nale for decision making regarding the appropriate inspection interval and scope. The effects of design, loading, and condition attributes on the potential for failures are considered and docu- mented through the process. For most bridges, the design attributes and loading attributes will not change over time. The RAP assessment should include criteria for modifying the selected inspection interval, and/or for reassessment of a bridge, based on the results of the RBI. These criteria will typically be based on the condition attributes identified during the RAP assessment. If loading conditions change significantly, reassessment may be necessary.

39 40 B 1 Introduction 41 B 1.1 Definitions 41 B 2 General Descriptions of Consequence Scenarios 42 B 2.1 Low Consequence Event 42 General Description 43 Requirements for Selection 43 B 2.2 Moderate Consequence Event 43 General Description 43 Requirements for Selection 46 B 2.3 High Consequence Event 46 General Description 46 Requirements for Selection 48 B 2.4 Severe Consequence Event 48 General Description 48 Requirements for Selection 50 B 3 Use of Expert Elicitation for Determining the Consequence Scenario 50 B 4 References A P P E N D I X B Guideline for Evaluating the Consequence Factor

40 Proposed Guideline for Reliability-Based Bridge Inspection Practices B 1 Introduction Within an RBI, the Consequence Factor (CF) is used to categorize the outcome or the result of the failure of a bridge element due to a given damage mode. For example, brittle fracture is one of the key damage modes pertaining to steel bridges. Should brittle fracture of a girder occur, the next logical question becomes, “what is the consequence?” This would obviously depend on the specific scenario for the fracture. If the member were classified as fracture critical, such an event may be catastrophic, or one that would be considered to be a severe consequence. How- ever, if the girder were one member of a multi-girder short-span bridge, the consequence of that fracture would likely to be much less serious, perhaps requiring a lane closure or even tempo- rary closure of the bridge, or a high consequence. (“Multi-girder” bridges described herein are bridges with four or more main load bearing members.) In fact, in some cases, such an event may only have moderate consequences. The CF is used to categorize the consequence of failure of a bridge element into one of four broad categories: Low, Moderate, High, and Severe. Table B1 indicates the general descriptions for each of the CF categories used for the RBI assessment. The general descriptions are indi- cated in terms of safety and serviceability of the bridge, graduated with qualitative descriptions. Both long- and short-term consequences should/may be considered. To assess the appropriate category for a particular element and damage mode, typical scenarios or outcomes of a failure must be considered. In some cases, there may be a single scenario that could result from the failure of an element; in other cases, more than one possible scenario needs to be considered. Using the example of brittle fracture of a single beam in a multi-girder, short- span bridge as noted above, it is unlikely that the result from a brittle fracture is a low consequence, which has a minor effect on serviceability and no effect on public safety. It is much more likely that such a fracture may have a moderate consequence, which has a moderate effect on serviceability and a minor effect on public safety. It is also possible that the fracture will have a high consequence, which has a major effect on serviceability and a moderate effect on public safety, and may require urgent repair. There may also be a remote possibility that the fracture causes a catastrophic col- lapse, or a severe consequence. It is necessary to determine which of these consequences is most realistic and establish sufficient rationale based on experience, engineering judgment, and/or theo- retical analysis to exclude those consequences that are not credible scenarios. While the immediate effect on the structure is primarily what is evaluated (e.g., collapse after member failure), it is also appropriate to consider longer term consequences. For example, in the example cited above, if the fracture were to result in a lane closure on a portion of interstate that car- ries a very high ADTT, the consequence on the traveling public could be high to even severe, though no concerns regarding the structural performance of the bridge may actually exist. Rather, the result- ing impacts on serviceability could be such that a more frequent inspection interval is justified. There are many cases in which the critical consequence is obvious. There are also many that require considerable judgment and/or analytical effort to ensure the appropriate CF is selected. Level Category Consequence on Safety Consequence on Serviceability Summary Description 1 Low None Minor Minor effect on serviceability, no effect on safety 2 Moderate Minor Moderate Moderate effect on serviceability, minor effect on safety 3 High Moderate Major Major effect on serviceability, moderate effect on safety 4 Severe Major Major Structural collapse/loss of life Table B1. General description of the CF categories.

Guideline for Evaluating the Consequence Factor 41 In these cases, it is important that the rationale used to support the determination is recorded. There are many situations in which analysis and/or experience can be used to justify selecting one scenario over another. However, the level and the type of analysis that is required must be defined, as well as what constitutes sufficient “experience” and when it is appropriate to use experience to justify the categorization of the consequence. This section describes, through example, situations in which analysis or experience is needed to justify the selection of an appropriate CF. Since not every situation can be included or foreseen, the reader must use the information provided and consider it a road map or framework on how to select the appropriate consequence. The Reliability Assessment Panel (RAP) may use this guid- ance to develop basic rules or common practices for very common scenarios they anticipate in the analysis. The RAP should consider existing rules, policies, or common practices within its state regarding the considerations for identifying structural redundancy and other factors that may influence the assessment of the consequences. If no rules, policies, or common practices exist, it may be necessary for the RAP to develop its own basic guidelines before performing consequence assessments. B 1.1 Definitions This section provides definitions for the terms “analysis” and “experience” as used in the context of this document to support the selection of the most appropriate CF. Analysis: As used herein, refers to the effort put forth using accepted methods of structural analysis to quantitatively evaluate the outcome of a given event or scenario based on certain initial conditions. Laboratory and field experimental testing are also acceptable methods that can be used to demonstrate, quantitatively, the outcome of a given event or scenario. Analysis requirements may be beyond the scope of most engineering specifications currently used for design and rating, and special assessments may be required in certain conditions. Hence, the owner and the engineer must agree upon the level of analysis, loading, material properties, etc. that will be used for the basis of the analysis. Similarly, any laboratory or field testing must properly simulate or represent in-situ conditions (i.e., scale of the specimen or test, loading, failure mode, etc.) in order to be considered acceptable. Experience: As used herein, refers to the use of previous knowledge alone to qualitatively evaluate the outcome of a given event based on certain initial conditions. In order to use experi- ence, the user must be able to demonstrate at least the following: 1. The characteristics of the structure being evaluated are identical or sufficiently similar to the structure for which the RAP has previous documented experience. 2. The result of the damage mode is identical for the bridge(s) used as a reference. For example, strand fracture as a result of corrosion or impact may be effectively the same. In both cases, the strand failed. The information on which the decision is based must be included in the documentation of the RBI assessment. It may consist of the location, structure type, damage type, reason for selection, or other rationale and evidence used to form the decision so that a permanent record is available for future RAPs. B 2 General Descriptions of Consequence Scenarios This section provides guidance for the treatment of typical scenarios and situations for each of the four CF categories. A brief description of each CF category is provided, as well as typical examples or scenarios for each category. Methods for selecting the appropriate CF for a given

42 Proposed Guideline for Reliability-Based Bridge Inspection Practices failure scenario are described. This section is intended as guidance for evaluation. Specific situa- tions and scenarios may vary, and the RAP should utilize good engineering judgment supported with analysis or documented experience where necessary. Local rules, policies, and practices of the bridge owner should be considered in the assessments. As stated, when assessing the CF, the immediate and short-term outcomes, or the results of the failure of an element should be considered. The immediate consequence refers to the structural integrity and safety of the traveling public when the failure occurs. Considerations include whether a bridge will remain standing when the damage mode occurs, and whether the traveling public will remain safe. For example, failure of a load bearing member in a multi-girder redundant bridge is not expected to cause loss of structural integrity, excessive deflections, or collapse. As a result, the traveling public is not immediately affected when the failure occurs. Another scenario would be for a fracture-critical bridge, where the loss of a main member could cause excessive deflec- tion or collapse thereby causing the bridge to be immediately unsafe for the traveling public. The safety of the structure and the public should be considered for determining the immediate consequence. The short-term consequence refers to serviceability concerns and short-term impacts to the traveling public after a given damage mode occurs. Load posting, repairs, and speed reductions can be considered serviceability concerns. Lane, sidewalk, or shoulder closures as a result of the damage mode impact the traveling public and can cause delays. For example, a multi-girder redundant bridge that experiences the loss of a load bearing member is expected to remain standing; however, once the failure is discovered, a typical response is to close a lane or shoulder until the bridge is repaired. Therefore, the traveling public will be affected. The serviceability of the structure and the impact to the traveling public should be considered when determining the short-term consequence. For example, the failure of a member in a multi-girder bridge may be a moderate immediate consequence because the bridge is expected to remain standing and no excess deflections are expected to occur. However, if this bridge is located on an interstate located downtown in a major city, the short-term consequence of the member failure may be high or severe because a lane closure may be required, which would cause significant traffic delays. Therefore, the CF for this bridge may be high based upon the short-term consequence. Tables B2 through B6 provide additional guidance for commonly encountered situations for bridge decks, typical superstructures, and substructures. These tables provide descriptions of typical immediate and short-term effects from common damage modes and sample situations. The tables also include factors the RAP may consider in differentiating CF categories. For exam- ple, for the damage mode of spalling in a bridge deck, the CF may be different for a low ADT bridge than for a high ADT bridge, based on serviceability considerations. The CF may be differ- ent for a bridge that crosses a roadway than one that crosses a small stream, based on concerns regarding debris falling into traffic, etc. These tables are not intended to be comprehensive, but rather are intended to provide guidance and examples to assist an RAP with developing criteria for determining the CF for typical damage modes for common bridge designs under analysis. B 2.1 Low Consequence Event General Description • Minor effect on serviceability, no effect on safety. This scenario is the least serious of all the CF categories. The likelihood of structural collapse resulting from the damage mode is not credible and the effect on the serviceability of the bridge is minor.

Guideline for Evaluating the Consequence Factor 43 Requirements for Selection In order to select the lowest consequence category, the user must be able to clearly demonstrate that the consequence of the damage will be benign. Generally speaking, this decision will most often be based on engineering judgment and experience. Situations in which selection of this consequence scenario may be appropriate are as follows: • Failure of thin deck overlay. • Spalling in a concrete deck bridge on a low-volume and/or low-speed roadway. • Spalling/corrosion damage in an abutment where the bridge is over a non-navigable waterway or unused right-of-way land. B 2.2 Moderate Consequence Event General Description • Moderate effect on serviceability, minor effect on safety. This scenario can be characterized by consequences that are classified as moderate in terms of their outcome. The likelihood of collapse and loss of life is very remote, and there is a minor effect on the safety of the traveling public. Requirements for Selection In order to classify the consequence of a given failure scenario as moderate, the user must demonstrate that the damage mode will typically result in a serviceability issue. The damage mode Table B2. Consequence table for deck elements. Assumed damage mode is spalling. Consequence for Deck Descripon Sample Situaons Factors to Consider Low Immediate: Damage to the top of the deck does not present a safety concern for the traveling public. Falling debris from the boom of deck does not affect the safety of the public. Short term: Minimal serviceability concerns may require maintenance. Lile or no impact to traveling public. Bridge carrying low volume and/or low speed roadway Bridge with concrete deck over a non navigable waterway or unused right of way land ADT/ADTT Feature under Feature carried Stay in place forms Moderate Immediate: Damage to the top of the deck presents a minimal safety concern to the traveling public. Falling debris from the boom of deck presents a minimal safety concern. Short term: Moderate serviceability concerns. Speed reducon may be needed. Traffic is moderately impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Moderately traveled roadway where damage would cause minimal delays Bridge with stay in place forms over roadway where spalls would not reach roadway or waterway High Immediate: Damage to the top of the deck presents a moderate safety concern to the traveling public. Falling debris from the boom of deck presents a moderate safety concern. Short term: Major serviceability concerns. Repairs or speed reducon may be required. Traffic is greatly impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. High volume roadway where damage would cause reducon in posted speed or potenal for loss of vehicular control Bridge without stay in place forms over heavily traveled waterway or high volume roadway Severe Immediate: Damage to the top of the deck presents a major safety concern to the traveling public. Falling debris presents a major safety concern. Possible loss of life. Short term: Potenal for significant traffic delays on or under the bridge. Bridge over feature where spalling concrete would result in lane closure, loss of life, or major traffic delays

44 Proposed Guideline for Reliability-Based Bridge Inspection Practices poses no serious threat to the structural integrity of the bridge or to the safety of the public. Generally, damage that will require repairs that can be addressed in a programmed fashion (i.e., non-emergency) would be classified as having a moderate consequence. Member or structural redundancy should be a consideration, and, in cases where the member is non-redundant, it may be prudent to classify an event higher in consequence. Situations in which the selection of this CF may be appropriate are as follows: • Spalling damage in a deck soffit or concrete girder for a bridge over multi-use path, railroad, or low-volume (<10 ADT) roadway. • Spalling in a concrete deck bridge on a moderate-volume roadway. • Lane or shoulder closure on a bridge carrying a moderate-volume urban roadway or a high- volume rural roadway that would cause moderate delays for drivers. • Fatigue cracks that require repair but are not the result of primary member stresses, such as out-of-plane distortion cracks in redundant members The examples above illustrate some of the element failure scenarios that would typically be categorized as having moderate consequence. In some cases, failure scenarios that could be considered more serious can be categorized as having moderate consequences, if analysis or past experience can be used to better define the outcome of a given scenario. For example, out-of-plane fatigue cracks are not uncommon in some older steel bridges, and are included in Table B3. Consequence table for steel superstructure elements. Assumed damage mode is loss of one primary load carrying member. Consequence for Steel Superstructure Descripon Sample Situaons Factors to Consider Low Immediate: Lile to no impact on structural capacity is expected based upon structural analysis or documented experience. Public safety is unaffected. Short term: Minimal serviceability concerns may require maintenance. Lile or no impact to traveling public. Bridge over non navigable waterway or unused right of way land Rural bridge with low ADT/ADTT ADT/ADTT Feature under Feature carried Redundancy Composite construc­on Load carrying capacity/ra­ng Moderate Immediate: Structural capacity is expected to remain adequate based upon structural analysis or documented experience. Short term: Moderate serviceability concerns. Speed reduc­on or load pos­ng may be needed. Traffic is moderately impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Bridge over mul­ use path, railroad, or lightly traveled waterway Bridge on or over moderate volume urban roadway or high volume rural roadway that would cause moderate delays for drivers High Immediate: Structural capacity is expected to remain adequate. Short term: Major serviceability concerns. Load pos­ng, repairs, or speed reduc­on may be needed. Traffic is greatly impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Bridge with alternate load path(s) that has an expecta­on of adequate remaining structural capacity Lane or shoulder closure on or under roadway that would cause major delays for drivers Severe Immediate: Structural collapse. Possible loss of life. Short term: Poten­al for significant traffic delays on or under bridge. Bridge with high ADT/ADTT that requires closure

Guideline for Evaluating the Consequence Factor 45 the examples above. However, other types of fatigue cracks may be more serious. For example, consider cracking in a single plate of a built-up riveted girder. These types of cracks would nor- mally be expected to be much more serious. They may require categorization as having high or a severe consequence, if it is assumed that the crack propagates such that the load carrying capacity of the girder is lost. However, in many cases, riveted built-up members are composed of two or three cover plates, two angles, and the girder web. If it could be shown by analysis that even after complete cracking of one of these individual components (e.g., complete cracking of one of the cover plates) the member still has plenty of reserve capacity, then it might be reasonable to classify the event as a moderate consequence scenario. The individual making this assessment would also want to consider overall system redundancy and other factors. Hence, if analysis can be used to show that a condition that is generally perceived to be more serious, but is actually not so, then it may be justified to classify the event as having a moderate con- sequence. Experience may also be utilized to assess if a given failure scenario is a high consequence event or a moderate consequence event. In cases where a given owner may have had the same or very similar experience with several other identical or sufficiently similar bridges, the owner may Consequence for Concrete Superstructure Descripon Sample Situaons Factors to Consider Low Immediate: Lile to no impact on structural capacity is expected based upon structural analysis or documented experience. Falling debris does not affect the safety of the public. Short term: Minimal serviceability concerns may require maintenance. Lile or no impact to traveling public. Bridge over non navigable waterway or unused right of way land Rural bridge with low ADT/ADTT ADT/ADTT Feature under Feature carried Redundancy Composite construcon Load carrying capacity/rang Moderate Immediate: Structural capacity is expected to remain adequate based upon structural analysis or documented experience. Falling debris presents a minimal safety concern to the public. Short term: Moderate serviceability concerns. Speed reducon or load posng may be needed. Traffic is moderately impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Bridge over mul use path, railroad, or lightly traveled waterway Bridge on or over moderate volume urban roadway or high volume rural roadway that would cause moderate delays for drivers High Immediate: Structural capacity is expected to remain adequate. Falling debris presents a moderate safety concern to the public. Short term: Major serviceability concerns. Load posng, repairs, or speed reducon may be needed. Traffic is greatly impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Bridge with alternate load path(s) that has an expectaon of adequate remaining structural capacity Lane or shoulder closure on or under roadway that would cause major delays for drivers Severe Immediate: Structural collapse. Falling debris presents a major safety concern to the public. Possible loss of life. Bridge over feature where spalling concrete would result in lane closure, loss of life, or significant traffic delays Short term: Potenal for significant traffic delays on or under bridge. Table B4. Consequence table for reinforced concrete superstructure elements. Assumed damage mode is loss of one primary load carrying member.

46 Proposed Guideline for Reliability-Based Bridge Inspection Practices be able demonstrate that a lower CF is justifiable. Very high load ratings (e.g., 150% of the mini- mum required) and redundancy could also be factors to consider when selecting this CF category. Of course, if experience and judgment are used to determine CF, then sufficient documentation would need to be available to justify the selection of a given CF. B 2.3 High Consequence Event General Description • Major effect on serviceability, moderate effect on safety. This scenario can be characterized by consequences that are more serious in terms of their outcome. The likelihood of collapse and loss of life may be more measurable, but still relatively remote. Requirements for Selection The user must be able to demonstrate that the possibility of collapse and loss of life are still relatively remote when identifying a given failure scenario as having a high consequence. Though the bridge may require repairs, the outcome would not be catastrophic in nature. Consequence for Prestressed Superstructure Descripon Sample Situaons Factors to Consider Low Immediate: Lile to no impact on structural capacity is expected based upon structural analysis or documented experience. Falling debris does not affect the safety of the public. Short term: Minimal serviceability concerns may require maintenance. Lile or no impact to traveling public. Bridge over non navigable waterway or unused right of way land Rural bridge with low ADT/ADTT ADT/ADTT Feature under Feature carried Redundancy Composite construcon Load carrying capacity/rang Moderate Immediate: Structural capacity is expected to remain adequate based upon structural analysis or documented experience. Falling debris presents a minimal safety concern to the public. Short term: Moderate serviceability concerns. Speed reducon or load posng may be needed. Traffic is moderately impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Bridge over mul use path, railroad, or lightly traveled waterway Bridge on or over moderate volume urban roadway or high volume rural roadway that would cause moderate delays for drivers High Immediate: Structural capacity is expected to remain adequate. Falling debris presents a moderate safety concern to the public. Short term: Major serviceability concerns. Load posng, repairs, or speed reducon may be needed. Traffic is greatly impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Bridge with alternate load path(s) that has an expectaon of adequate remaining structural capacity Lane or shoulder closure on or under roadway that would cause major delays for drivers Severe Immediate: Structural collapse. Falling debris presents a Bridge over feature wheremajor safety concern to the public. Possible loss of life. Short term: Potenal for significant traffic delays on or under bridge. spalling concrete may result in lane closure, loss of life, or significant traffic delays Table B5. Consequence table for prestressed concrete superstructure elements. Assumed damage mode is loss of one primary load carrying member.

Guideline for Evaluating the Consequence Factor 47 Examples of high consequence events would include scenarios that require short-term closures for repairs, lane restrictions that have a major impact on traffic, load postings, or other actions that majorly affect the public. Situations where the selection of this CF may be appropriate are as follows: • Failure of a main member in a multi-girder bridge with sufficient load path redundancy. • Spalling damage in a deck soffit or concrete girder for a bridge over a navigable waterway or a moderate-/high-volume roadway. • Spalling in a concrete deck bridge on a high-volume roadway. • Lane or shoulder closure on or under a roadway that would cause major delays for drivers. • Impact damage on a multi-girder bridge. Again, using brittle fracture of a girder as an example, consider the response to the fracture of an exterior girder in a multi-girder bridge. If the girders are spaced relatively closely, a reasonable strategy would be to place barriers on the bridge to keep traffic off the shoulder and hence, off the faulted girder. Though one girder out of several was compromised, experience indicates the remaining girders have sufficient capacity to carry traffic safely. In the above example, it is important to note the reaction to the fracture was not based on calculations, but was based entirely upon experience. If the owner performed calculations that Table B6. Consequence table for substructure elements. Assumed damage mode is spalling. Consequence for Substructure Descripon Sample Situaons Factors to Consider Low Immediate: Falling debris does not affect the safety of the public. Structural capacity of the bridge remains adequate. Short term: Minimal serviceability concerns may require maintenance. Lile or no impact to traveling public. Bridge over non navigable waterway or unused right of way land ADT/ADTT Feature under Load carrying capacity Moderate Immediate: Falling debris from substructure presents a minimal safety concern to the public. Structural capacity is expected to remain adequate based upon structural analysis or documented experience. Short term: Moderate serviceability concerns. Speed reducon or load posng may be needed. Traffic is moderately impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Bridge over mul use path, railroad, or lightly traveled waterway High Immediate: Falling debris from substructure presents a moderate safety concern to the public. Structural capacity is expected to remain adequate. Short term: Major serviceability concerns. Load posng, repairs, or speed reducon may be needed. Traffic is greatly impacted as a result of lane, shoulder, or sidewalk closure on or under bridge. Lane or shoulder closure on roadway that would cause major delays for drivers Severe Immediate: Structural collapse, bearing area failure, or loss of load carrying capacity. Falling debris presents a major safety concern to the public. Possible loss of life. Short term: Potenal for significant traffic delays on or Bridge adjacent to high volume roadway with insufficient horizontal clearance where spalling concrete may result in lane closure, loss of life, or major under bridge. traffic delays Bearing area failure resulng in deck misalignment

48 Proposed Guideline for Reliability-Based Bridge Inspection Practices quantifiably showed that the bridge had sufficient reserve capacity in the faulted condition, i.e., with one girder fractured, it might be reasonable to identify the event as having a moderate consequence. Guidance on such analysis exists in the literature and it can be performed for common bridges and common bridge types. However, simplified analytical procedures may also suffice. For exam- ple, there is considerable discussion regarding redundancy of multi-girder systems, both concrete and steel, as reported in NCHRP Report 406: Redundancy in Highway Bridge Superstructures (1). This document provides direction on determining the capacity and the redundancy as a func- tion of span, girder spacing, and the number of loaded lanes using system factors. The research resulted in the development of system factors that quantify redundancy based on an assessment of the reliability of the bridge systems, rather than simply the individual bridge members. Using the recommended system factors may greatly reduce the analytical effort needed in assessing a bridge. The major conclusion from this research was that bridges designed to AASHTO bridge specifications generally possess sufficient reserve capacity. In addition, NCHRP Project 12-87, “Fracture-Critical System Analysis for Steel Bridges” was underway at the time this report was prepared and once complete may be of use in performing system analysis. If experience is used as the reason to justify a reduction from a high consequence to a moderate consequence, the experience referenced would have to be for a type of structure and a damage mode outcome that is nearly identical to the one under consideration, as described in section B 1.1. (For example, corrosion, fatigue, or fracture can all lead to a failed girder. Hence, although the dam- age modes are different, the outcome is the same.) Therefore, the RAP would have to adequately document and demonstrate that the cited case(s) are of sufficient similarity. Owners may cite examples both in their own state and from other states. Another desirable characteristic would be whether or not the experience with a given response has been observed more than once. For example, an owner may have experience with a certain type of rolled steel beam bridge and truck impact. Experience with truck impacts on several similar steel bridges may demonstrate that for the bridge under consideration, impact to the superstructure would not result in a set of circumstances that justify identifying the event as having a high consequence. Based on this experience, it may be appropriate to identify the event as having only moderate consequences. Another example would be a case in which there is severe spalling at the bearing of a member in a prestressed, multi-girder bridge that is over a small creek or a flood plain. Hence, there is no concern regarding spalled concrete hitting someone or something below the bridge (minor effect on public safety). If calculations could be made to show that if the bearing were to com- pletely fail, there would only be moderate effects on serviceability, then it would be reasonable to state this is a moderate consequence event. In the absence of detailed calculations and/or substantial experience regarding the specific scenario, it would be required to be identified as having a high consequence, based on the criteria discussed. B 2.4 Severe Consequence Event General Description • Major effect on serviceability and safety. This is the most critical CF category and can be characterized by events that, should they occur, are anticipated to result in catastrophic outcomes. Structural collapse and loss of life are likely should the failure occur. Requirements for Selection Due to the catastrophic nature implied by this consequence scenario, it should not be selected arbitrarily as a catch-all or just “to be conservative.” The user must have reasonable justification

Guideline for Evaluating the Consequence Factor 49 that shows that the failure scenario being considered is likely to be consistent with a severe con- sequence event. Examples of severe consequence events would include failure of the pin or hanger in a bridge with a suspended truss span or a two-girder system, or strand fractures in a pre- or post- tensioned element that results in a non-composite member falling into a roadway below, such as what was observed in Washington Township, PA (2). Failure of a pier due to severe corrosion of the reinforcement or to a lack of reinforcement would also be an example of a severe con- sequence event. Situations in which the selection of this CF may be appropriate are as follows: • Fracture in a non-redundant steel bridge member. • Failure of a non-composite girder over traffic. • Spalling of a concrete soffit, concrete girder, or concrete abutment over a high-volume road- way or pedestrian walkway. • Lane or shoulder closure on a major roadway that would cause significant delays for the traveling public. • Bearing area failure resulting in deck misalignment. Cases for which there is insufficient experience or where reliable calculations cannot be made (due to lack of analytical models or data for use in the models) may also be categorized as severe. Examples would be unique, one-of-a-kind bridges or other structural systems for which the result of failure associated with a given damage mode is essentially unknown. In such cases, the only reasonable approach is to conservatively assume and select the worst-case consequence (i.e., a severe consequence), as the actual outcome cannot be well defined. A common example of a failure that would result in a severe consequence is primary member failure in a fracture-critical bridge. Due to the perceived lack of redundancy, fracture of a main member is assumed to result in a total collapse of a bridge or a portion of a bridge. Though this is a reasonable conclusion in the absence of more rigorous analysis, the bridges can also be good exam- ples of where more rigorous analysis can be used to show redundancy actually exists. For exam- ple, a literature review conducted as part of NCHRP Synthesis 354: Inspection and Management of Bridges with Fracture-Critical Details, revealed that there were no documented cases of catastrophic failure for any two-girder bridges or cross girders where fractures had occurred (3). In some of the failures, an entire girder fractured, but due to inherent redundancy of the unaccounted-for load paths, such as the deck and lateral system, and overall system behavior, the bridges did not collapse. In fact, in some cases, there is little perceived deflection in the faulted state. In light of the above, owners may wish to perform an after-fracture redundancy analysis to demonstrate that a given bridge possesses sufficient alternate load paths such that the most likely outcome would have only high consequences. Obviously, the owner must select the appropriate live load that must be carried in the faulted state for the analysis. Further, consideration should be given to the fact that the bridge may need to remain in service for some time with the fracture undetected. For example, if the fracture occurred immediately after a scheduled inspection and there was little or no evidence that would alert anyone to the condition and to take action (e.g., no deflection). Obviously, there are other damage modes that may result in a severe consequence. For those, analysis may also be used to demonstrate that the most likely outcome would have only a high consequence. Downgrading to the less serious consequence scenario is permitted but only through the use of analysis. Experience alone may not be used to justify downgrading from a severe consequence to a high consequence, due to the catastrophic outcomes associated with the more severe scenario. While experience may be used in conjunction with analytical studies to make a stronger case for downgrading to a lower consequence scenario, experience alone is not deemed to be sufficient.

50 Proposed Guideline for Reliability-Based Bridge Inspection Practices B 3 Use of Expert Elicitation for Determining the Consequence Scenario An expert elicitation of the RAP can be a useful tool for evaluating the appropriate CF for situ- ations that are not well matched to the examples given above, or to establish basic ground rules for the assessment of common situations. The expert elicitation process can be used to establish or to build consensus among the RAP and to assist in identifying the most likely outcomes of damage modes assessed during the reliability analysis. The process should be carefully controlled and systematic to ensure that the judgments of the RAP are effectively ascertained. The process involves a few basic, but critical, steps as follows: 1. Statement of the Problem: The RAP should be presented with a clear statement of the problem and supporting information to allow for expert judgment to be made. Care should be taken to ensure the problem statement does not contain information that could lead to a biased decision. The problem statement typically includes data regarding the bridge design, location, typical traffic patterns, and the failure scenario under consideration. 2. Expert Elicitation: Independently, each member of the RAP is asked, based on his or her judg- ment, experience, available data, and given the scenario presented, to determine what the most realistic consequence is resulting from the damage mode under consideration. The expert is asked to express this as a percentage, with the smallest unit of estimate typically being 10%. The expert provides a written statement on what factors they considered in making the estimate. 3. Comparison of Results: Once each member of the RAP has rated the situation, the results of the elicitation are aggregated. Generally, there will be consensus regarding the most critical consequence. However, in some cases, the most critical choice will not be clear and there will not be consensus. 4. Identify Consequence Factor: If there is consensus among the panel regarding the appropriate CF, then the rationale for making the determination is recorded. This rationale should be consistent with the general guidance herein, or document deviations, changes, and associated rationale. For cases in which consensus is not reached in the initial elicitation, the experts should discuss their rankings, their assumptions, and rational for their specific judgments. The members of the RAP should then be given the opportunity to discuss the various judgments and to revise their scores based on the discussion. In some cases, additional information may be needed to support developing a consensus regarding the appropriate CF. For example, analysis may need to be con- ducted or previous experience documented. If consensus cannot be reached, a potential approach would be to adopt the most conservative consequence scenario that was included among the revised scores. Exceptions to the selected CF should also be documented. When consensus cannot be reached, the RAP may determine that additional analysis is required to determine the appropriate consequence for a given failure scenario. In some cases, additional data collection may be required in order to reach a consensus. Regardless of the approach, the individual RAP should have the flexibility to develop its own methodologies to handle cases for which there is no consensus. However, at the conclusion of the analysis, the method still must result in the selection of the most appropriate consequence scenario, based on the guidelines provided herein and on good engineering judgment. B 4 References 1. Ghosn, M., Moses, F., NCHRP Report 406: Redundancy in Highway Bridge Superstructures. 1998, TRB, National Research Council: Washington, D.C. 2. Clay, N., et al., Forensic Examination of a Noncomposite Adjacent Precast Prestressed Concrete Box Beam Bridge. Journal of Bridge Engineering. 15(4): p. 408–418. 3. Connor, R. J., R. Dexter, and H. Mahmoud, NCHRP Synthesis 354: Inspection and Management of Bridges with Fracture-Critical Details. 2005, Transportation Research Board of the National Academies: Washington, D.C.

51 A P P E N D I X C 52 C 1 Inspection Intervals 52 C 1.1 Important or Essential Bridges Guideline for Determining the Inspection Interval

52 Proposed Guideline for Reliability-Based Bridge Inspection Practices C 1 Inspection Intervals Inspection intervals are determined based on the reliability analysis using a simple four by four matrix as shown in Figure C1, which illustrates a risk matrix for a typical highway bridge. Engi- neering judgment is required for establishing the specific divisions applied to the risk matrix; the divisions are generally applied to ensure that the likelihood of damage remains low during the interval between inspections, such that there are multiple inspections conducted before there is a high likelihood of failure occurring. When consequences are relatively high, should the failure occur, the interval is further reduced to provide an extra margin of safety. For the risk matrix shown in Figure C1, divisions have been made to separate the bridges requiring more frequent inspections (Category I) from those requiring less frequent inspec- tions (e.g., Categories III, IV, and V). The inspection interval categories are shown in Table C1. Bridges with elements falling in Category II require the typical inspection interval of 24 months, currently used under the NBIS. The inspection intervals and the divisions on the risk matrix are engineering-based to ensure a high margin of safety and that multiple periodic inspections take place before the likelihood of failure becomes high. In other words, the intervals are determined such that the likelihood of failure remains low, and the intervals are further reduced as consequences increase to provide additional levels of safety. For example, recall that the RAP assessment of the likelihood of a damage mode resulting in a “failure” (as defined in Section 2.1) is based on a 72-month time- frame. For a given element, if there is low likelihood of a failure (OF = 2), and the consequence of that failure is moderate (CF = 2), the inspection interval of 72 months (Class IV) is identified on the matrix. This is justified because the analysis has indicated that there is a low likelihood of failure, and even if the failure occurs, there will be only a moderate effect on the serviceability of the bridge. However, if the consequence of the failure were high, then the inspection inter- val is reduced to 48 months (Class III) and 24 months (Class II) if the consequence is severe. Alternatively, if the likelihood of failure is moderate (OF = 3) over 72 months, the maximum inspection interval is less than 72 months, regardless of the consequence; 48 months if the con- sequence were only low (benign) (CF = 1) or moderate (CF = 2) and 24 months if the conse- quence were high (CF = 3). Similarly, if the likelihood of failure were remote over the 72-month timeframe, it may be justified to have a maximum interval of more the 72 months, particularly if the consequences are assessed to be benign (CF = 1). As the consequences increase, this interval is reduced. C 1.1 Important or Essential Bridges As noted, the divisions on the risk matrix require engineering judgment to determine which inspection intervals are acceptable and necessary. For certain bridges, for example, essential bridges along key transportation routes, an owner may wish to provide an additional margin of reliability. Under these circumstances, the divisions on the risk matrix may be adjusted down Figure C1. Risk matrix for a typical highway bridge. Category Maximum Interval I 12 months or less II 24 months III 48 months IV 72 months V 96 months Table C1. Maximum inspection interval categories.

Guideline for Determining the Inspection Interval 53 and toward the lower left corner of the matrix. For example, Figure C2 illustrates a risk matrix an owner could apply to bridges for which an additional measure of reliability is desired. This may be due to the importance of the bridge to the effectiveness of the transportation system overall, and/or because the bridge serves essential purposes. Criteria for identifying these essential or important bridges should be developed by the bridge owner, but would typically consider such factors as ADT, functional classification of the route, and importance to local transportation functions. Owners may already have criteria for identifying essential or important bridges for which added measures of reliability are desired. Figure C2. Risk matrix that may be applied to “essential” bridges.

54 A P P E N D I X D 55 D 1 Introduction 55 D 1.1 NDE Method Technical Readiness Levels and Costs 56 D 2 Inspection Methods and Technologies Inspection Technologies

Inspection Technologies 55 D 1 Introduction This appendix provides general guidance for the inspection methods to be utilized in a risk-based inspection (RBI) practice. The section includes a description of nondestructive evaluation (NDE) technology’s technical readiness and relative costs to assist decision mak- ers in determining appropriate and practical technologies for the detection and evaluation of typical damage modes and deterioration mechanisms in highway bridges. This section also includes tables that indicate the relative reliability of different inspection methods and NDE technologies to assist decision makers regarding the application and effectiveness of the technologies. D 1.1 NDE Method Technical Readiness Levels and Costs This section provide general guidance on the technical readiness levels (TRLs) and costs of the most common NDE technologies that may be applied for damage detection in highway bridges. Technologies have been evaluated on relative scales using expert judgment and experience. Table D1 indicates the scale used to assess the TRL of the technologies. This scale is intended to assist engineers in understanding the practicality and the availability of NDE technologies, and to discriminate between those techniques that are readily available and well proven, from those that may be more experimental in nature. The scales provide a five-level discrimination that indicates if an NDE technology is experimental in nature, or if it is a widely available and widely implemented technology. NDE technologies are rated according to the cost scales shown in Table D2. These scales are intended to provide engineers with general information regarding the relative costs of imple- menting NDE technologies for bridge inspection. Relative costs are based on a typical, multi- girder highway bridge approximately 150 ft in length. The TRL and costs for NDE technologies are shown in Table D3. Table D1. TRLs for NDE technologies. Technical Readiness Level Description TRLNo. Examples Fundamental Research: basic research in the laboratory 1 Fundamental sensor research, nano-sensors, laser-induced breakdown spectroscopy (LIBS) In Development: laboratory equipment, starting field testing and experimental applications, proof of concept testing 2 In-situ corrosion sensors, positron annihilation, backscatter x-ray, thermal crack detection Application Development: Applications for the technology are being developed, commercially available research equipment, field testing is experimental/developmental, initial assessments of effectiveness in the field, reliability unknown 3 Electromagnetic-acoustic transducer (EMAT) sensors, ultrasonic stress measurement, magnetic flux leakage for embedded strands thermal crack detection Controlled Implementation: Commercially available equipment and service, application by specialist/consultant, (certification may be available), assessments of reliability/effectiveness are ongoing 4 Ground penetrating radar (GPR), radiography, impulse response, phased array ultrasonics, infared thermography (IR) Widespread Implementation: Certification available, widely used, commercially available equipment, commonly available, application by suitably trained technician, generally accepted reliability/effectiveness 5 Ultrasonic pulse velocity (UPV), dye penetrant, eddy current, magnetic particle, covermeters, half cell

56 Proposed Guideline for Reliability-Based Bridge Inspection Practices D 2 Inspection Methods and Technologies The tables included in this section (Tables D4 through D9) qualitatively describe the reliability and effectiveness of NDE technologies and inspection methods including routine inspection and hands-on inspections. In making the assessments of reliability and effectiveness, it was assumed that a routine inspection was conducted without hands-on access to the bridge element. The reliability assessment indicated in Tables D5 through D9 is intended to provide general guid- ance on effective inspection methods for detecting and evaluating certain damage modes and deterioration mechanisms. Key monitoring or sampling methods that provide tools for assessing the likelihood of corrosion damage developing in concrete have been included. The key to the symbolic guide is shown in Table D4. Methods that are low reliability typical do not provide effective detection or assessment, and are not recommended for the damage mode or deteriora- tion mechanism indicated. Table D2. Relative cost scales for NDE technologies. Cost Scales Description Symbol Examples Low cost, state forces, or $100s of dollars to apply/bridge Dye penetrant, magnetic particle, impact echo, ultrasonic thickness, thermography Moderate Cost, $1,000–$10,000 typical costs/bridge GPR, ultrasonic crack detection, impact echo High cost, >$10,000 to apply Health monitoring, x-ray diffraction, radiography Table D3. TRL and cost for typical NDE technologies. Code Name TRL Cost Material Primary Usage MP Magnetic particle testing 5 Steel Surface-breaking cracks in steel PT Dye penetrant testing 5 Steel Surface-breaking cracks in steel UT Ultrasonic testing 5 Steel Surface and subsurface cracks in steel, volumetric defects UT-T Ultrasonic thickness gage 5 Steel Plate thickness, section loss ET Eddy current testing 5 Steel Surface-breaking cracks in steel AE Acoustic emission 4 Steel Monitoring growth of fatigue cracks IR Infrared thermography 4 Concrete Subsurface delaminations in concrete GPR Ground penetrating radar 4 Concrete Detecting damage in concrete associated with corrosion, rebar depth, locating embedded metal objects UPV Ultrasonic pulse velocity 5 Concrete Deterioration of concrete, concrete moduli/strength, subsurface voids, cracks IE Impact echo 4/5 Concrete Delaminations in concrete, deterioration of concrete, subsurface voids CD Chain drag 5 Concrete Delaminations in concrete HC Half-cell potential 5 Concrete Corrosion potential RT Radiographic testing 4 Concrete Internal voids, loss of section/fracture in embedded steel S Sounding 5 Concrete Delaminations, deterioration of concrete SAW Surface acoustic wave 4 Concrete Cracking and deterioration in concrete, delaminations MFL Magnetic flux leakage 3 Concrete Loss of section for embedded steel element (prestressing strand, rebar)

Inspection Technologies 57 Table D4. Symbolic guide to inspection method reliability and effectiveness. Key Low Moderate - low Moderate - high High Table D5. Inspection methods for bare concrete decks. Damage Mode or Mechanism Routine Visual Hands-On Visual Sounding 1 IR GPR Impact Echo Chain Drag Half Cell Chloride Ion Content NA NA Delamination (dry) NA NA Deck cracking (distributed) NA NA Corrosion damage NA NA Freeze-thaw/ pulverized/ cracks NA NA Delamination in soffit2 NA NA NA ASR NA NA Active corrosion/ corrosion potential 1 Based on FHWA visual inspection study results. 2 NDE technologies applied to the soffit surface. Spalling/patches Table D6. Inspection methods for concrete decks with overlays. Damage Mode or Mechanism Routine Visual Hands-On Visual Sounding IR GPR Impact Echo Chain Drag Half Cell Chloride Ion Content Spalling/patches NA NA Delamination NA NA Debonding/ overlay delamination NA NA Corrosion damage NA NA Freeze-thaw/ pulverized/ cracks NA NA Delamination in soffit1 NA NA NA ASR NA NA corrosion potential Active corrosion/ 1 NDE technologies applied to the soffit surface.

58 Proposed Guideline for Reliability-Based Bridge Inspection Practices Damage Mode or Mechanism Routine Visual Hands-On Visual Sounding IR GPR Impact Echo Chain Drag SAW Delamination/ debonding Overlay cracking Spalling/patches Table D7. Concrete deck overlays. Damage Mode or Mechanism Routine Visual Hands-on Visual1 PT 2 MT2 UT2 UT-T ET2 Fatigue cracks – primary stress NA Out of plane distortion cracking Section loss NA NA NA Coatings failure NA NA NA NA Steel pins – pack rust NA NA NA Cracks in steel pins NA NA NA 1 Assumes inspectors have been adequately trained. 2 Assumes Level II certification; Level III procedure development. Table D8. Inspection methods for steel members. Damage Mode or Mechanism Routine Visual Hands-on Visual Sounding IR GPR IE MFL RT UPV Chloride Ion Content NA Delamination NA Strand corrosion NA Freeze-thaw/ pulverized/ cracks NA Delamination in soffit1 NA ASR NA Active corrosion/ corrosion potential Spalling/patches 1 NDE technologies applied to the soffit surface. Table D9. Inspection methods for open prestressed girders.

59 61 Introduction 61 Scoring Scheme 62 Screening Attributes 62 S.1 Current Condition Rating 62 S.2 Fire Damage 63 S.3 Susceptible to Collision 63 S.4 Flexural Cracking 64 S.5 Shear Cracking 65 S.6 Longitudinal Cracking in Prestressed Elements 65 S.7 Active Fatigue Cracks Due to Primary Stress Ranges 65 S.8 Details Susceptible to Constraint-Induced Fracture (CIF) 66 S.9 Significant Level of Active Corrosion or Section Loss 66 S.10 Design Features 67 Design Attributes 67 D.1 Joint Type 67 D.2 Load Posting 67 D.3 Minimum Vertical Clearance 69 D.4 Poor Deck Drainage and Ponding 69 D.5 Use of Open Decking 69 D.6 Year of Construction 71 D.7 Application of Protective Systems 72 D.8 Concrete Mix Design 73 D.9 Deck Form Type 73 D.10 Deck Overlays 73 D.11 Minimum Concrete Cover 75 D.12 Reinforcement Type 76 D.13 Built-Up Member 76 D.14 Constructed of High Performance Steel 76 D.15 Constructed of Weathering Steel 77 D.16 Element Connection Type 78 D.17 Worst Fatigue Detail Category 78 D.18 Skew 78 D.19 Presence of Cold Joints 79 D.20 Construction Techniques and Specifications 79 D.21 Footing Type 79 D.22 Subsurface Soil Condition A P P E N D I X E Attribute Index and Commentary

60 Proposed Guideline for Reliability-Based Bridge Inspection Practices 80 Loading Attributes 80 L.1 ADTT 81 L.2 Dynamic Loading from Riding Surface 81 L.3 Exposure Environment 82 L.4 Likelihood of Overload 82 L.5 Rate of De-icing Chemical Application 83 L.6 Subjected to Overspray 83 L.7 Remaining Fatigue Life 84 L.8 Overtopping/High Water 84 Condition Attributes 84 C.1 Current Condition Rating 85 C.2 Current Element Condition State 86 C.3 Evidence of Rotation or Settlement 86 C.4 Joint Condition 86 C.5 Maintenance Cycle 87 C.6 Previously Impacted 87 C.7 Quality of Deck Drainage System 88 C.8 Corrosion-Induced Cracking 88 C.9 General Cracking 89 C.10 Delaminations 89 C.11 Presence of Repaired Areas 90 C.12 Presence of Spalling 90 C.13 Efflorescence/Staining 91 C.14 Flexural Cracking 92 C.15 Shear Cracking 92 C.16 Longitudinal Cracking in Prestressed Elements 93 C.17 Coating Condition 93 C.18 Condition of Fatigue Cracks 94 C.19 Presence of Fatigue Cracks due to Secondary or Out of Plane Stress 94 C.20 Non-Fatigue-Related Cracks or Defects 94 C.21 Presence of Active Corrosion 95 C.22 Presence of Debris 95 References

Attribute Index and Commentary 61 Introduction This section includes suggested attributes for the reliability assessment of bridges. Users can select attributes from this listing. It is also recommended that users develop additional attributes that meet the needs of their individual agencies. This commentary is organized into four sec- tions: Screening, Design, Loading, and Condition. The Screening section describes attributes that may be used to quickly identify bridges that should not be included in a particular analysis, either because they already have significant damage or they have attributes that are outside the scope of the analysis being developed. In many cases, these attributes may require engineering analysis beyond that which is typically conducted during a reliability assessment using this Guideline. Screening attributes are typically attributes that: • Make the likelihood of failure very high. • Make the likelihood of failure unusually uncertain. • Identify a bridge with different anticipated deterioration patterns than other bridges in a group. Design attributes are characteristics of a bridge element that are part of the element’s design. Design attributes are frequently intrinsic characteristics of the element that do not change over time, such as the amount of concrete cover or material of construction [concrete, high perfor- mance concrete (HPC), etc.]. In some cases, preservation or maintenance activities that con- tribute to the durability of the bridge element may be a design attribute, such as the use of penetrating sealers as a preservation strategy. Loading attributes are characteristics that describe the loads applied to the bridge element. These may include structural loading, traffic loading, or environmental loading. Environmental loading may be described in macro terms, such as the general environment in which the bridge is located, or on a local basis, such as the rate of de-icing chemical application on a bridge deck. Loading attributes describe key loading characteristics that contribute to the damage modes and deterioration mechanisms under consideration. Condition attributes describe the relevant bridge element conditions that are indicative of its future reliability. These can include its current element or component level rating, or may be a specific condition that will affect the durability of the element. For example, if the deterioration mechanism under consideration is corrosion at the bearing areas, the condition of the bridge joint may be a key attribute in determining the likelihood that corrosion will occur in the bearing area. The listing of attributes included here is not intended to be comprehensive or mandatory. Users should consider adding attributes that are important to their specific inventory. Users are encouraged to document the rationale for including additional attributes in the reliability assess- ment, along with an appropriate scoring scheme. Users may also wish to omit certain attributes if they are not relevant to their inventory or do not contribute to the reliability and durability of bridges within their inventory. The suggested weightings are also exemplary in nature and may need to be adjusted to meet the needs of a particular bridge inventory. Scoring Scheme Attributes are assigned points based on the importance or contribution of the attribute in terms of the durability and the reliability of the element being assessed. In general, the scoring scheme utilizes a three-stage assessment of the importance of the attribute as shown in Table E1. The Ranking Descriptor is intended to provide some verbal description of the weight associated with each score. As shown, three relative course levels are presented: Low, Moderate, and High. The RAP may wish to modify the suggested scoring for a given attribute, based on local condi- tions, past experience, and previous performance within its bridge inventory and operational

62 Proposed Guideline for Reliability-Based Bridge Inspection Practices environment. The scoring scheme should effectively develop sound engineering rationale to support risk-based inspection practices. Screening Attributes S.1 Current Condition Rating Reason(s) for Attribute. The current condition rating characterizes the overall condi- tion of the component being rated according to the NBIS rating scale. Bridge components that have condition ratings of 4 or less have been rated to be in poor condition. In some cases, these components may already be on a reduced (12 month or less) inspection frequency. Users may wish to use this criterion to screen bridges that are already in poor condition and, as a result, require more in-depth analysis to identify their inspection needs. Users could also assign the OF of “high” without further assessment, since the component is already in poor condition. For element-level inspection approaches, National Bridge Elements (NBEs) or Bridge Man- agement Elements (BMEs) could be utilized within the screening criteria, as appropriate for specific bridge inventories and inspection practices. Generally, elements indicated with condi- tion states of 4 would be appropriate for consideration as a screening tool for elements selected to match the needs and practices within the specific bridge inventory. Assessment Procedure. This screening attribute is scored based on whether the current condition rating is 4 or less or greater than 4. The current condition rating from the most recent inspection report should be used. If using an element-level approach, the RAP should identify appropriate elements and condition states for screening. Current condition rating is less than or equal to 4 Component is screened from general reliability assessment Current condition rating is greater than 4 Continue with procedure S.2 Fire Damage Reason(s) for Attribute. Incidences of fire on or below a highway bridge are not uncom- mon. This type of damage is most frequently caused by vehicular accidents that result in fire, but secondary causes such as vandalism, terrorism, or other damage initiators should not be discounted. If fire does occur on or below a bridge, an appropriate follow-up assessment should be conducted to determine how the fire has affected the load carrying capacity and the durabil- ity characteristics of the main structural members and the deck. This assessment is typically performed during a damage inspection immediately following the incident. Damage to bridge components resulting from a fire is either immediately apparent during the damage inspection, or may manifest within the first 12- to 24-month interval following the Table E1. Suggested rank scoring for attributes. Ranking Descriptor Total Points High 20 Moderate 15 Low 10

Attribute Index and Commentary 63 fire. Based on this observation, bridges that have experienced a fire may be screened from the reliability assessment until an inspection, which has been conducted approximately 12 months or more after the fire, confirms that the fire has not affected the typical durability characteristics of the bridge components. The purpose of this screening is to ensure that damage from the fire has not manifested after the damage inspection. Assessment Procedure. This attribute is scored based only on the occurrence of a fire on or below the structure being assessed. It is assumed that an appropriate assessment immediately following the fire incident (i.e., damage inspection) has been performed. Fire incident has occurred and an inspection 12 months after the fire has not occurred Bridge is not eligible for reliability assessment until inspection confirms that the bridge is undamaged There have been no incidence of fire on or below the bridge, or inspections conducted approximately 12 months or more an after the fire have confirmed that the bridge is undamaged Continue with procedure S.3 Susceptible to Collision Reason(s) for Attribute. This screening attribute can be used to screen an inventory or a family of bridges to identify those bridges with specific vulnerabilities to random or near- random damage from collision. This attribute is intended to apply to a limited number of bridges for which the risk of collision is unusually high or special. Simply because a bridge could be subjected to impact does not mean the likelihood of impact is high, and, in fact, it could actu- ally be quite remote. However, there are some structures that have been impacted many times in the past, where a channel or a roadway is particularly difficult to navigate, vertical clearance is inadequate, etc. that are much more likely to be struck. Examples include collisions from barges, debris, or heavy trucks. This attribute would typically be used to screen specific bridges that have an unusual or a unique risk of collision damage than a larger group or family of bridges which do not. In such cases, individual reliability analysis may be required. Assessment Procedure. This screening attribute should be assessed based on sound engi- neering judgment and is intended to screen bridges with unusual or special collision risks from an assessment of a group of bridges that do not. Highly susceptible to collisions Requires specialized assessment and/or mitigation Structure is not susceptible to collisions Continue with procedure S.4 Flexural Cracking Reason(s) for Attribute. When the primary load-bearing members in a concrete bridge exhibit flexural cracking, it may indicate that the members were either inadequately designed for the required loading, that overloads have occurred, or that deterioration has occurred that has reduced the load-bearing capacity of the members. In any case, large flexural cracks can be indicative of an inadequate load-bearing capacity that may require an engineering analysis in order to determine the cause of the cracking and the resulting effect on the load capacity of the structure. As a result, bridges exhibiting moderate to severe flexural cracking should be screened from the general reliability assessment unless appropriate engineering analysis indicates that the cracking is benign or corrective repairs have been made.

64 Proposed Guideline for Reliability-Based Bridge Inspection Practices The effects on the strength and the durability of a prestressed element due to flexural cracking are generally more significant than for a reinforced concrete element. Assessment Procedure. Flexural cracks will typically present themselves with a vertical ori- entation either near the bottom flange at mid-span or near the top flange over intermediate supports, if the member is continuous. Engineering judgment must be exercised in determining whether any present flexural crack- ing is moderate to severe. Crack widths in reinforced concrete bridges exceeding 0.006 inches to 0.012 inches reflect the lower bound of “moderate cracking.” The American Concrete Institute Committee Report 224R-01 (1) presents guidance on what could be considered reasonable or tolerable crack widths at the tensile face of reinforced concrete structures for typical conditions. These values range from 0.006 inches for marine or seawater spray environments to 0.007 inches for structures exposed to de-icing chemicals, to 0.012 inches for structures in humid, moist envi- ronments. In prestressed concrete bridge structural elements, tolerable crack width criteria have been adopted in the Precast/Prestressed Concrete Institute (PCI) Manual for the Evaluation and Repair of Precast, Prestressed Concrete Bridge Products (MNL-37-06). The PCI Bridge Committee recommends that flexural cracks greater in width than 0.006 inches should be evaluated to affirm adequate design and performance. Presence of moderate to severe flexural cracking in reinforced or prestressed concrete bridge elements Assess individually to determine source, extent, and effect of cracking Flexural cracking is not present, or it has been determined to be benign or repaired Continue with procedure S.5 Shear Cracking Reason(s) for Attribute. If the primary load-bearing members in a reinforced or a pre- stressed concrete bridge exhibit shear cracking, it may indicate that the members were either inadequately designed for the required loading, an overload has occurred, or that deteriora- tion has occurred that has reduced the load-bearing capacity of the members. In any case, shear cracks can be indicative of an inadequate load-bearing capacity requiring an engi- neering analysis in order to determine the cause of the cracking and the resulting effect on the load capacity of the structure. As a result, bridges exhibiting cracking attributable to a deficiency in shear strength should be screened from the reliability assessment unless appro- priate engineering analysis indicates that the cracking is benign or corrective repairs have been made. Assessment Procedure. Engineering judgment must be exercised in determining whether any present shear cracking is attributed to a shear strength deficiency. Shear cracks will typically present themselves with a roughly 45 degree diagonal orientation for conventionally reinforced concrete and down to roughly 30 degrees for prestressed elements, and will generally radiate toward the mid-span of the member. The ends of the member and any sections located over piers should be checked for this type of cracking. Presence of unresolved shear cracking Assess individually to determine source and extent of cracking Shear cracking is not present or it has been determined to be benign Continue with procedure

Attribute Index and Commentary 65 S.6 Longitudinal Cracking in Prestressed Elements Reason(s) for Attribute. This attribute is for the assessment of prestressed bridge elements. Longitudinal cracking in prestressed elements can be indicative of corrosion or fracture of the embedded prestressing strands. As a result, prestressed elements with reported longitudinal cracking should be individually assessed to determine the source of the cracking and the condi- tion of the prestressing strands. Assessment Procedure. This attribute is assessed based on data in the inspection report and engineering judgment. If longitudinal cracking is reported, further assessment may be required. Significant longitudinal cracking is present Assess individually to determine source and extent of cracking and condition of strand No significant longitudinal cracking Continue with procedure S.7 Active Fatigue Cracks Due to Primary Stress Ranges Reason(s) for Attribute. Active fatigue cracks in steel bridge elements due to primary stresses can propagate quickly and potentially lead to a fracture in the element. These cracks are distinguished from distortion cracks or out-of-plane fatigue cracks, which are more commonly observed, but generally less critical. Assessment Procedure. If any active fatigue cracks due to primary stresses are found in the element, it is strongly recommended that the element be retrofitted before continuing with this procedure. It is noted that a “stable” fatigue crack can potentially propagate to brittle fracture depending on the toughness of the material, the total applied stress, and the temperature. A fatigue crack can be considered “not active” if previous inspection reports show that the crack has not grown over a set period of time (e.g., the longest inspection interval plus 1 year). Pri- mary stresses are those stresses (i.e., stress ranges) that are readily calculated using traditional mechanics principles (e.g., MC/I or P/A) and are typically obtained during design or rating. Active fatigue crack(s) due to primary stresses present Retrofit before continuing No active fatigue crack(s) due to primary stresses present Continue with procedure S.8 Details Susceptible to Constraint-Induced Fracture (CIF) Reason(s) for Attribute. Details that are susceptible to CIF can lead to brittle fracture in the absence of any observable cracking. An example of this is the failure of the Hoan Bridge in December 2000 in Milwaukee, WI (2). The bridge had been in service for approximately 25 years before two of the three girders experienced full-depth fractures and the third girder had a crack that arrested in the flange. Inspection is not a valid method to prevent these types of failures from occurring (the Hoan Bridge was inspected a few days prior to the failure). Hence, the attribute is included as a screening criterion. Assessment Procedure. Details susceptible to CIF have a much higher probability of frac- ture failure than other types of details. It is recommended that CIF details be retrofitted or examined more closely before continuing with this process. Structure contains details susceptible to CIF Retrofit before continuing Structure does not contain details susceptible to CIF Continue with procedure

66 Proposed Guideline for Reliability-Based Bridge Inspection Practices S.9 Significant Level of Active Corrosion or Section Loss Reason(s) for Attribute. This attribute is intended to be used to screen bridges that have a significant level of existing or active corrosion sites that make the likelihood of severe corro- sion damage relatively high. A significant amount of active corrosion and/or section loss in an element increases the probability of severe corrosion damage developing in the near future. As a result, individual engineering assessments may be required to effectively assess the reliability characteristics for the element. Significant section loss would normally be visible for steel struc- tural members. Assessment Procedure. If a significant amount of active corrosion with section loss is found on a steel element it is recommended that the element be repaired before continuing with this process. Engineering judgment must be used to determine what is defined as a significant amount of active corrosion with section loss and assess its effects. Previous inspection reports and engineering judgment must also be used to determine whether or not the corrosion is active. Corrosion damage in steel elements that is inactive is explicitly distinguished from corrosion that is active. For example, section loss on a girder web that was the result of a leaking expansion joint that was corrected (the joint was replaced and the girder was repainted), could be classified as inactive corrosion if the expansion joint repair eliminates the vulnerability to corrosion. It is assumed that the owner has either determined that the existing section loss is insignificant or has taken it into account in the rating procedures and load posting, if needed, is in place. Significant level of active corrosion and section loss Repair before proceeding Active corrosion or section loss is not significant or has been repaired Continue with procedure S.10 Design Features Reason(s) for Attribute. This attribute is intended to be used to screen bridges that have unusual or unique design features that make the likelihood of serious damage either usually high or unusually uncertain, relative to other bridge in the same family, or identify bridges with dif- ferent anticipated deterioration patterns than other bridges in a group or family. This attribute can be used to subdivide a family of bridges into two or more groups with similarly anticipated deterioration patterns, based on specific design features that are not common to each sub-group. Design features for use as screening items should be identified by the RAP. Two examples below are provided to illustrate the way in which this attribute might be used. Bridges with pin and hanger connections: Pin and hanger connections generally have a history of presenting maintenance challenges. As such, it may be desirable to screen a bridge that includes this particular type of connection from a larger family, such as a family of steel multi-girder bridges. Jointless bridges: Jointless bridges are typically less susceptible to corrosion-related damage associated with leaking joints in the bearing areas. As such, the deterioration patterns may differ from other bridges of similar materials and general overall design. Assessment Procedure. Unique or unusual design features should be identified through review of bridge plans or other documentation describing the design features of a bridge. Bridge has unique or unusual design feature Screen Bridge does not have unique or unusual design features Proceed

Attribute Index and Commentary 67 Design Attributes D.1 Joint Type Reason(s) for Attribute. Bridge joint types can be categorized as either closed systems or open systems. Compared to open joint systems, closed joint systems provide for higher durability based on the way their designs shield the inner workings of the joint from dirt and debris. This, in turn, increases the amount of time before a joint begins to leak onto other bridge components. The presence of open-type deck joints increases the probability of chloride-contaminated water leaking onto bridge elements below the deck, thus increasing the likelihood of corrosion-related damage. Assessment Procedure. This attribute is rated based on the presence of open joints. Open joint system 10 points Closed joint system 0 points D.2 Load Posting Reason(s) for Attribute. The presence of a load posting typically indicates that the given bridge was either not designed to carry modern loading or that the bridge has become damaged and its structural capacity has been reduced. A structure of this type may be more likely to expe- rience damage from heavy traffic and dynamic loading. This attribute is intended to consider the contribution of high and possibly even excessive loads on accelerating damage generally for a given bridge or a family of bridges. Engineering judgment is necessary to evaluate if this attribute is applicable. Considerations include the likelihood of the applied loading being higher than (i.e., illegal) or near the load posting. In some cases, traffic patterns are such that the fact that the bridge is load posted will not affect the rate of damage accumulation on the bridge. For example, a bridge is load posted for the state’s legal truck load, but is located on a parkway where trucks are prohibited. Assessment Procedure. This attribute is scored based only on whether or not a bridge has been load posted; the level of the rating does not need to be considered. This assessment should consider if the load posting has a significant effect on the durability of the bridge. Structure is load posted 20 points Structure is not load posted 0 points D.3 Minimum Vertical Clearance Reason(s) for Attribute. This attribute is intended to consider the likelihood that a bridge may be impacted by an over-height vehicle and damaged such that the deterioration rate of the superstructure elements may be increased. For concrete bridges, impacts may damage the embedded reinforcement or the prestressing strands, or damage the typical concrete cover exposing the steel to the environment. For steel bridges, impacts can deform members and damage coating systems in the areas of the impact. Impact damage that affects the structural capacity of the bridge requires a damage inspection and an assessment beyond the scope of a typical reliability assessment. Users may wish to use this attribute to include the potential for increased deterioration rates for bridges that experience frequent impact damage. The bridge superstructure’s minimum vertical clearance influences on how often it will be impacted. A bridge with a lower vertical clearance will be more likely to experience impact damage than a bridge with higher vertical clearance. The likelihood of being hit may also

68 Proposed Guideline for Reliability-Based Bridge Inspection Practices depend on the traffic composition of the roadway below, such as the average daily truck traf- fic (ADTT). This attribute is generally based on the total vertical clearance between the bottom of the girders and the riding surface of the roadway below. The functional classification of the roadway below the bridge may also be a consideration. NBIS data fields record the vertical clearance and the functional classification of the route passing under the bridge, and are rated using the model provided in the coding guide (3), which is provided in Table E2. Assessment Procedure. This attribute should be scored based on appropriate measure- ments or on the information stored in the bridge file. The suggested scoring models shown below consider only the vertical clearance of the bridges. Users may wish to consider the func- tional classification or the typical traffic pattens below the bridge in their assessment. In the scoring models shown, increased importance is given to over height clearances for prestressed concrete bridges relative to steel and conventionally reinforced bridges. This is due to the poten- tial for strand corrosion when the concrete cover is damaged by impact, and the increased rate of deterioration for strands relative to mild steel. Prestressed Concrete Girders Vertical clearance is 15 feet or less 20 points Vertical clearance is between 15 feet and 16 feet 15 points Vertical clearance is between 16 feet and 17 feet 10 points Vertical clearance is greater than 17 feet or no under traffic present 0 points Steel and Concrete Girders Vertical clearance is 14 feet or less 15 points Vertical clearance is between 14 feet and 15 feet 12 points Vertical clearance is between 15 feet and 17 feet 7 points Vertical clearance is greater than 17 feet or no under traffic present 0 points Table E2. FHWA coding guide minimum vertical underclearance provisions. Underclearance Code Minimum Vertical Underclearance Functional Class Railroad Interstate and Other Freeway Other Principal and Minor Arterials Major and Minor Collectors and Locals 9 >17 ft >16.5 ft >16.5 ft >23 ft 8 17 ft 16.5 ft 16.5 ft 23 ft 7 16.75 ft 15.5 ft 15.5 ft 22.5 ft 6 16.5 ft 14.5 ft 14.5 ft 22 ft 5 15.75 ft 14.25 ft 14.25 ft 21 ft 4 15 ft 14 ft 14 ft 20 ft 3 Rating <4 and requiring corrective action 2 Rating <4 and requiring replacement 1 No value indicated 0 Bridge closed

Attribute Index and Commentary 69 D.4 Poor Deck Drainage and Ponding Reason(s) for Attribute. This attribute is intended to consider the adverse effects of poorly designed deck drainage systems and the possibility of ponding on the deck surface, as well as for inadequate provisions for preventing scuppers and drains from splashing de-icing chemicals onto the superstructure below. Ineffective deck drainage increases the likelihood of bridge ele- ments developing corrosion related damage. This results from drainage onto the superstructure and the substructure elements. Both concrete and steel elements will have an increased suscepti- bility to corrosion damage when exposed to prolonged periods of wetness and/or frequent wet- dry cycles. The presence of chlorides from de-icing chemicals applied to the deck also increases the likelihood of corrosion damage to these elements. This attribute can also be used to characterize decks with ponding or with drain diversion issues. When water is allowed to sit on the surface of the deck, there is an increase in the likeli- hood that corrosion of the reinforcing steel will initiate and damage will propagate. Water and chlorides are more likely to penetrate to the level of the reinforcement when periods of wetness are prolonged and chloride concentrations at the surface are high. Assessment Procedure. This attribute is scored based on the drainage design of the bridge and any known ponding or drainage issues, as noted in the inspection report. Drainage systems which normally allow water to run off onto the components below the bridge deck are con- sidered ineffective, regardless of whether they have sustained any damage or not. Deck drains through curb openings, where the water from the decks typically drains onto superstructure elements, are an example of poor deck drainage. Decks with ponding issues may need to be individually scored. Ponding or ineffective drainage 10 points No problems noted 0 points D.5 Use of Open Decking Reason(s) for Attribute. The presence of an open deck increases the likelihood that corro- sion of the steel superstructure will occur. An open deck allows water, de-icing chemicals, and other debris to fall directly onto the superstructure instead of running into deck drains and then to downspout pipes, as they would in a closed deck system. As a result, the likelihood of damage occurring in superstructure elements, bearing, and substructure elements is greatly increased. Users may also use this as a screening attribute. Assessment Procedure. The attribute is scored based on whether or not the bridge contains an open deck. Common types of open decks include timber or open grating decks. Bridge has an open deck 20 points Bridge does not have an open deck 0 points D.6 Year of Construction Reason(s) for Attribute. This attribute reflects the influence of bridge age and historic design on the most prevalent aging mechanisms in highway bridges—deterioration of concrete associated with corrosion of embedded reinforcement, and corrosion damage and/or fatigue and fracture for steel structures. The corrosion of embedded reinforcing steel occurs due to the penetration of chlorides, water, and oxygen to the level of the reinforcement. For intact concrete, the penetration of the chlorides

70 Proposed Guideline for Reliability-Based Bridge Inspection Practices is presently modeled as a diffusion process, using Fick’s Law, which depends on time, tempera- ture, the permeability of the concrete, and the concentration of chlorides at the component’s surface. Additionally, if the concrete has suffered damage, such as cracking or spalling, chlorides can more easily concentrate at the reinforcement, effectively expediting the corrosion process. The quality of the concrete used in bridge construction has generally improved over time due to concrete technology innovation, improvements in quality control, and in better sup- plier understanding of optimal material selection for strength and durability. Therefore, it is reasonable to expect that a concrete component constructed to modern standards is likely to have improved corrosion resistance characteristics compared to older components. Addition- ally, older structures have been exposed to the surrounding environment for a longer period of time, and are therefore more likely to be affected by corrosion. With respect to steel girders, the year the bridge was designed can provide valuable information about the susceptibility of the bridge to fatigue cracking and fracture. Over the years, there have been numerous changes in design specifications that have resulted in the improved fatigue and fracture resistance of bridges. Four key dates have been identified; 1975, 1985, 1994, and 2009, with regard to changes in design specifications. These dates were selected for the following reasons: 1975 Fatigue The “modern” fatigue design provisions, based on the research of Fisher and others, were fully incorporated into the AASHTO Specifications with the 1974 Interims. The basic detail categories have not changed significantly since their introduction. Hence, 1975 was selected as a differen- tiator regarding fatigue design of steel bridges. Prior to 1975, fatigue design was based on prin- ciples that were not generally appropriate for welded structures. Although these early provisions appeared in the 1965 version of the specifications and were in place through 1976, it was felt that it was reasonably conservative to ignore the earlier provisions and set the cutoff date at 1975. Fracture In 1974, partly in response to the Point Pleasant Bridge collapse (1967), mandatory Charpy V-Notch (CVN) requirements were set in place for welds and base metals as a part of the AASHTO/AWS Fracture Control Plan. The purpose of these CVN requirements was to ensure adequate fracture toughness of materials used in bridges. Furthermore, “modern” fatigue design provisions, based on the research of Fisher and others, were fully incorporated into the AASHTO Specifications as previously discussed. Hence, 1975 was selected as a differentiator regarding fatigue and fracture design of steel bridges. 1985 In 1985, AASHTO introduced changes to address and to prevent distortion-induced fatigue cracking. A common example of distortion-induced fatigue cracking is web-gap cracking. Hence, considering the specifications introduced in 1975 and 1985, bridges designed after 1985 are less likely to be susceptible to fatigue due to primary or secondary stress ranges than bridges built prior to these revisions. 1994 In 1994, the AASHTO design specifications changed from load factor design (LFD) to load and resistance factor design (LRFD). The LRFD method is intended to ensure greater reliability in bridge design. There were several changes regarding the load models and the load distribution factors used for the fatigue limit state. These changes were intended to result in a more realistic and reliable fatigue design. Hence, for the fatigue limit state, bridges designed after 1994 would be expected to have improved reliability.

Attribute Index and Commentary 71 2009 In 2008, language was introduced into the AASHTO LRFD Bridge Design Specifications which directly addressed the issue of CIF. The article provided prescriptive guidance to ensure that details susceptible to CIF are avoided. It is included in the 2009 and later versions of the AASHTO LRFD Bridge Design Specifications. Assessment Procedure. The year of construction is intended to characterize the years of environmental exposure a component has experienced or the fatigue susceptibility of the design. The suggested values are intended to put elements into four broad classes that range from very old to relatively new. For elements that have been replaced, the year of the replacement should be used. Elements that have been rehabilitated should use the original construction date. These ranges are advisory; users may consider modifying these categories based on experience with their bridge inventory or significant changes to construction practices that may have occurred within their state. For steel-girder categories, users should consider if the design specifica- tion used in the design of the bridge matched the contemporary specifications at the time, as described above. If, for example, the LRFD provisions of 1994 were not implemented in the state until 2000, then the ranges should be adjusted accordingly. Concrete Bridge Decks, Prestressed Girders, Substructures Built before 1950 10 points Built between 1950 and 1970 6 points Built between 1970 and 1990 3 points Bridge is less than 20 years old 0 points Steel Girders, Fatigue Bridge designed before 1975/unknown 20 points Bridge designed between 1976 and 1984 10 points Bridge designed between 1985 and 1993 5 points Bridge designed after 1994 0 points Steel Girders, Fracture Bridge designed before 1975/unknown 20 points Bridge designed between 1975 and 1984 10 points Bridge designed between 1985 and 1993 5 points Bridge designed between 1994 and 2008 3 points Bridge designed after 2009 0 points D.7 Application of Protective Systems Reason(s) for Attribute. Protective systems such as membranes, overlays, or sealers may be applied to the surface of a concrete element to reduce the ingress of water, which may contain dissolved chlorides or other corrosive substances. When these corrosive materials diffuse to the level of the reinforcement, the likelihood of reinforcement corrosion increases, which may lead to the propagation of damage. Protective systems delay or prevent this process from occurring thereby reducing the likelihood for future corrosion damage. Some overlays have also been shown to delay the development of spalling as a result of an increased resistance to cracking and an increased ability to confine delamination damage (4).

72 Proposed Guideline for Reliability-Based Bridge Inspection Practices An overlay is defined herein as an additional layer of protective material, which is applied on top of the concrete deck and that also serves as the riding surface. Overlays may consist of asphalt, latex-modified concrete, low-slump dense concrete, silica fume concrete, polymer con- crete, or other materials. A membrane is defined herein as a barrier that is placed on top of the concrete deck and is then covered by another material, which serves as the riding surface. Common membranes may consist of hot-rubberized asphalt, resin, bitumen-based liquid, or prefabricated sheets. Sealers are somewhat different from overlays and membranes in that they are applied thinly to concrete surfaces and penetrate the porosity of the concrete to seal it from moisture. Ini- tially, sealers were used to counteract freeze-thaw damage and de-icing chemical-application related scaling. With the proper use of air-entraining admixtures, the primary purpose of sealers changed to preventing or slowing the ingress of chlorides (5). Types of sealers include silanes, siloxanes, silicates, epoxies, resins, and linseed oil. Surface coatings such as epoxy, polyurethane, or polyurea may also be applied to the concrete elements of a bridge in order to increase their resistance to water intrusion and consequently reduce their probability of developing corrosion damage. The application of these coatings can improve the durability and corrosion resistance of concrete elements. Each of these protective systems is intended to delay or prevent corrosion damage in concrete bridge elements. If the protective systems are effective, then the likelihood of corrosion-related damage will be reduced compared to unprotected elements of similar design characteristics and environmental conditions. As a result, the application of protective systems may be considered in the reliability assessment. Assessment Procedure. If protective systems such as membranes, overlays, or sealers have been applied to a concrete element, their effectiveness should be evaluated based on engineering judgment and local experience or test data along with any documented research and field test- ing data that is available. Important factors to consider include the effectiveness of the applied system as well as how often that system is applied or maintained. This attribute assumes that overlays and sealers generally have similar effects in terms of corrosion protection for the deck. Based on their experience, users may wish to separate certain overlays or membrane systems. For example, an owner may have experience that indicates that low-slump overlays are having a significant effect on extending the service life of bridge decks. In that case, the owner may wish to increase the importance of this attribute to a moderate or high level, and distribute the scor- ing appropriately. The suggested scoring assumes the protective system has a low importance relative to other design characteristics. Never applied, poor functioning, or non-functioning 10 points Yes, penetrating sealer, crack sealer, limited effectiveness 5 points Yes, periodically applied, effective 0 points D.8 Concrete Mix Design Reason(s) for Attribute. Concrete mix designs, such as those considered to be “HPC,” typically have a lower permeability and a higher durability than other traditional concrete mixes. Therefore, high performance mixes provide an increased resistance to de-icer or marine environment-based chloride ion penetration. This in turn can increase the time to corrosion initiation in reinforcing steel. This design attribute is intended to consider the increased durabil- ity provided by HPC mixes.

Attribute Index and Commentary 73 The permeability of a concrete mix depends on several factors including the water to cementi- tious ratio, the use of densifying additives, and the use of mix-improving additives. Supplemen- tary cementitious materials such as fly ash, ground-granulated blast furnace slag, and silica fume have been shown to reduce permeability. Additionally, a properly designed and placed concrete mix with a lower water to cementitious ratio will have a lower permeability. Materials and criteria that have been identified as being beneficial in enhancing the per- formance of concrete bridge decks can be found NCHRP Synthesis 333: Concrete Bridge Deck Performance (5). Assessment Procedure. The evaluation of a bridge’s concrete mix design should be based on information contained in the bridge’s design plans and on engineering judgment. Many different types of concrete mixtures can be considered to be high performance, therefore, users should consider the corrosion resistance characteristics of the particular mixture and assess if the concrete mix used is expected to provide an increased durability relative to a typical concrete mix design. Past experience with concrete mixes of similar characteristics should be considered. The concrete used is not considered to be high performance 15 points The concrete used satisfies high performance conditions 0 points D.9 Deck Form Type Reason(s) for Attribute. Concrete decks constructed with stay-in-place (SIP) forms have the surface of the deck soffit hidden from visual inspection. Signs of corrosion damage such as efflorescence, rust staining, and cracking in the deck soffit cannot typically be observed. As a result, there can be increased uncertainty in the condition of the deck determined through visual inspection. This attribute is intended to consider the increased level of uncertainly in the deck condition that may exist when SIP forms are used. Assessment Procedure. This attribute is assessed based on whether the deck has SIP forms. SIP forms 10 points Removable forms 0 points D.10 Deck Overlays Reason(s) for Attribute. Similar to SIP forms, deck overlays prevent the visual observation of the deck condition. Signs of deterioration, corrosion damage, and cracking of the deck cannot typically be observed. As a result, there can be increased uncertainty in the condition of the deck determined through visual inspection. This attribute is intended to consider the increased level of uncertainty in the deck condition that may exist for decks with overlays. Assessment Procedure. This attribute is assessed based on whether or not the deck has an overlay. Deck has an overlay 10 points Bare deck 0 points D.11 Minimum Concrete Cover Reason(s) for Attribute. This attribute is intended to consider the improved corro- sion resistance and the increased durability associated with adequate concrete cover, and the

74 Proposed Guideline for Reliability-Based Bridge Inspection Practices historically poor performance of bridge elements with inadequate cover. The depth of concrete cover characterizes how far corrosive agents need to travel in order to reach the embedded steel reinforcement. Several studies have identified that the depth of concrete cover over the top rein- forcing steel mat is the most significant factor contributing to the durability of decks (5). The importance of adequate concrete cover is also an important durability factor for other concrete elements. The value used for this attribute should be the actual amount of concrete cover, which may not necessarily be the design cover. If quality control procedures are adequate to ensure that the design cover matches the as-built cover, the design cover may be used. If such quality control procedures have not been utilized or have historically been inadequate, it may be necessary to assess the as-built cover. In 1970, the general recommendation for concrete cover was a minimum clear concrete cover of 2 inches over the top-most steel. Currently, the AASHTO Standard Specifications for Highway Bridges (2002) requires a minimum concrete cover of 2.5 inches for decks that have no posi- tive corrosion protection and are frequently exposed to de-icing chemicals. Positive corrosion protection may include epoxy coated bars, concrete overlays, and impervious membranes. The AASHTO LRFD Bridge Design Specifications (2004) also requires a minimum concrete cover of 2.5 inches for concrete that is exposed to de-icing chemicals or on deck surfaces that are subject to stud or chain wear. The concrete cover may be decreased to 1.5 inches when epoxy coated reinforcement is used. It is also important to note that the type of damage and the rate of damage development vary with the amount of concrete cover. It has been reported that the type of damage changes from cracks and small, localized surface spalls to larger delaminations and spalling as the concrete cover increases (4). There is also an increase in the time to corrosion initiation and a reduction in the rate of damage development when cover increases, as shown schematically in Figure E1. In summary, as concrete cover increases, the time to corrosion initiation increases due to the increased depth that chloride ions must penetrate to initiate the corrosion process. As corrosion progresses, an increased concrete cover provides confinement that reduces the rate and the type of damage that develops at the surface of the concrete element. It should be noted that concrete cover greater than 3 inches can result in increased cracking, providing pathways for the intrusion of water and chlorides. This may be a consideration in special cases in which the concrete cover is unusually large. Figure E1. Effect of concrete cover on the time to corrosion initiation and development of damage (4).

Attribute Index and Commentary 75 Assessment Procedure. This attribute is scored based on the actual, physical clear cover which with the specified bridge element operates. The user should consider whether quality control practices used at the time of construction were adequate to provide confidence that the as-built concrete cover conforms to the design concrete cover, or if there are indications that the concrete cover may not be adequate. In these cases, the as-built concrete cover may be required and can be easily obtained using a covermeter. 1.5 inches or less, unknown 20 points Between 1.5 inches and 2.5 inches 10 points Greater than or equal 2.5 inches 0 points D.12 Reinforcement Type Reason(s) for Attribute. This attribute is intended to characterize whether or not the embedded reinforcing steel has a barrier to protect it against corrosion. The most commonly used barrier is an epoxy coating; however, galvanized bars and stainless steel, either as cladding or as solid bars, have also been used. Uncoated steel reinforcement will corrode easily and significantly when under attack from corrosive elements such as chloride ions, oxygen, and water. Since this exposure is inevitable in an operating structure, one way to slow the corrosion process is to coat the mild steel bars with either an organic or a metallic coating or to use an alternate solid metal bar, such as stain- less steel. These coatings or alternate bars help slow the corrosion process by providing either a physical or a metallurgical barrier against the action of the corrosive elements. The most commonly used barrier coating is fusion-bonded epoxy powder. This type of coat- ing has been used since 1973 and has been the subject of a significant body of research. It has been shown that, in reinforced concrete decks, if only the top mat is coated, for every year required to consume a given amount of mild steel, it will take 12 years for the epoxy coated bar to lose that same amount of metal. If both the top and bottom mats are coated, it may take up to 46 years (6). This significant increase when both mats are coated is due to increased electrical resistance, which further slows corrosion. Two of the more common metallic coatings used are zinc and stainless steel. Zinc coated bars are also known as galvanized bars. Conflicting reports have been given on the performance of galvanized bars, mostly with respect to varying levels of the water to cement ratio and to whether or not galvanized bars are used in conjunction with mild steel bars. Research suggests that gal- vanized bars may add 5 more years to the 10 to 15 years required for corrosion-induced stress to manifest in unprotected bridge decks (6). Solid stainless steel or stainless steel clad mild steel bars have also been used, although to a lesser extent due to their higher costs. Research conducted by the State of Virginia compared the perfor- mance of stainless steel clad and stainless steel bars with uncoated carbon steel bars. The research concluded that defect-free stainless steel clad bars performed nearly identically to the solid stainless steel bars. These types of bars were determined to tolerate at least 15 times more chloride than the carbon steel bars (6). Regardless of the specific coating or reinforcement material used, protected bars generally have a higher resistance to corrosion damage than uncoated, mild steel bars. As such, the scoring for this attribute considers only if the rebar is protected by one of these methods, or if it is not. Assessment Procedure. The type of reinforcement is scored based on the presence of bar- rier coatings or the use of alternative metal for the embedded reinforcement. This information

76 Proposed Guideline for Reliability-Based Bridge Inspection Practices can typically be identified from the structure’s design plans. If suitable information is unavail- able, engineering judgment should be used. Reinforcement is uncoated carbon steel 15 points Reinforcement has a protective coating or is produced from an alternate corrosion resistant metal (e.g., stainless steel) 0 points D.13 Built-Up Member Reason(s) for Attribute. Many bridges, especially older structures, contain built-up members. These built-up members are sometimes more susceptible to corrosion than nor- mal rolled steel sections because they contain pockets or crevices, which can retain water, salt, debris, etc. This has been known to result in an accelerated corrosion rate since debris and moisture can remain trapped. Bridge washing, if thoroughly performed, can mitigate these effects. Assessment Procedure. For this attribute, a built-up member refers to riveted or bolted members. Welded members should not be included in this assessment because they do not con- tain the type of pockets or crevices that can trap corrosion inducing materials. Element is a built-up member 15 points Element is not a built-up member 0 points D.14 Constructed of High Performance Steel Reason(s) for Attribute. In addition to possessing higher yield strengths than normal steels, high performance steels (HPSs) generally have greater fracture toughness than that required by ASTM A709, and of other common bridge steels. Improved fracture toughness results in steel that is more resistant to fracture than normal steels. This is because it is more likely that cracks will propagate at a slower rate, and could even arrest, in HPS compared to normal steels. At this time, the CVN levels required for HPS in ASTM A709 are not established with the objective of achieving any particular level of fracture resistance or crack tolerance. Hence, the benefits provided by using HPS, if the steel just meets the ASTM A709 specification, are limited. Therefore, the suggested ranking of HPS is low in terms of contribution to durability and reli- ability (10 pts), relative to normal steel. This may change as future research becomes available and the minimum required CVN values increase for HPS. Assessment Procedure. This attribute should be scored based on whether or not the ele- ment is constructed out of HPS. If there is no documentation or it is unknown if the element is constructed of HPS, the attribute should be scored accordingly. Element is not constructed of HPS/unknown 10 points Element is constructed of HPS 0 points D.15 Constructed of Weathering Steel Reason(s) for Attribute. Weathering steel is a type of steel that contains alloying elements that increase the inherent corrosion resistance of the steel. For this reason, weathering steels are less susceptible to corrosion than normal black steels. However, this is only true if the steel is used in the proper environment and is detailed properly.

Attribute Index and Commentary 77 Assessment Procedure. This attribute is scored based on whether or not the element is constructed using weathering steel and is detailed and located in a manner that minimizes the contact of the steel with de-icing chemicals and moisture. If it is unknown if the element is com- posed of weathering steel, the element should be scored accordingly. The assessment procedure assumes that the steel is used in the proper environment and is detailed properly. Guidance on the appropriate application of uncoated weathering steel can be found in FHWA Techni- cal Advisory T-5140.22 (7). The document also includes recommendations for maintenance to ensure continued successful performance of the steel. Element is not constructed of weathering steel or location and detailing may allow impact of ambient or de-icing chemicals on steel surfaces 10 points Element is constructed of weathering steel and properly detailed consistent with FHWA Technical Advisory T-5140.22 0 points D.16 Element Connection Type Reason(s) for Attribute. Welded connections are usually more susceptible to the effects of fatigue damage than other types of connections, as there is a direct path for cracks to propagate between connected elements. For example, a crack in a flange can grow into the web through the web-to-flange weld. Fatigue cracking is generally of greatest concern for welded details that have low fatigue resistance, such as D, E, and E’, along with residual stresses and weld toe defects. Riveted connections, unlike welded connections, do not offer a direct path for cracks to prop- agate from one element to another. Using the web-to-flange connection example, cracks in an angle used to make up a flange are not able to grow directly into the web plate because the ele- ments are not fused together. Hence, there is a certain amount of redundancy at the member level. Nevertheless, the quality of the rivet hole (e.g., punched vs. drilled) and a lack of consistent pretension in rivets results in these details being classified as category D. Similar to riveted connections, high strength (HS) bolted connections are more resistant to a fatigue crack propagating from one component of a member to another, as compared to welded members. A properly tightened HS bolt generates very high compressive forces in the connection. The pretension force is much greater and is much more consistently achieved in a HS bolted connection than in a riveted connection. As a result of the significant pretention in a fully tightened A325 or A490 bolt, the quality of the hole itself has little or no effect on the fatigue resistance of the connection (in contrast to riveted joints). As a result, they are classified as category B details. It is noted that considering the element connection type may appear to be a double penalty when considered in conjunction with D.17 Worst Fatigue Detail Category. However, it is clear that should cracking occur at a welded detail in a main member, it is more likely to become an issue than in, say, the equivalent bolted detail simply due to the fact that there is no direct path for cracks to grow from component to component in the bolted joint. Hence, it is considered a “better” condition even though both welded and bolted details may both be classified as category B. Riveted details, which do not have as high a fatigue resistance as HS bolted con- nections, but are not as susceptible to crack propagation as welded joints, have been arbitrarily scored in the middle. Assessment Procedure. If the element has multiple types of connections, the worst type of connection should be scored for this attribute.

78 Proposed Guideline for Reliability-Based Bridge Inspection Practices Element connected with welds 15 points Element connected with rivets 7 points Element connected with HS bolts 0 points D.17 Worst Fatigue Detail Category Reason(s) for Attribute. The likelihood of fatigue cracking is influenced by the type of fatigue detail category present. It is generally accepted that poor fatigue details are more likely to develop cracks than more fatigue resistance details. This is implied in the current AASHTO LRFD Bridge Design Specifications, which discourages the use of details lower than category C and encourages design for infinite life. Fortunately, since the introduction of the modern AASHTO fatigue provisions in 1975, the use of poor details (D, E, and E’) has been greatly reduced. Hence, details in bridges designed over the past 30 years or so will typically be of higher fatigue resistance. Assessment Procedure. The worst type of detail subjected to tensile stress ranges in the element or member should be used for this attribute. The AASHTO fatigue details A through E’ should be used. Fatigue detail category E or E’ 20 points Fatigue detail category D 15 points Fatigue detail category C 5 points Fatigue detail category A, B, or B’ 0 points If the element has multiple types of connections, the worst type of connection should be scored for this attribute. Element connected with welds 15 points Element connected with rivets 7 points Element connected with HS bolts 0 points D.18 Skew Reason(s) for Attribute. Bridge skew can introduce unanticipated forces in a bridge deck, deck joints, and superstructures. Thermal expansion of the superstructure and deck may intro- duce uneven strain distributions and/or torsional forces. As a result, bridges with high skew angles may suffer atypical deterioration patterns including cracking in bridge decks, failure of joints and bearing, and distortion-induced cracking at diaphragms (8–12). Assessment Procedure. This attribute is typically scored based on the recorded skew angles for a bridge. Angles of 30 degrees or greater may be used as a value for evaluating the potential for adverse skew angle effects. This attribute may also be used as a screening attribute. Skew 30° or more 20 points Skew 20–30° 10 points Skew less than 20° 0 points D.19 Presence of Cold Joints Reason(s) for Attribute. Cold joints or construction joints within deck spans can some- times result in leakage of water and de-icing chemicals through the deck and onto the supporting

Attribute Index and Commentary 79 superstructure. This may result in accelerated deterioration patterns including coating failure and section loss for steel members, corrosion damage in concrete members, and/or corrosion damage in the deck. Assessment Procedure. This attribute is typically scored based on the presence of known cold joints within the deck span. Data to support this assessment may come from inspection reports, because cold joints that are performing as designed may not be known. Presence of cold joints 10 points No known cold joints 0 points D.20 Construction Techniques and Specifications Reason(s) for Attribute. Construction techniques and specifications have evolved over time to improve the durability and performance characteristics of bridges. Certain construc- tion techniques and specifications used during previous eras may be problematic, and result in deterioration and damage patterns that can be associated with the techniques or specification in use at the time of bridge construction. For example, reduced bridge deck thickness may have been typical during a certain era. Over time, the reduced deck thickness may be shown to reduce the durability of the bridge deck and result in deck damage such as punch-through. As a result, decks constructed during that era may be more likely affected by a certain damage mode than bridges constructed during other eras. Assessment Procedure. This attribute will typically be identified by RAP members based on experience of bridge inspection and maintenance personnel. Historical records documenting the evolution of design standards and construction techniques may be necessary to identify the specific era, or estimates based on experience may be used. This attribute may also be used as a screening attribute. Bridge constructed during identified era 20 points Bridge not constructed during identified era 0 points D.21 Footing Type Reason(s) for Attribute. Spread-type footings may be susceptible to the adverse effects of scour, soil sliding, or rotations due to uneven settlement or subsidence. In contrast, pile foun- dations may be unaffected by these phenomena. As such, deterioration patterns and damage modes that affect spread footings may not be relevant for pile foundations. Assessment Procedure. This attribute can typically be determined from the design drawing available in the bridge file. This attribute may be used as screening criteria for specific damage modes that affect spread footings, but would not affect pile foundations. Spread-type footing 15 points Pile foundation 0 points D.22 Subsurface Soil Condition Reason(s) for Attribute. Footings on certain soils may be susceptible to the effects of soil sliding or rotations due to uneven settlement or subsidence. This attribute is typically utilized in conjunction with D.21 to reflect the increased likelihood of damage modes such as substructure rotations, cracking, or displacements for bridges in certain geographic regions.

80 Proposed Guideline for Reliability-Based Bridge Inspection Practices Assessment Procedure. Subsurface soil conditions susceptible to these effects are typically known to geotechnical engineers and/or maintenance personnel. This attribute may be identi- fied based on soil testing results or experience. Poor or unknown subsurface soil conditions 20 points Acceptable soil condition or pile foundations 0 points Loading Attributes L.1 ADTT Reason(s) for Attribute. The ADTT on a bridge is used to characterize the frequency of occurrence of large external loads on the bridge due to heavy vehicles. Large transport trucks or other heavy vehicles place stress on a bridge as static and dynamic loads, the latter reflecting impact and other dynamic amplification effects. As ADTT levels increase, the rate of damage formation and accumulation in concrete is typi- cally expected to increase. This is in part because the stresses caused by traffic loads accelerate the effects of the internal expansion forces from reinforcement corrosion (4). These loads, especially when placed on a bridge with existing deterioration, will open cracks and possibly allow corro- sive elements to enter the cracks or increase the crack density. Experience has shown that bridge decks exposed to heavy truck traffic generally deteriorate at a much higher rate than decks with little or no truck traffic. For steel girders, research has shown that trucks produce nearly all of the fatigue damage in highway bridges. Hence, a bridge with high truck traffic (high ADTT) will have a higher prob- ability of fatigue damage. Of course, the converse is also true, bridges with little or no truck traffic (e.g., HOV bridges) are unlikely to experience fatigue cracking. It is important to note that ADTT only considers the “load” side of the equation. The likeli- hood of fatigue cracking also depends on the “resistance” side of the equation, which is addressed by the D.16 Element Connection Type and D.17 Worst Fatigue Detail Categories. Although ADTT does not provide an exact correlation to the stress ranges an element will experience, it does provide a reasonably good understanding of how quickly fatigue damage may accumulate. Assessment Procedure. This attribute should be scored based on the ADTT. For steel structures, the scoring limits for ADTT were taken from a recent study on fracture critical bridges titled A Method for Determining the Interval for Hands-On Inspection of Steel Bridges with Fracture Critical Members (13). Although these limits were developed primarily with fracture critical bridges in mind, it was decided these limits could be applied to other highway bridges as well for the fatigue limit state. The reasoning behind the limits as documented in Parr and Connor’s report is as follows: “The ADTT limit of 15 comes from the fact that for bridges where the ADT is less than 100, the ADT is generally not reported in the NBIS. During the Purdue University Workshop, it was agreed than an ADTT of 15% (of the ADT) was a reasonably conservative estimate of the pro- portion of trucks crossing a typical low volume bridge. Hence, 15% of the lowest ADT reported in the NBIS (ADT = 100) yields an ADTT of 15. The lower bound value of 100 was set such to separate bridges in rural areas versus ‘moder- ately’ traveled bridges. The upper bound limit of an ADTT equal to 1,000 was obtained by simply increasing the ‘moderate’ limit by a factor of 10. It was included simply to create a boundary between ‘heavily’ and ‘moderately’ traveled bridges.”

Attribute Index and Commentary 81 For concrete bridges, high ADTT will likely have the most significant effect on the durabil- ity of the bridge deck. Superstructure components will be affected to a much lesser extent; if designed to modern standards, high ADTT may have little effect on the durability of superstruc- ture components. Deck joints may also deteriorate more rapidly in the presence of high ADTT. Users may wish to adopt different thresholds for the scoring model, depending on typical traffic patterns and needs. Concrete Bridge Deck, Prestressed Concrete Girder ADTT is greater than 5,000 20 points ADTT is moderate 10 points ADTT is minor 5 points No heavy trucks 0 points Steel Girders ADTT is greater than 1,000 20 points ADTT is between 100 and 1,000 15 points ADTT is between 15 and 100 5 points ADTT is less than 15 0 points L.2 Dynamic Loading from Riding Surface Reason(s) for Attribute. This attribute is intended to consider the detrimental effects of dynamic loading on the deterioration patterns for concrete bridge decks. This attribute would typically be used to adjust assessments to consider a reduction of the durability of bridge decks with high dynamic loads (i.e., high speed traffic and high ADTT). This attribute is included to consider cases where the riding surface or the deck joint becomes damaged, such as through the development of potholes, rough patches, or a bump at the end of the bridge, and increased dynamic forces are created due to the traffic loading. These forces place additional stress on the structure leading to a perpetual cycle of damage propagation that accelerates the rate of deterio- ration for the deck element (14). Assessment Procedure. This attribute is based on engineering judgment. Considerations in assessing this attribute include the roughness of the riding surface, the existence of potholes and patches, durability of deck joints, ADTT, and traffic speeds. Dynamic forces leading to increased rate of deterioration a significant consideration 15 points Dynamic forces not a significant consideration 0 points L.3 Exposure Environment Reason(s) for Attribute. The environment surrounding a bridge can have a significant effect on the rate of deterioration, particularly for corrosion. This attribute is intended to charac- terize the macro-environment surrounding a bridge and account for the likelihood of increased deterioration rates in environments that are particularly aggressive, such as coastal or marine environments. Aggressive environments typically have high ambient levels of chlorides, high ambient moisture levels (high humidity or frequent wet/dry cycles, increased temperature), and the presence of other harmful chemicals (i.e., high levels of carbon dioxide, sulphates, etc.).

82 Proposed Guideline for Reliability-Based Bridge Inspection Practices Assessment Procedure. The assessment procedure is similar to other environmental expo- sure classifications that are already in practice. Marine environments are deemed to be the most severe due to the high levels of ambient chlorides and moisture. “Moderate” environments are those in which corrosive agent levels (water and chlorides) are elevated but lower than those found in marine or other severe exposures. “Industrial” environments are less severe than marine but may contain other harmful chemicals. Under modern regulatory constraints, air- borne pollutant levels associated with industrial environments are minimized, and this should be considered in the assessment of industrialized environments. “Benign” environments are those in which application of de-icing chemicals is minimal or nonexistent; the environments may be arid and atmospheric pollutants typical. Severe/Marine 20 points Moderate/Industrial 10 points Benign 0 points L.4 Likelihood of Overload Reason(s) for Attribute. This attribute can be used when the likelihood of overload is a consideration for the bridge or a family of bridges being assessed. The likelihood of overload is used to characterize the chance that a bridge will be loaded beyond its inventory load rating. Such overloads generally increase the deterioration rate for structural elements. The probability of this occurring may be greater for bridges with a reduced capacity, such as those that have already been load posted. Assessment Procedure. This attribute is scored based on how likely it is that a bridge will be overloaded. Sound engineering judgment should be used to assess this attribute. High likelihood of overload 15 points Moderate likelihood of overload 10 points Low likelihood of overload 0 points L.5 Rate of De-icing Chemical Application Reason(s) for Attribute. This attribute is intended to characterize the volumes of the de-icing chemicals containing chloride ions that are being applied regularly to the surface of the deck. The detrimental effects of de-icing chemicals on the durability of bridge elements are well known. The intrusion of chloride ions to the level of the reinforcing steel provides an important driving force for corrosion of the reinforcing steel (15). When combined with oxygen and water, higher levels of de-icing chemical application generally lead to more rapid and severe reinforcement corrosion rates. The presence of increased chloride concentrations at the surface of the concrete increases chloride diffusion rates, shortening the time for the initiation of corrosion in the steel. If faulty deck joints or a substandard drainage system are present, which permit water seepage, bridge elements below the deck may also be affected by increased chloride ion levels. This will lead to increased levels of corrosion and consequently to corrosion-related damage. Assessment Procedure. This attribute can be scored based on the average annual number of applications of de-icing chemicals to the deck surface. The application rates may either be expressed quantitatively, if the bridge owner keeps such records, or on a qualitative scale. Factors that could be used to help estimate the rate of salt application include the ADT of the roadway and the amount of snowfall the bridge experiences. Typically, bridges with high ADT lie along

Attribute Index and Commentary 83 critical roadways that may receive the focus of local maintenance crews for the application of de- icing chemicals. Obviously, the more frequent the snowfall, the more often de-icing chemicals are likely to be applied. Users may have other data or information regarding the application of de-icing chemicals that can be used to develop rationales identifying those bridges exposed to high levels of de-icing chemicals and those where de-icing chemical use is minimal. High (more than 100 applications per year) 20 points Moderate 15 points Low (less than 15 applications per year) 10 points None 0 points L.6 Subjected to Overspray Reason(s) for Attribute. Overspray refers to the de-icing chemicals on a roadway that are being picked up and dispersed by traveling vehicles onto adjacent highway structures, including bridges and their substructures. Bridges that are located over roadways may receive overspray from the road below. Since overspray typically consists of salt or other de-icing chemicals, more exposure increases the likelihood of developing a corrosion problem. It is noted that L.6 Subjected to Overspray is explicitly considered to be a separate item from L.5 Rate of De-icing Chemical Application. This because some bridges may not have de-icing chemicals directly applied to their decks, but still can be exposed to overspray from below. An example of this would be a rural road over an interstate. However, to address the more severe condition where de-icing chemicals are applied to the bridge directly and by overspray, the items are considered separately. Assessment Procedure. Similar to the rate of de-icing chemical application, a quantitative estimate of overspray exposure may be difficult. The frequency of de-icing chemical application on the highway that the bridge crosses (if applicable) can be used to aid in estimating the over- spray exposure. The vertical clearance of the bridge is also a consideration. For example, a bridge with greater than 20 feet of vertical clearance over the roadway below may experience minimal effects from overspray. In any case, sound engineering judgment should be used. The suggested scoring scheme is based on the generally more significant effect of overspray on steel bridge elements. These suggested scales should be modified appropriately based on local experience. Concrete Bridge Deck, Prestressed Girder, Substructure Severe overspray exposure 15 points Moderate overspray exposure 7 points Low exposure overspray or not over a roadway 0 points Steel Girder Severe overspray exposure 20 points Moderate overspray exposure 10 points Low exposure overspray or not over a roadway 0 points L.7 Remaining Fatigue Life Reason(s) for Attribute. The remaining fatigue life of an element is somewhat related to the probability of a fatigue crack propagating to the point of brittle fracture. Obviously, for elements

84 Proposed Guideline for Reliability-Based Bridge Inspection Practices that have longer remaining fatigue lives, there is a lower probability of failure due to fatigue cracking than for elements with shorter remaining fatigue lives. Assessment Procedure. The remaining fatigue life of an element can be determined using any established method. Insufficient fatigue life refers to a fatigue life that is less than the required service life or some other interval defined by the owner (e.g., less than 10 years). It is noted that it is possible to calculate a life of less than the length of time the bridge has been in service (i.e., a “negative fatigue life”). In many cases, although a negative fatigue life has been calculated, there is no evidence of fatigue cracking on the structure. Although a negative fatigue life does not make physical sense, it does suggest that the probability of failure due to fatigue cracking is greater. In such cases, more in-depth evaluation efforts are justified, such as field testing or monitoring to obtain in-service stress range histograms or a more accurate finite element model of the struc- ture. Often, the more in-depth evaluations reveal that there is significant remaining fatigue life. Sufficient fatigue life refers to a fatigue life that exceeds the expected service life, or a defined life required by the owner (e.g., 10 years until replacement) of the element, but is not infinite. Infinite life is the case in which fatigue cracking is not expected to propagate during the life of the structure. It is noted that a greater penalty is placed on not having any knowledge of the remaining fatigue life than on having performed a fatigue analysis that determined a negative fatigue life. Unknown remaining fatigue life 10 points Insufficient remaining fatigue life 7 points Sufficient remaining fatigue life 3 points Infinite remaining fatigue life 0 points L.8 Overtopping/High Water Reason(s) for Attribute. Certain bridges are susceptible to periodic overtopping or high water condition in which the bridge superstructure is partially or totally immersed in water. Such condition may not adversely affect the loading carrying capacity of the structure; however, this condition may increase the likelihood that A) the structure is impacted by debris or ice in the water, or B) debris is deposited on the flanges and surrounding the bearing areas of the bridge. Impact from debris or ice in the water may increase the likelihood that a certain bridge suffers impact damage, even though the structure is not over a roadway. Debris deposited on the superstructure or at the bearing will retain moisture and may accelerate corrosion damage. Assessment Procedure. Bridges that are likely to be overtopped during periods of high water are typically documented in the NBIS data submitted annually to the FHWA. Experience may also be used to identify bridges susceptible to the adverse effects of high water. Scoring of this attribute may be different values for conditions A and B. Periodic overtopping/high water 20 points No overtopping/high water 0 points Condition Attributes C.1 Current Condition Rating Reason(s) for Attribute. The condition rating for a bridge component describes the exist- ing, in-place bridge as compared with the as-built condition. The condition ratings provide

Attribute Index and Commentary 85 an overall characterization of the general condition of the entire component. It is reason- able to assume that a given element that has already shown signs of damage is more likely to deteriorate to a serious condition than an element showing little or no signs of damage. It is typical for a concrete component with a condition rating of 5 or less to have observ- able corrosion damage in the form of cracking or spalling (either as open spalls or patched spalls). Such damage provides pathways for the increased penetration of chlorides ions and for increased rates of damage accumulation. For steel elements, low condition ratings are frequently emblematic of significant corrosion damage. Fatigue cracking or member distor- tions due to unexpected settlement, etc. may be present. Conversely, components with a high condition rating (6 or above) typically have lower levels of existing deterioration. Conse- quently, some consideration should be given to the overall component rating when assessing the durability of the bridge element. Assessment Procedure. For this attribute, a condition rating of 5 or less is considered to have a much higher likelihood for accelerated damage than component with higher condi- tion ratings. A condition rating of 6 is considered to have a smaller likelihood of accelerated damage. Condition rating is 5 or less 20 points Condition rating is 6 5 points Condition rating is 7 or greater 0 points C.2 Current Element Condition State Reason(s) for Attribute. When element-level inspections are conducted under the AASHTO Bridge Element Inspection Manual, element condition states (CS) that are linked to specific evi- dence of damage or deterioration to the subject bridge element are defined. Elements or por- tions of elements in CS 1 typically have very little or no evidence of deterioration. Elements or portions of elements in CS 2 have some evidence of damage. As such, it is reasonable to assume that if a given element is entirely in CS 1, the likelihood of severe damage occurring in the near future is lower than an element with portions of the element in CS 2, 3, or 4. This attribute is intended to consider the positive attributes of an element in CS 1. Assessment Procedure. For this attribute, the current CS for a given bridge element is con- sidered. For elements entirely in CS 1, the scoring of 0 points is suggested, for elements where CS 3 is indicated for any portion of the element, a score of 20 points is suggested. Users may wish to utilize appropriate gradations for elements with conditions indicated as CS 2. The severity and the significance of CS 2 vary by element, and the RAP may wish to develop alternative scoring schemes based on specific elements and CS apportionment. Element-level inspection imple- mentation varies at the owner level, and therefore appropriate scoring should be considered by the RAP according to existing inspection practices. CS 2 is indicated for a significant portion of the element, or CS 3 is indicated for any portion of the element 20 points Condition State 2 is indicated for a minor portion of the element 10 points Condition State 1 is indicated for entire element 0 points

86 Proposed Guideline for Reliability-Based Bridge Inspection Practices C.3 Evidence of Rotation or Settlement Reason(s) for Attribute. This attribute is intended to consider the effects of unexpected rotation or settlement of abutments and piers. Use of this attribute is for minor settlements or rotations that do not affect the structural capacity, but may result in atypical or accelerated deterioration patterns. Significant rotations or settlements may require engineering analysis. The rotation of a bridge substructure beyond its design tolerances may result in damage that is manifested by cracking, skewing, and/or misaligned bridge components. Unexpected settle- ments may result in cracking that provides pathways for intrusion of water and chlorides, lead- ing to accelerated corrosion of reinforcing steel. Assessment Procedure. Evidence of rotation or settlement should be rated based on their severity using engineering judgment. Rotation or settlement resulting in cracking of concrete, misaligned joints, or misaligned members 15 points Minor evidence of rotation or settlement with the potential to result in unexpected cracking or poor joint performance 5 points No evidence of rotation 0 points C.4 Joint Condition Reason(s) for Attribute. The presence of one or more leaking joints will dramatically increase the possibility for corrosion related deterioration on the elements below the deck. This is because joints that are leaking will usually leak chloride-contaminated water directly onto other bridge components such as the superstructure, substructure, and bearing areas. This allows corrosion to initiate and propagate at a faster rate in the affected elements. Assessment Procedure. This attribute should be rated based on either visual observation or on information contained in bridge inspection reports. For this attribute, the presence of a leaking joint is considered to be severe. If a joint has become debris filled, there is an increased probability that that joint will become damaged and start to leak in the near future. Users should consider historical experience with typical joints in their inventory in evaluating this attribute. For example, if certain typical joint types are expected to have a service life of less than 5 years, it may be appropriate to assume that this joint is a leaking joint, because even if it is not leaking currently, it is expected to leak in near future. Open joints should be expected to allow for the passage of water and debris, and thus should be scored accordingly if this effect is unmitigated. For bridges that are jointless, it is assumed that the bridge is performing as intended and deck drainage is not affecting the bearing areas. Significant amount of leakage at joints 20 points Joints have moderate leakage or are debris filled 15 points Joints are present but not leaking 5 points Bridge is jointless 0 points C.5 Maintenance Cycle Reason(s) for Attribute. This attribute is intended to consider the positive benefits of consistent maintenance and preservation activities on the durability and the reliability of bridge elements. Activities such as deck cleaning, maintenance of drainage, debris removal,

Attribute Index and Commentary 87 washing out joints, and periodic application of the sealers help preserve bridge elements and extend their service lives. Conversely, a bridge that does not receive periodic mainte- nance and preservation activities is likely to experience damage and deterioration much earlier in its service life, and deteriorate at a higher rate relative to a bridge receiving consistent, periodic maintenance. Assessment Procedure. This attribute is scored based on the bridge maintenance policies and practices within the particular inventory being assessed. The RAP panel should consider the policies and practices within its state with regard to the intensity of maintenance activi- ties within particular regions, districts, or municipalities. For example, state-owned bridges typically receive more consistent and thorough maintenance than locally-owned bridges. Bridges located in rural areas may receive less intense maintenance than those located near population centers, etc. The RAP should consider specific situations within its bridge inven- tory when assessing this attribute, and develop criteria for establishing which bridges receive regular maintenance that can be expected to prevent deterioration, and those bridges which do not. Bridge does not receive routine maintenance 20 points Some limited maintenance activities 10 points Bridge is regularly maintained 0 points C.6 Previously Impacted Reason(s) for Attribute. If a bridge has been previously struck or impacted by a vehicle, it is reasonable to assume that there is an increased probability of further impact damage. The element could also have been damaged as a result of previous impact, which has been shown to decrease, for example, a steel girder’s resistance to brittle fracture (16). For concrete bridge ele- ments, impacts can compromise the concrete cover, resulting in the exposure of embedded steel elements. The occurrence of previous impacts should be considered in the analysis for potential impact damage. Assessment Procedure. This attribute is scored based only on whether or not the bridge has been previously impacted. If the impact risks have been mitigated, this should be considered in the analysis. Bridge has been previously impacted 20 points Bridge has not been previously impacted 0 points C.7 Quality of Deck Drainage System Reason(s) for Attribute. The purpose of the deck drainage system is to get water, de-icing chemicals, and debris off of the bridge deck effectively, without draining directly onto other ele- ments of the bridge, such as the superstructure and the substructure elements. This attribute is intended to address leakage or deck drainage onto other bridge elements as a result of damage, deterioration, or the ineffective performance of a deck drainage system. Deck drainage systems with ineffective designs would typically be address using attribute D.4 Poor Deck Drainage and Ponding. Assessment Procedure. This attribute is based on the performance of the drainage system in place on the bridge deck. Since estimating the quality of the drainage system is subjective, it should be based on experience, engineering judgment, and common sense. Some key factors to

88 Proposed Guideline for Reliability-Based Bridge Inspection Practices consider when scoring this attribute include build-up at the deck inlet grates, clogged drains or pipes, section loss in pipes, etc. Deck drains directly onto superstructure or substructure components, or ponding on deck results from poor drainage 20 points Drainage issues resulting in drainage onto superstructure or substructure components, or moderate ponding on deck; effects may be localized 10 points Adequate quality 0 points C.8 Corrosion-Induced Cracking Reason(s) for Attribute. This attribute considers the presence of corrosion-induced crack- ing in concrete bridge elements. Corrosion-induced cracking typically occurs due to the expan- sion of reinforcing steel caused by the development of corrosion by-products on the surface of the bar. This expansion leads to cracking of the concrete, providing pathways for water and chlo- rides to penetrate to the reinforcement level. Frequently, this type of cracking is accompanied by rust staining. Such evidence of active corrosion would typically be detected during a typical visual inspection of a bridge. The presence of active corrosion increases the likelihood for cor- rosion damage to occur to a severe extent in the future. Assessment Procedure. This attribute is scored based on the presence and the severity of corrosion-induced cracking in concrete bridge elements. The determination of the significance of the cracking should be based on engineering judgment. Significant corrosion-induced cracking 20 points Moderate corrosion-induced cracking 10 points Minor corrosion-induced cracking 5 points No corrosion-induced cracking 0 points C.9 General Cracking Reason(s) for Assessment. This attribute is used to characterize the presence non- structural cracks in concrete. These cracks may result from shrinkage, thermal forces, or other non-structural effects. These cracks can provide pathways for the intrusion of chlorides to the level of the reinforcement. It is generally recognized that cracks perpendicular to the reinforcing bars hasten the corrosion of the intersected reinforcement by facilitating the ingress of moisture, oxygen, and chloride ions. Cracks that follow the line of a reinforcing bar are much more serious, since the length of the bar equal to the length of the crack is exposed to corrosive elements. The presence of cracking also reduces the concrete’s ability to contain spalling as the reinforcement corrodes. This attribute is generally used for cracking other than corrosion-induced cracking, which is described in attribute C.8. Assessment Procedure. The rating of this attribute depends on engineering judgment. More specific guidance to classifying crack sizes and density can be found in the 2010 edition of the AASHTO Bridge Element Inspection Manual. Widespread or severe cracking 15 points Moderate cracking present 10 points Minor or no cracking present 0 points

Attribute Index and Commentary 89 C.10 Delaminations Reason(s) for Attribute. Delaminations are subsurface cracks in concrete generally parallel to the concrete surface. Delaminations are caused by the formation of horizontal cracking as a result of volumetric expansion of the reinforcing steel during the corrosion process. Delamina- tions are typically emblematic of the corrosion of embedded steel, and thus provide an early indicator of where future spalling is likely to occur. This attribute is intended to consider that concrete elements with delaminations are more likely to experience deterioration and damage in the future, relative to elements in which delaminations are not present. The detection of delami- nations in concrete can reduce the uncertainty in determining if there is active corrosion that is manifesting in damage to the concrete. This attribute may also be used to characterize conditions for a deck overlay. Under these conditions, delaminations are indicative of a loss of bond between the overlay and the substrate. Overlays that are debonding are likely to deteriorate more rapidly than an overlay with good bonding characteristics. It is implied that some form of NDE has been conducted to address this attribute, as delami- nations are not visibly detectable. This typically includes hammer sounding or chain drag, but may include other techniques such as infrared thermography, impact echo, or other methods. Assessment Procedure. This attribute is scored based on inspection results that indicate the extent of delaminations present in a given concrete element. This attribute should be scored based on the amount of surface area of the structure that includes delaminations. Suggested values for the significant levels of delamination are indicated below. Significant amount of delaminations present (greater than 20% by area) or unknown 20 points Moderate amount of delaminations present (5% to 20% by area) 10 points Minor, localized delaminations (less than 5% by area) 5 points No delaminations present 0 points C.11 Presence of Repaired Areas Reason(s) for Attribute. Repaired spalls and patches are a way to temporarily seal reinforce- ment exposed as a result of damaged concrete. However, even though the reinforcement is again sealed from the environment, the existing corrosion can continue to propagate. Patches frequently have a relatively short service life, especially when traffic loading is high. The service life of deck patches ranges from 4 years to 10 years (17), although an FHWA TechBrief indicates that the service life of a patch ranges from 4 years to only 7 years (18). The service life of the patch depends largely on the corrosivity of the surrounding concrete and the development of the halo effect. When concrete is contaminated with chlorides in concentrations greater than the threshold level in the area surrounding the patches, inadvertent acceleration of the corrosion rate can occur. The patched area acts as a large non-corroding site (i.e., cathodic area) adjacent to corroding sites (i.e., anodic areas), and thus corrosion cells are created. Assessment Procedure. The presence of repaired areas should be scored based on the total surface area of the bridge that has repaired areas. Engineering judgment should be exercised. If the repaired areas result from impact damage or other non-corrosion–related damage, and chlo- rides levels for the intact concrete are expected to be nominal, a reduced score may be assigned.

90 Proposed Guideline for Reliability-Based Bridge Inspection Practices Significant amount of repaired areas 15 points Moderate amount of repaired areas 10 points Minor amount of repaired areas 5 points No repaired areas 0 points C.12 Presence of Spalling Reason(s) for Attribute. This attribute is intended to consider the presence of spalling on concrete bridge elements. Open spalls are sections of concrete that have separated from the larger mass of concrete and fallen off of the structure, usually exposing the underlying reinforce- ment. Unrepaired spalling allows corrosive elements to directly contact the exposed reinforce- ment and prestressing steel, if present. This will lead to accelerated rates of corrosion damage in the area surrounding the spall. Users may wish to include repaired spalls under this attribute, or utilize the attribute C.11 Presence of Repaired Areas. Assessment Procedure. This attribute is scored based on the severity and the extent of spalling as reported in bridge inspection reports. Users should consider the importance of the spalling in terms of the structural performance of the element under consideration in devel- oping their scoring methodology. Spalling that leads to the exposure of prestressing strands is considered significantly more important than spalling in a reinforced element exposing the mild steel bars. Significant spalling (greater than 10% of area with spalling, rebar or strands exposed) 20 points Moderate spalling (greater than 1 inch deep or 6 inches in diameter or exposed reinforcement) 15 points Minor spalling (less than 1 inch deep or 6 inches in diameter) 5 points No spalling present 0 points C.13 Efflorescence/Staining Reason(s) for Attribute. This attribute is intended to consider the increased likelihood of corrosion damage associated with the presence of efflorescence on the surface of concrete elements. Efflorescence is a white stain on the face of a concrete component which results from the crystallization of dissolved salts. While efflorescence is typically considered an aes- thetic problem, it may be indicative of a problem with the concrete mix and may contribute to corrosion initiation. Efflorescence on the soffit of a bridge deck typically indicates that water is passing freely through the deck, likely carrying with it chlorides that may cause cor- rosion of the reinforcing steel. When rust stains are present, the corrosion of reinforcing steel is assured. Extensive leaching causes an increase in the porosity and the permeability of the concrete, thus lowering the strength of the concrete and making it more vulnerable to hostile environ- ments (e.g., water saturation and frost damage, or chloride penetration and the corrosion of embedded steel). Those concretes that are produced using a low water-cement ratio, adequate cement content, proper compaction, and curing are the most resistant to leaching that results in efflorescence on the surface of the concrete (19).

Attribute Index and Commentary 91 Assessment Procedure. This attribute is scored based on inspection results. The scoring for this attribute is based on the existence of efflorescence stains and whether or not rust stains have also been deposited from corroding reinforcement. Moderate to severe efflorescence with rust staining; severe efflorescence without rust staining 20 points Moderate efflorescence without rust staining 10 points Minor efflorescence 5 points No efflorescence 0 points C.14 Flexural Cracking Reason(s) for Attribute. When the primary load-bearing members in a concrete bridge exhibit flexural cracking, it may indicate that the members were either inadequately designed for the required loading, that overloads have occurred, or that deterioration has occurred that has reduced the load-bearing capacity of the members. In any case, large flexural cracks can be indicative of an inadequate load-bearing capacity that may require an engineering analysis in order to determine the cause of the cracking and the resulting effect on the load capacity of the structure. As a result, bridges exhibiting moderate to severe flexural cracking should be screened from the general reliability assessment unless appropriate engineering analysis indicates that the cracking is benign. Flexural cracking in a prestressed element is generally more significant than in a reinforced concrete element. In cases where flexural cracking is minor or appropriate assessment has indicated that the cracking is not affecting the adequate load capacity of the element, the cracking nonetheless may provide pathways for the ingress of moisture and chlorides that may cause corrosion of the embedded steel. This attribute is intended to consider the increased likelihood of corrosion resulting from the cracking in the concrete. Assessment Procedure. Flexural cracks will typically present themselves with a vertical ori- entation either near the bottom flange at mid-span or near the top flange over intermediate supports, if the member is continuous. Engineering judgment must be exercised in determining whether any present flexural crack- ing is moderate to severe. Crack widths in reinforced concrete bridges exceeding 0.006 inches to 0.012 inches reflect the lower bound of “moderate cracking.” The American Concrete Insti- tute Committee Report 224R-01 (1) presents guidance for what could be considered reasonable or tolerable crack widths at the tensile face of reinforced concrete structures for typical condi- tions. These range from 0.006 inches for marine or seawater spray environments to 0.007 inches for structures exposed to de-icing chemicals, to 0.012 inches for structures in a humid, moist environment. In prestressed concrete bridge structural elements, tolerable crack width criteria have been adopted in the PCI MNL-37-06 Manual for the Evaluation and Repair of Precast Pre- stressed Concrete Bridge Products (20). The PCI Bridge Committee recommends that flexural cracks greater in width than 0.006 inches should be evaluated to affirm adequate design and performance. Note that this attribute is a companion to the screening attribute S.4 Flexural Cracking, in which any moderate to severe flexural cracking should exclude the bridge from a risk-based assessment unless appropriate engineering analysis has been completed showing that the crack- ing is benign or has been repaired. Generally, cracking in prestressed elements is more problem- atic than cracking in reinforced concrete elements.

92 Proposed Guideline for Reliability-Based Bridge Inspection Practices Crack widths equal to or less than 0.006 inches to 0.012 inches, depending on environment for reinforced concrete; crack widths equal to or less than 0.006 inches for prestressed concrete 10 points No flexural cracking 0 points C.15 Shear Cracking Reason(s) for Attribute. Similar to flexural cracking, if the primary load-bearing members in a concrete bridge exhibit shear cracking, it can be assumed that the members were either inadequately designed for the required loading or that deterioration has occurred, which has reduced the load-bearing capacity of the members. In either case, large shear cracks can be indicative of an inadequate load-bearing capacity, which may require an engineering analysis in order to determine the cause of the cracking and the resulting effect on the load capacity. As a result, bridges exhibiting moderate to severe shear cracking should be screened from the reli- ability assessment unless appropriate engineering analysis indicates that the cracking is benign in terms of the load-bearing capacity. Assessment Procedure. Engineering judgment must be exercised in determining the sever- ity of any present shear cracking. Shear cracks will typically present themselves with a roughly 45 degree diagonal orientation and will radiate towards the mid-span of the member for conven- tionally reinforced concrete. For prestressed concrete, angles down to roughly 30 degrees may be observed. The ends of the member and any sections located over piers should be checked for this type of cracking. Note that this attribute is a companion to the screening attribute S.5 Shear Cracking, where any moderate to severe flexural cracking should exclude the bridge from a risk- based assessment until adequate assessments have been conducted. Minor, hairline to less than 0.0625 inch shear cracking 10 points No shear cracking 0 points C.16 Longitudinal Cracking in Prestressed Elements Reason(s) for Attribute. This attribute is for the assessment of prestressed concrete bridge elements. Longitudinal cracking in prestressed elements can be indicative of the corrosion or the fracture of the embedded prestressing strands. As a result, elements with reported longitudinal cracking in the soffit, web, or flange should be individually assessed to determine the source of the cracking and to assess the condition of the prestressing strands (21). Assessment Procedure. Longitudinal cracking in prestressed elements can be indicative of strand corrosion and damage, and, as such, significant longitudinal cracking is a screening attri- bute. The use of longitudinal cracking in prestressed elements as a condition attribute assumes the cracking in question is minor in nature, and significant strand corrosion is not currently present. In this case, the longitudinal cracking provides pathways for the intrusion of moisture and chlorides to the prestressing strands and the mild steel bars. As a result, a prestressed element with minor longitudinal cracking is more likely to experience deterioration and damage than an uncracked element. This attribute is scored based on inspection results. Minor longitudinal cracking in beam soffit 15 points No longitudinal cracking in beam soffit 0 points

Attribute Index and Commentary 93 C.17 Coating Condition Reason(s) for Attribute. This attribute considers the effect of the coating condition on the likelihood of corrosion damage occurring in steel bridge elements. Coatings are applied to steel elements to provide protection from corrosion and for aesthetic reasons. Elements with coatings in good condition, and performing as intended, are generally less susceptible to corrosion damage. Elements with significant rusting and corrosion in areas in which that paint system has failed are more likely to experience further corrosion damage in the future. Assessment Procedure. Depending on the condition of the coating, the likelihood of corro- sion damage varies. Coatings typically deteriorate more rapidly where drainage from the bridge deck is allowed to flow onto the steel surface. As a result, conditions for the accelerated corrosion of steel may already exist. If the coating is already in poor condition, the likelihood of severe corrosion damage is greater than for a coating in good condition. If the element is constructed with weathering steel (assuming it is placed in the proper environment and is detailed correctly), it should be scored as though the coating is in good condition. The development of an effective patina for the weathering steel should be confirmed. Coating system in very poor condition, limited or no effectiveness for corrosion protection, greater than 3% rusting 10 points Coating system is in poor condition, 1% to 3% rusting, substantially effective for corrosion protection 5 points Coating is in fair to good condition, effective for corrosion protection 0 points C.18 Condition of Fatigue Cracks Reason(s) for Attribute. Active fatigue cracks due to primary stress ranges will continue to grow until the failure of the member, either by brittle or by ductile fracture. An arrested or repaired fatigue crack is better than having an active crack, but it is still worse than having no crack at all, as it suggests that the conditions necessary for cracking to initiate were or still may be present in the structure. In other words, other similar details (that have not been preemptively retrofitted) may be susceptible to cracking in the future. Assessment Procedure. To determine whether or not a fatigue crack is arrested, a comparison must be made between previous inspection reports. In order to be considered arrested, a crack must have not grown in a specified amount of time (e.g., the inspection interval plus one year). It is noted that although no fatigue cracks may have been observed, a detail still may be highly sus- ceptible to fatigue. Hence, other attributes such as D.16 Element Connection Type, D.17 Worst Fatigue Detail Category, and L.1 ADTT are included in the assessment procedure to address the susceptibility to cracking. Fatigue crack exists and is active/unknown 20 points (see S.7) Fatigue crack exists and has arrested or been retrofitted 10 points No fatigue cracks are present 0 points

94 Proposed Guideline for Reliability-Based Bridge Inspection Practices C.19 Presence of Fatigue Cracks Due to Secondary or Out-of-Plane Stress Reason(s) for Attribute. Fatigue cracks due to secondary or out-of-plane stresses are the most common type of fatigue cracks found on highway bridges. Most of these cracks occur due to incompatibility or relative movement between bridge components. Assessment Procedure. The scoring for this attribute is based on the existence or non- existence of fatigue cracks. Some common types of fatigue cracks due to secondary stresses include web-gap cracks, deck plate cracking in orthotropic bridge decks, and floor beam connections. Fatigue cracks are present and are active/unknown 15 points Fatigue cracks are present but have been arrested or have been retrofitted 5 points No fatigue cracks are present 0 points C.20 Non-Fatigue-Related Cracks or Defects Reason(s) for Attribute. This attribute refers to steel bridge elements that may be suscep- tible to fatigue-induced cracking. Fatigue cracks generally start from some initial crack or defect. As a result of this, fatigue and brittle fracture is less likely if there are no cracks or defects from which cracks can propagate. Assessment Procedure. This attribute should be scored based on whether or not cracks or other defects are found in the element. Previous inspection reports should be used when evalu- ating this attribute. Non-fatigue-related cracks or defects are present 10 points Non-fatigue-related cracks or defects are not present 0 points C.21 Presence of Active Corrosion Reason(s) for Attribute. The presence of visible active corrosion on steel bridge elements indicates that severe corrosion damage in the future is possible, since the environment and the bridge features are vulnerable to the initiation and the propagation of corrosion. It is also well known that corrosion damage typically propagates at an accelerated rate, once initiated, and that elements that show no signs of active corrosion are very unlikely to develop severe corrosion damage during the assessment interval of 72 months. Maximum rates of section loss under the most severe marine conditions typically do not exceed 10 mils/year (0.010 inches/ year). For moderate conditions, rates are typically on the order of 4 mils/year (0.004 inches/ year) or less. Corrosion damage that is inactive is explicitly distinguished from corrosion that is active. For example, section loss on a girder web that was the result of a leaking expansion joint that was corrected (the joint was replaced and the girder was repainted), may be assumed to have inactive corrosion. It is assumed that the owner has determined that the existing section loss is either insignificant or has taken it into account in the rating procedures and that load posting, if needed, is in place. Assessment Procedure. This attribute should be scored based on the amount of active cor- rosion present on the element. Engineering judgment should be used in determining whether

Attribute Index and Commentary 95 or not the corrosion is active. This attribute may also be used as a screening tool in a reliability assessment. Significant amount of active corrosion present 20 points Moderate amount of active corrosion present 15 points Minor amount of active corrosion present 7 points No active corrosion present 0 points C.22 Presence of Debris Reason(s) for Attribute. The presence of debris on bridge elements can substantially increase the probability of corrosion damage by maintaining a moisture-rich environment on the surface of the steel. Debris can be especially damaging if it is allowed to remain on the bridge without maintenance action, such as washing or cleaning. This attribute is intended to charac- terize bridges susceptible to having debris deposited on the flanges, bearings, connections, or other details that results in atypical (e.g., accelerated) deterioration patterns. Assessment Procedure. This attribute should be assessed based on if debris is present or likely to be present on the element, resulting in an atypical deterioration pattern. Debris is or is likely to be present 15 points Debris not likely to be present 0 points References 1. ACI Committee 224, ACI 224R-01: Control of Cracking in Concrete Structures. 2001, American Concrete Institute: Farmington Hill, Michigan. 2. Fisher, J. W., and Lichtenstein, A., Hoan Bridge Forensic Investigation Failure Analysis Final Report. 2001, Wisconsin Department of Transportation: Madison, WI. 3. FHWA, Recording and Coding Guide for the Structure Inventory and Appraisal of the Nation’s Bridges. 1995, Federal Highway Administration: Washington, D.C. 4. Skeet, J., G. Kriviak, and M. Chichak, Service Life Prediction of Protective System for Concrete Bridge Decks in Alberta. 1994, Edmonton, Alberta, Canada: Alberta Transportation and Utilities, Research & Development. 5. Russell, H. G., NCHRP Synthesis 333: Concrete Bridge Deck Performance. 2004, Transportation Research Board of the National Academies, Washington, D.C. 6. Clemena, G. G. and Y. P. Virmani, Corrosion Protection: Concrete Bridges. 1998, McLean, VA: U.S. Dept. of Transportation, Federal Highway Administration, Research and Development, Turner-Fairbank Highway Research Center. 7. FHWA, Uncoated Weathering Steel in Structures, in Federal Highway Administration Technical Advisory T-5140.22. 1989, FHWA: Washington, D.C. 8. Huang, H., Shenton, H. W., and Chajes, M. J. Load distribution for a highly skewed bridge: Testing and analysis. Journal of Bridge Engineering, 2004, 9(6), 558–562: ASCE, Reston, VA. 9. Coletti, D., B. Chavel, and W. J. Gatti, Challenges of Skew in Bridges with Steel Girders. Transportation Research Record: Journal of the Transportation Research Board, No. 2251 2011: Transportation Research Board of the National Academies, Washington, D.C., pp. 47–56. 10. Fu, G., Feng, J., Dimaria, J., and Zhuang, Y., Bridge Deck Corner Cracking on Skewed Structures, 2007. MDOT Report RC 1490. 11. Menassa, C., Mabsout, M., Tarhini, K., and Frederick, G. Influence of Skew Angle on Reinforced Concrete Slab Bridges. Journal of Bridge Engineering, 2007. 12(2), pp. 205–214: ASCE, Reston, VA. 12. Tindal, T. T. and Yoo, C. H. Thermal Effects on Skewed Steel Highway Bridges and Bearing Orientation. Journal of Bridge Engineering, 2003, 8(2), p. 57–65: ASCE, Reston, VA. 13. Connor, R. J. and M. J. Parr, A Method for Determining the Interval for Hands-On Inspection of Steel Bridges with Fracture Critical Members. 2008: Purdue University. p. 32.

96 Proposed Guideline for Reliability-Based Bridge Inspection Practices 14. McLean, D. I., et al., NCHRP Synthesis 266: Dynamic Impact Factors for Bridges,. 1998, TRB, National Research Council: Washington, D.C. 15. Silano, L. G. and P. Brinckerhoff, Bridge Inspection and Rehabilitation: A Practical Guide. 1993, New York, NY: John Wiley & Sons, Inc. 16. Connor, R. J., M. R. Urban, and E. J. Kaufmann, NCHRP Report 604: Heat Straightening Repair of Dam- aged Steel Bridge Girders—Fatigue and Fracture Performance,. 2008, Transportation Research Board of the National Academies, Washington, D.C. 17. Weyers, R. E., et al., SHRP-S-360: Bridge Protection, Repair, and Rehabilitation Relative to Reinforcement Corrosion: A Methods Application Manual, Strategic Highway Research Program, report. 1993, Transportation Research Board of the National Academies: Washington, D.C. 18. FHWA, FHWA-RD-99-177: Portland Cement Concrete (PCC) Partial-Depth Spall Repair, 1999, Federal Highway Administration: McLean, VA. 19. Oak Ridge, N. L., Primer on Durability of Nuclear Power Plant Reinforced Concrete Structures—A Review of Pertinent Factors. 2006, U.S. Nuclear Regulatory Commission. p. 114. 20. PCI, Manual for the Evaluation and Repair of Precast, Prestressed Concrete Bridge Products: Including Imper- fections or Damage Occurring During Production, Handling, Transportation, and Erection. 2006: Chicago, IL. 21. Naito, C., Sause, R., Hodgson, I., Pessiki, S., and Macioce, T., Forensic Examination of a Noncomposite Adja- cent Precast Prestressed Concrete Box Beam Bridge. Journal of Bridge Engineering, 2010, 15(4), p. 408–418: ASCE, Reston, VA.

97 98 F 1 Introduction 98 F 2 Example 1: Prestressed Concrete Bridge 98 F 2.1 Bridge Profile 98 F 2.1.1 Overview 99 F 2.1.2 Concrete Bridge Deck 99 F 2.1.3 Prestressed Girders 99 F 2.1.4 Substructure 99 F 2.2 Assessment 100 F 2.2.1 Concrete Bridge Deck 102 F 2.2.2 Prestressed Girder 105 F 2.2.3 Substructure 106 F 2.3 Consequence Assessment 107 F 2.4 Scoring Summary 107 F 2.5 Criteria for a Family of Bridges 109 F 3 Example 2: Steel Girder Bridge 109 F 3.1 Bridge Profile 109 F 3.1.1 Overview 109 F 3.1.2 Concrete Bridge Deck 110 F 3.1.3 Steel Girders 110 F 3.1.4 Substructure 110 F 3.2 Assessment 110 F 3.2.1 Concrete Bridge Deck 112 F 3.2.2 Asphalt Overlay 112 F 3.2.3 Steel Girders 114 F 3.2.4 Substructure 115 F 3.3 Consequence Assessment 116 F 3.4 Scoring Summary 117 F 3.5 Inspection Data 117 F 4 Example 3: Reinforced Concrete Bridge 117 F 4.1 Bridge Profile 117 F 4.1.1 Overview 117 F 4.1.2 Concrete Bridge Deck 118 F 4.1.3 Reinforced Concrete Girders 118 F 4.1.4 Substructure 119 F 4.2 Assessment 119 F 4.2.1 Concrete Bridge Deck 121 F 4.2.2 Reinforced Concrete Girders 123 F 4.2.3 Substructure 124 F 4.3 Consequence 126 F 4.4 Scoring Summary 126 F 4.5 Inspection Data A P P E N D I X F Illustrative Examples

98 Proposed Guideline for Reliability-Based Bridge Inspection Practices F 1 Introduction This section provides three illustrative examples of applying reliability-based analysis to establish an inspection interval and strategy. The first is an example of a bridge constructed with a superstructure composed of prestressed girders, the second example is a bridge with a multi-girder steel superstructure, and the third example is a multi-girder reinforced concrete superstructure. The RAP assembled by a bridge owner would typically conduct this analysis. For these examples, typical attributes that could be identified by a RAP have been selected for illustrative purposes. Attribute scoring sheets are shown to illustrate the process of applying a numerical scoring process for identified attributes to estimate the reliability of bridge elements, and to develop rationale for determining the appropriate inspection interval. In the examples shown, Occurrence Factor (OF) categories were determined by applying the following equation: 4X S S i o ∑ ∑= ∗ Where Si is the score recorded for each attribute and So is the maximum score for each attribute, such that the ratio S Si o∑ ∑ is a value between 0 and 1. OFs were then applied such that values of X between 0 and <1 were identified as “Remote,” values 1 or greater but less than 2 “Low,” etc. This provides a simple methodology for ranking bridges according to their important attributes that contribute to the durability and reliability of the bridge, and estimating the appropriate OF. This scoring methodology should be calibrated by the RAP for its specific bridge inventory to ensure results are consistent with sound engineering judgment. The examples also describe the Consequence Factors that were selected for each bridge, along with the rationale for selection. Based on these results, an appropriate inspection interval is identified for each bridge based on the risk matrix (Figure C1). The IPN for each damage mode is also calculated to illustrate how the process prioritizes damage modes to support inspection procedures for that bridge. F 2 Example 1: Prestressed Concrete Bridge F 2.1 Bridge Profile F 2.1.1 Overview This example bridge is constructed of prestressed girders with a composite concrete deck (Figure F1). The bridge has a typical reinforced concrete deck, seven prestressed AASHTO Type IV girders, and a reinforced concrete substructure. The bridge was constructed in 2006. Epoxy-coated reinforcement has been used in the deck and in parts of the prestressed girders. The substructure contains regular, uncoated reinforcement. The rate of de-icing chemical application is moderate, and the environment is also moderate. The reported ADTT is 210 vehicles. An element-level inspection had been conducted on the bridge, and data from the element-level inspection including inspector notes were used in deter- mining values for the condition attributes. All elements were rated 100% in Condition State (CS) 1.

Illustrative Examples 99 F 2.1.2 Concrete Bridge Deck The deck for this structure was cast-in-place and constructed with normal concrete and epoxy-coated rebar. From the design plans, the concrete cover for the top of the deck is 1-½ inches. Asphaltic plug joints in the deck are in good condition. Some transverse cracks, spaced 2 to 3 feet apart, have been noted on the underside of the deck. Efflorescence is present near these cracks, though there is no rust staining. No other damage has been observed. The current condition rating is 7-Good Condition, based on the most recent inspection. F 2.1.3 Prestressed Girders The superstructure of this bridge consists of 7 AASHTO Type IV prestressed concrete girders. There is at least 2 inches of clear cover for all surfaces as determined from the design plans, and the mild reinforcing is epoxy coated. No sealers or coatings have been applied to the girders. The maximum span length is 99 feet. The superstructure has no observed spalling or cracking and was most recently rated as being condition 8-Very Good Condition. F 2.1.4 Substructure The substructure was constructed of normal concrete with uncoated carbon steel reinforce- ment. The minimum design cover was determined to be 2 inches. Water from the deck does not contact the substructure either through the drainage system or through the joints. There are no observed signs of cracking or spalling. No evidence of unusual rotation or settlement has been noted, and the bridge is founded on rock. The substructure is rated to have a condition rating of 8-Very Good Condition based on the most recent inspection report. F 2.2 Assessment This section will show how the methodology is applied to determine the OFs, the Consequence Factors, and the corresponding inspection intervals for this bridge. A detailed scoring of each damage mode will be presented with written descriptions of how the Figure F1. Elevation view of Example Bridge 1.

100 Proposed Guideline for Reliability-Based Bridge Inspection Practices consequence of damage was considered. The results are then summarized in a table that provides the maximum inspection interval based on the risk matrix and the IPN determined from the analysis. The primary elements of this bridge are a concrete bridge deck, prestressed concrete girders, piers, and abutments. For the concrete bridge deck element, the RAP identified typical damage modes of widespread corrosion-induced cracking and spalling. Since each of these damage mode results from the effects of corrosion, these damage modes were combined into a single damage mode named “Corrosion Damage.” For the prestressed concrete girders, the RAP identified the following damage modes: • Bearing Area Damage, • Corrosion Between Beam Ends, • Flexural and Shear Cracking, and • Strand Fracture. For the substructure, the damage mode considered was: • Corrosion Damage (cracking and spalling due to the effects of corrosion). Considering the damage modes identified for each element, attributes relating to each damage mode were identified and ranked, as described in the Guideline. The following sec- tions contain illustrative examples of attribute scoring sheets developed for the different elements and damage modes for the bridge and the estimated OFs based on the attribute scoring. F 2.2.1 Concrete Bridge Deck The RAP determined that certain attributes of a bridge deck that contribute to the likelihood of corrosion damage are common and well known, and that these same attributes would generally apply to other bridge decks in its inventory, as well as other typical concrete elements. Addition- ally, because corrosion will affect most concrete elements and associated damage modes, repeti- tion of certain common attributes could be reduced by having a single corrosion profile for an element. This corrosion profile could then be applied to all damage modes stemming from cor- rosion for a given element more efficiently. As such, a corrosion profile was developed to assess the corrosion-resistance characteristics of a concrete bridge deck or other concrete element. This profile included typical attributes that were well known to affect the durability of concrete, but did not depend on the current condition or individual characteristics of an element. The attributes identified included: • Poor Deck Drainage and Ponding, • Years of Construction, • Application of Protective Systems, • Concrete Mix Design, • Minimum Concrete Cover, • Reinforcement Type, • Exposure Environment, • Rate of De-icing Chemical Application, and • Maintenance Cycle. Supporting rationale for each of these attributes from the commentary (Appendix E) was used. Utilizing these corrosion profile attributes and the suggested rankings in the commentary, the RAP developed a simple scoring sheet to calculate the corrosion profile for a bridge deck as shown in the table below.

Illustrative Examples 101 Corrosion Profile, Concrete Bridge Deck Attribute Score D.4 Poor Deck Drainage and Ponding • The deck drainage system is of modern design and is effective 0 D.6 Year of Construction • Bridge constructed in 2006 0 D.7 Application of Protective Systems • Protective systems never applied to deck 10 D.8 Concrete Mix Design • Constructed of normal grade concrete, no admixtures 15 D.11 Minimum Concrete Cover • Design cover is 1.5 inches 10 D.12 Reinforcement Type • Epoxy-coated reinforcement used 0 L.3 Exposure Environment • Deck environment is moderate 10 L.5 Rate of De-icing Chemical Application • Rate of de-icing chemical application is moderate 15 C.5 Maintenance Cycle • Bridge receives regular, periodic maintenance 0 Corrosion Profile score 60 out of 140 Attributes were identified by the RAP that affected the reliability and durability of a bare concrete deck. These attributes include the corrosion profile score, plus attributes based on the loading and the condition of a particular deck. The RAP identified screening criteria of the Cur- rent Condition Rating and Fire Damage for concrete bridge decks, to identify decks that may require further assessment. Other attributes of bare concrete decks were identified and ranked. The scoring plan was then applied to the subject concrete deck. Corrosion Damage, Concrete Bridge Deck Attribute Score S.1 Current Condition Rating • Current deck condition rating is greater than 4 Pass S.2 Fire Damage • No fire damage in the past 12 months Pass Corrosion Profile score 60 L.1 ADTT • ADTT is moderate (210 vehicles) 10 C.1 Current Condition Rating • Current deck condition rating is 7 0 C.8 Corrosion-Induced Cracking • Minor corrosion-induced cracking noted 5

102 Proposed Guideline for Reliability-Based Bridge Inspection Practices Corrosion Damage, Concrete Bridge Deck Attribute Score C.9 General Cracking • No general cracking observed 0 C.10 Delaminations • No delaminations found 0 C.11 Presence of Repaired Areas • No repaired areas 0 C.12 Presence of Spalling • No spalling noted 0 C.13 Efflorescence/Staining • Minor efflorescence without rust observed 5 Corrosion Damage total 80 out of 290 Corrosion Damage ranking 1.1 Low This bridge deck is still relatively new, was built to modern standards for durability and cor- rosion resistance, and has very little damage accumulation. As a result, the deck received very low scores for the attributes identified. Based on the attribute score, the RAP estimated that the likelihood of the failure for the deck (based on the criteria described in Section 2.1) in the next 72 months was low, i.e., the OF was Low (OF = 2). F 2.2.2 Prestressed Girder For the assessment of a prestressed girder, the corrosion profile scoring model was also used. As with the corrosion profile for bridge decks, this basic profile can be applied across many concrete elements. In this case, the prestressed girder scored the same as the deck. Corrosion Profile, Prestressed Girder Attribute Score D.4 Poor Deck Drainage and Ponding • The deck drainage system is of modern design and is effective 0 D.6 Year of Construction • Bridge constructed in 2006 0 D.7 Application of Protective Systems • Protective systems never applied 10 D.8 Concrete Mix Design • Constructed of normal grade concrete 15 D.11 Minimum Concrete Cover • Minimum concrete cover is 2 inches 10 D.12 Reinforcement Type • Reinforcement is epoxy coated 0 L.3 Exposure Environment • Superstructure environment is moderate 10 L.5 Rate of De-icing Chemical Application • Rate of salt application is moderate 15

Illustrative Examples 103 Corrosion Profile, Prestressed Girder Attribute Score C.5 Maintenance Cycle • Bridge receives regular, periodic maintenance 0 Corrosion Profile point total 60 out of 140 The RAP then considered the identified damage modes for a prestressed girder element, iden- tified and ranked attributes, and applied the scoring model for each damage mode as shown below. Bearing Area Damage, Prestressed Girder Attribute Score Corrosion Profile score 60 D.1 Joint Type • Bridge contains a closed joint system 0 C.4 Joint Condition • Joints are not leaking 5 C.8 Corrosion-Induced Cracking • No corrosion-induced cracking noted 0 C.9 General Cracking • No general cracking observed 0 C.11 Presence of Repaired Areas • No repaired areas 0 C.12 Presence of Spalling • No areas of spalling noted 0 Bearing Area Damage point total 65 out of 240 Bearing Area Damage ranking 1.08 Low Corrosion Between Beam Ends, Prestressed Girder Attribute Score Corrosion Profile score 60 C.8 Corrosion-Induced Cracking • No corrosion-induced cracking noted 0 C.10 Delaminations • No delaminations found 0 C.11 Presence of Repaired Areas • No repaired areas 0 C.12 Presence of Spalling • No spalling present 0 C.13 Efflorescence/Staining • No signs of efflorescence 0

104 Proposed Guideline for Reliability-Based Bridge Inspection Practices Corrosion Between Beam Ends, Prestressed Girder Attribute Score Corrosion Between Beam Ends point total 60 out of 235 Corrosion Between Beam Ends ranking 1.02 Low Flexural/Shear Cracking, Prestressed Girder Attribute Score S.4 Flexural Cracking • No flexural cracking Pass S.5 Shear Cracking • No shear cracking Pass D.2 Load Posting • Bridge is not load posted 0 L.4 Likelihood of Overload • Likelihood of overload is low 0 C.14 Flexural Cracking • No flexural cracking 0 C.15 Shear Cracking • No shear cracking 0 Flexural/Shear Cracking point total 0 out of 55 Flexural/Shear Cracking ranking 0 Remote Strand Fracture, Prestressed Girder Attribute Score S.1 Current Condition Rating • Superstructure condition rating is greater than 4 Pass S.6 Longitudinal Cracking in Prestressed Elements • Significant cracking is not present Pass Corrosion Profile score 60 L.6 Subjected to Overspray • Bridge not over a roadway, not exposed to overspray 0 C.1 Current Condition Rating • Superstructure condition rating is 8 0 C.4 Joint Condition • Joints are present but not leaking 5 C.8 Corrosion-Induced Cracking • No corrosion-induced cracking noted 0 C.10 Delaminations • No delaminations found 0 C.11 Presence of Repaired Areas • No repaired areas 0 C.12 Presence of Spalling • No spalling present 0

Illustrative Examples 105 Strand Fracture, Prestressed Girder Attribute Score C.16 Longitudinal Cracking in Prestressed Elements • No longitudinal cracking in the girders 0 Strand Fracture point total 65 out of 285 Strand Fracture ranking 0.91 Remote Based on the attributes identified by the RAP, the OF for the bearing area damage and corro- sion between the beam ends was estimated to be Low (OF = 2). For the damage modes of shear cracking, flexural cracking and strand fracture, the OF was Remote (OF = 1). F 2.2.3 Substructure For the piers and abutments, the RAP considered that the most likely damage modes were corrosion-induced cracking and spalling, or a settlement or rotation of one of the substruc- ture elements. However, settlement and rotations were determined to not be relevant damage modes because the bridge substructure is founded on rock. To estimate the likelihood for the corrosion damage mode, the panel once again used the generalized corrosion profile scoring. The panel then considered appropriate attributes for estimating the OF for the corrosion damage mode, identified and ranked key attributes, and scored the piers and abutments for the bridge, as shown below. Corrosion Profile, Substructure Attribute Score D.4 Poor Deck Drainage and Ponding • Deck does not drain onto the substructure 0 D.6 Year of Construction • Bridge constructed in 2006 0 D.7 Application of Protective Systems • Protective systems never applied 10 D.8 Concrete Mix Design • Substructure constructed with normal grade concrete 15 D.11 Minimum Concrete Cover • Minimum design cover is 2 inches 10 D.12 Reinforcement Type • Reinforcement is uncoated carbon steel 15 L.3 Exposure Environment • Environment is rated as moderate 10 L.5 Rate of De-icing Chemical Application • Rate of de-icing chemical application is moderate 15 C.5 Maintenance Cycle • Bridge receives regular, periodic maintenance 0 Corrosion Profile point total 75 out of 140

106 Proposed Guideline for Reliability-Based Bridge Inspection Practices Corrosion Damage—Piers and Abutments, Substructure Attribute Score Corrosion Profile score 75 C.1 Current Condition Rating • Current substructure condition rating is 8 0 C.4 Joint Condition • Joints present but not leaking 5 C.8 Corrosion-Induced Cracking • No corrosion-induced cracking noted 0 C.9 General Cracking • No cracking observed 0 C.10 Delaminations • No delaminations found 0 C.11 Presence of Repaired Areas • No repaired areas present 0 C.12 Presence of Spalling • No spalling noted 0 C.13 Efflorescence/Staining • No signs of efflorescence 0 Corrosion Damage point total 80 out of 290 Corrosion Damage ranking 1.10 Low Based on the attribute scoring, the OF for the damage mode of “Corrosion Damage” was assessed to be Low (OF = 2). F 2.3 Consequence Assessment Once the likelihood for each damage mode has been ranked, the RAP must perform a con- sequence analysis for each damage mode considered. For the concrete bridge deck, based on the damage mode of corrosion damage, the RAP considered the scenario of significant spalling of the deck as a result of extensive corrosion damage. Since the bridge is over a non-navigable waterway, spalling of concrete from the soffit would have a low consequence. Considering the ADT and the posted speed limit, spalling on the deck surface was determined to have only a moderate effect on serviceability for the bridge and a planned repair. The consensus of the RAP was that the appropriate Consequence Factor was Moderate (CF = 2). The RAP’s consequence assessment will be included in the file for the bridge. For the prestressed girder superstructure, in order to determine the consequence of failure, the RAP considered the scenario that one of the prestressed beams lost 100% of its load carrying capacity due to the damage modes of strand fracture, flexural and shear cracking, or corrosion between the beam ends. The RAP reviewed data from two very similar bridges for which truck impacts severely damaged one or more of the prestressed girders. The RAP determined that these two bridges could be considered “very similar” as their span lengths were within 10% of the bridge under consideration, and had nearly identical girder spacing and deck configuration. In both cases, the impact severely damaged at least one of the girders such that its load carrying capacity was effectively reduced to 0. The bridges exhibited little or no additional dead load deflection and were capable of carrying normal live loads. Temporary barriers were installed to shift traffic away from the shoulder area above the fascia girders that were damaged. Further, the load rating information

Illustrative Examples 107 for this bridge was reviewed and the bridge possessed a capacity far in excess of the required Inven- tory and Operating ratings. Hence, the RAP concluded that the loss of one girder would at most have a Moderate (CF = 2) consequence based on the following rationale: • The bridge is redundant, based on AASHTO definitions; • The bridge is very similar to other bridges for which a member failure has occurred, but did not result in collapse of the bridge or excessive deflection; • The bridge capacity far exceeds required Inventory and Operating ratings; • The bridge has low ADT, such that there will not be a major impact on traffic; and • The bridge is located over a non-navigable stream. Thus, the risks to people or property under the bridge are minimal. For the damage mode of bearing area damage, two scenarios were considered. The first scenario considered that the bearing area damage was sufficient to result in a downward displacement of the bridge deck. The most likely consequence was assessed by the panel to be Moderate, because such a displacement would result in only moderate disruption of service and require a planned repair. This was based on the rationale that the deck is composite with the superstructure and the bridge is a multi-girder bridge with normal beam spacing, such that any displacement would be minor and localized in nature, because loads could transfer to adjacent girders and the composite deck would limit displacements. The second scenario considered was that the bearing area damage resulted in severe cracking in the shear area of the beam, resulting in damage to the development length of the strands or shear cracking. The RAP considered that such a scenario would, at worst, result in 100% loss in load carrying capacity, as was considered for the damage modes of strand fracture, flexural or shear cracking, and corrosion between the beams ends in the previous scenario. Based on these two scenarios, the CF of Moderate (CF = 2) was selected for this damage mode. The RAP’s consequence assessment will be included in the file for the bridge. For the reinforced concrete substructure, the RAP considered the scenario that there was wide- spread corrosion damage (cracking and spalling) to the piers and abutments. The bridge is over a small creek, and hence there is little concern of injury from spalling concrete. The piers and abut- ments are short. Past experience of the panel with many piers and abutments of similar character- istics indicated that serious corrosion damage has a benign immediate effect on serviceability and safety. Therefore, the consensus of the panel was that the appropriate consequence category was Low (CF = 1). The data from the RAP assessment was then applied to the appropriate risk matrix (Figure C1) to determine the maximum inspection interval for the bridge. A summary of the scoring and maxi- mum inspection interval for the bridge are shown below. F 2.4 Scoring Summary Table F1 shows a summary of the analysis for this bridge. The maximum inspection interval based on the RAP analysis was determined to be 72 months, based on the low likelihood of serious damage (failure) to the elements of the bridge, and the moderate consequences associated with that damage. F 2.5 Criteria for a Family of Bridges The RAP assessed that it has many bridges in its inventory of very similar design characteris- tics. Based on the key attributes developed by the RAP, the panel identified a series of criteria to apply to a family of bridges to extend this analysis to other bridges in its inventory. These criteria describe bridges of the same design type and characteristics, with similarly adequate load ratings, and similar environmental loading. Condition attributes were mapped to suitable surrogates in the element-level bridge inspection data that were being collected for the bridge. For example, for the prestressed concrete girders, the panel identified that the individual condition attributes identified

108 Proposed Guideline for Reliability-Based Bridge Inspection Practices by the analysis, such as shear or flexural cracking, corrosion-induced cracking, spalling, or efflores- cence, were either not present or minimal if the CS ratings for the element were CS 1 or CS 2. There- fore, for the prestressed girder element, elements that are rated as CS 1 and CS 2 would not have the damage characteristics the panel identified as key to the potential for serious damage to develop. Bridges with any portion of the prestressed element rated as CS 3 would likely have one or more of these condition attributes present, and therefore would require reanalysis and possibly a reduced inspection interval. Similar criteria were developed for each of the elements assessed by the RAP. The RAP also identified that “longitudinal cracking in prestressed elements” was a key con- dition attribute not adequately represented in its element-level inspection scheme. As a result, the RBI procedure for bridges in this family needed to include a requirement that longitudinal cracking be assessed during the inspection. This requirement was included in the RBI procedure as a special emphasis area for this family of bridges. The RAP developed a listing of criteria, including design characteristics and using surrogate element data for certain condition attributes, to apply to the overall family of similar bridges in its inventory. These criteria are based on the engineering assessment documented through the RAP analysis. Example criteria to identify the family of bridges included: • Maximum span length less than 120 feet; • Four or more AASHTO prestressed girders; • Beam spacing of 10 feet or less; • ADTT less than 1000; • Constructed in 1995 or later; • No structural element with CS 3 reported; • No joint element with CS 3 reported; • Load rating exceeds requirements; • No significant flexural, shear, or longitudinal cracking in the prestressed element; and • Bridge receives RBI-based inspections. The RAP determined that bridges meeting these criteria will be treated as a family under the RBI methodology. If a particular bridge violates any of these criteria, it must be reassessed according to the attribute scoring criteria developed for this family of bridges. Table F2 summarizes the information from the RAP analysis to be included in the RBI proce- dure for these bridges. Longitudinal cracking in the prestressed elements is indicated as a special emphasis area for the inspection, to ensure this key damage mode is assessed during subsequent inspections. Other IPNs for identified damage modes are low, indicating a standard RBI inspec- tion is required for the bridge. Element Damage Occurrence Factor (OF) Consequence Factor (CF) Maximum Interval OF x CF (IPN) Deck Corrosion Damage Low (2) Moderate (2) 72 months 4 Prestressed Girders Bearing Area Damage Low (2) Moderate (2) 72 months 4 Corrosion Between Beam Ends Low (2) Moderate (2) 72 months 4 Flexural/Shear Cracking Remote (1) Moderate (2) 72 months 2 Strand Fracture Remote (1) Moderate (2) 72 months 2 Substructure Corrosion Damage Low (2) Low (1) 72 months 2 Table F1. Reliability assessment scoring summary for Example Bridge 1.

Illustrative Examples 109 F 3 Example 2: Steel Girder Bridge F 3.1 Bridge Profile F 3.1.1 Overview This example bridge carries a state highway over a non-navigable river. The bridge was con- structed in 1954 with a continuous steel girder superstructure, a non-composite reinforced con- crete deck, and a reinforced concrete substructure (Figure F2). All steel reinforcement used in this bridge is regular uncoated mild carbon steel. The observed ADTT is 130 vehicles. The rate of salt application is determined to be high by the RAP, with more than 100 applications of de-icing chemicals per year. The exposure environment is considered moderate. F 3.1.2 Concrete Bridge Deck The reinforced concrete bridge deck was constructed of cast-in-place normal concrete. From the design plans, the minimum cover was determined to be 1-9⁄16 inches. The deck has a bitumi- nous wearing surface of unknown thickness which was assessed to be in fair condition. In some locations the wearing surface has come off the deck. No membranes or sealers have been applied. The deck has no reported drainage or ponding problems. Maximum Inspection Interval: 72 months Special Emphasis Items S.6 Longitudinal Cracking in Prestressed Elements RBI Damage Modes Element Damage Mode IPN Deck Corrosion Damage 4 Prestressed Girder Bearing Area Damage 4 Corrosion Between Beam Ends 4 Flexural/Shear Cracking 2 Strand Fracture 2 Substructure Corrosion Damage 2 Table F2. Table of information to be included in the RBI procedure. Figure F2. Elevation view of Example Bridge 2.

110 Proposed Guideline for Reliability-Based Bridge Inspection Practices The most recent inspection rated the deck condition as 6-Satisfactory. According to the inspection report, the underside of the deck has hairline transverse cracks, spaced 2 to 3 feet apart, with efflorescence stains. The underside of the approach span at abutment 1 has heavy efflorescence stains on the left side. F 3.1.3 Steel Girders The continuous steel girder superstructure is constructed from four painted steel girders with steel diaphragms. These girders are riveted at the connection plates. No problems were found at the connection plates during a recent in-depth inspection. The bottom flanges of the girders have corrosion with missing paint. These locations have some pack rust formation. The super- structure was assessed to have a condition rating of 6-Satisfactory. Based on the inspection report, no fatigue or fracture related damage is present. Based on the provided design plans, it was determined that the girders are riveted built-up members, so the worst fatigue detail category is D. F 3.1.4 Substructure The substructure was constructed of normal grade reinforced concrete with uncoated carbon steel reinforcement. The minimum cover was determined to be 3-3⁄8 inches. Drainage from the deck is leaking onto the substructure from the deck due to leaking joints. There is no observed evidence of rotation or settlement. The concrete piers have random hair- line cracks with some moderate surface scaling below the high water line. Hairline to 1⁄32 inch (0.03125 inch) diagonal and vertical cracks with minor efflorescence stains have been reported on the concrete abutments. The concrete pier caps have some hairline cracks but appear to be in good condition. There is spalling in the concrete piers exposing rebar. The substructure condition was assessed to be 6-Satisfactory. F 3.2 Assessment The primary elements of this bridge are a concrete bridge deck with an asphalt overlay, riveted steel girders, deck joints, piers, and abutments. For the concrete bridge deck element the typical damage modes identified were concrete cracking and spalling. Since each of these damage modes results from the effects of corrosion, these damage modes were again grouped into a single dam- age mode termed “Corrosion Damage.” The same corrosion profile as developed for the previ- ous example was used for the deck. The asphalt overlay for the deck was assessed individually for debonding and spalling/potholes. For the steel girders, the damage modes considered were: • Corrosion Damage, • Fatigue Damage, and • Fracture Damage. For the substructure, the damage mode considered was: • Corrosion Damage (cracking and spalling due to the effects of corrosion). The RAP determined through consensus that tilting of the piers or unexpected settlement were not credible damage modes. This was based on the rationale that the bridge had been in service for more than 50 years without any signs of tilt or rotation, the geographic area was not susceptible to subsurface erosion or unexpected settlements, and the roller bearings were insen- sitive to moderate displacements of the substructure. F 3.2.1 Concrete Bridge Deck The concrete deck was assessed for the damage mode of corrosion damage, using the corrosion profile for concrete elements and attributes identified for the deck, as shown below.

Illustrative Examples 111 Corrosion Profile, Concrete Bridge Deck Attribute Score D.4 Poor Deck Drainage and Ponding • No drainage problems noted 0 D.6 Year of Construction • Bridge constructed in 1954 6 D.7 Application of Protective Systems • Protective systems never applied to deck 10 D.8 Concrete Mix Design • Constructed of normal grade concrete, no admixtures 15 D.11 Minimum Concrete Cover • Design cover is between 1.5 inches and 2.5 inches 10 D.12 Reinforcement Type • Uncoated carbon steel reinforcement 15 L.3 Exposure Environment • Deck environment is moderate 10 L.5 Rate of De-icing Chemical Application • Rate of de-icing chemical application is high (100 times per year) 20 C.5 Maintenance Cycle • Maintenance cycle is at least limited 10 Corrosion Profile score 96 out of 140 Corrosion Damage, Concrete Bridge Deck Attribute Score S.1 Current Condition Rating • Current deck condition rating is greater than 4 Pass S.2 Fire Damage • No fire damage in the past 12 months Pass Corrosion Profile score 96 L.1 ADTT • ADTT is minor (130 vehicles) 5 C.1 Current Condition Rating • Current deck condition rating is 6 5 C.8 Corrosion-Induced Cracking • Minor corrosion-induced cracking noted 5 C.9 General Cracking • No general cracking observed 0 C.10 Delaminations • Unknown—Asphalt overlay prevents effective sounding 20 C.11 Presence of Repaired Areas • No repaired areas 0 C.12 Presence of Spalling • No spalling noted 0

112 Proposed Guideline for Reliability-Based Bridge Inspection Practices Corrosion Damage, Concrete Bridge Deck Attribute Score C.13 Efflorescence/Staining • Moderate efflorescence without rust observed 10 Extent of Damage total 141 out of 290 Corrosion damage ranking 1.94 Low Based on the attributes identified by the RAP, the OF for corrosion damage was assessed to be Low (OF = 2). F 3.2.2 Asphalt Overlay The asphalt overlay was assessed by the panel using a simple expert elicitation. The general consensus of the panel was that the typical service life of an asphalt overlay was less than 10 years. The RAP agreed that the likelihood of failure of the asphalt overlay was greater than 1% over a 72-month interval, given that the overlay was already in service. The OF for the overlay failure was determined to be High (OF = 4) by consensus of the panel. F 3.2.3 Steel Girders The steel girders were assessed for three damage modes: Fatigue Damage, Corrosion Damage, and Fracture Damage. Key attributes were identified by the RAP as shown below. Supporting data and rationale for each attribute are included in the commentary. Fatigue Damage, Steel Girder Attribute Score S.7 Active Fatigue Cracks due to Primary Stress Ranges • No active fatigue cracks due to primary stress Pass D.6 Year of Construction • Bridge was built in 1954 20 D.16 Element Connection Type • Element is connected by rivets 7 D.17 Worst Fatigue Detail Category • Worst fatigue detail category is D 15 L.1 ADTT • ADTT is 130 vehicles 15 L.7 Remaining Fatigue Life • Remaining fatigue life is unknown 10 C.18 Condition of Fatigue Cracks • No fatigue cracks present 0 C.19 Presence of Fatigue Cracks due to Secondary or Out-of-Plane Stress • No fatigue cracks due to secondary or out of plane stress 0 Fatigue Damage point total 67 out of 110 Fatigue Damage ranking 2.44 Moderate

Illustrative Examples 113 Corrosion Damage, Steel Girder Attribute Score S.9 Significant Level of Active Corrosion or Section Loss • Active corrosion present is not alarming Pass D.5 Use of Open Decking • Bridge does not have an open deck 0 D.13 Built-Up Member • Element is built up 15 D.15 Constructed of Weathering Steel • Element not constructed with weathering steel 10 L.3 Exposure Environment • Exposure environment is moderate 10 L.5 Rate of De-icing Chemical Application • Rate of de-icing chemical application is high (100 times per year) 20 L.6 Subjected to Overspray • Superstructure is not subjected to overspray 0 C.4 Joint Condition • Joints are moderately leaking 15 C.7 Quality of Deck Drainage System • Drainage system is of adequate quality 0 C.17 Coating Condition • Element is painted, with steel exposed on bottom flanges 10 C.21 Presence of Active Corrosion • Significant active corrosion is present 20 C.22 Presence of Debris • Element has no debris 0 Corrosion Damage point total 100 out of 190 Corrosion Damage ranking 2.1 Moderate Fracture Damage, Steel Girder Attribute Score S.7 Active Fatigue Cracks due to Primary Stress Ranges • No active fatigue cracks due to primary stress Pass S.8 Details Susceptible to Constraint-Induced Fracture • No details susceptible to constraint induced fracture Pass D.3 Minimum Vertical Clearance • Bridge is not over a roadway, max vertical clearance 0 D.6 Year of Construction • Bridge constructed in 1954 20 D.14 Constructed of High Performance Steel • Element is not constructed of HPS/unknown 10 L.1 ADTT • ADTT is 130 vehicles 15

114 Proposed Guideline for Reliability-Based Bridge Inspection Practices Fracture Damage, Steel Girder Attribute Score L.7 Remaining Fatigue Life • Remaining fatigue life is unknown 10 C.6 Previously Impacted • Bridge has not been impacted before 0 C.19 Presence of Fatigue Cracks due to Secondary or Out-of-Plane Stress • No fatigue cracks present 0 C.20 Non-Fatigue-Related Cracks or Defects • No fatigue cracks present 0 Fracture Damage point total 55 out of 125 Fracture Damage ranking 1.76 Low The RAP analysis of key attributes for the damage modes indicated that the steel superstruc- ture has a moderate likelihood of fatigue damage (OF = 3), a moderate likelihood of developing corrosion damage (OF = 3), and a low likelihood of fracture (OF = 2). F 3.2.4 Substructure The substructure was assessed for the damage mode of corrosion damage, using the corrosion profile for concrete elements and attributes identified for the piers and abutments. Corrosion Profile, Substructure Attribute Score D.4 Poor Deck Drainage and Ponding • No drainage problems noted 0 D.6 Year of Construction • Bridge constructed in 1954 6 D.7 Application of Protective Systems • Protective systems have not been applied 10 D.8 Concrete Mix Design • Substructure constructed with normal grade concrete, no admixtures 15 D.11 Minimum Concrete Cover • Minimum design concrete cover is 3-3⁄8″ 0 D.12 Reinforcement Type • Reinforcement is uncoated carbon steel 15 L.3 Exposure Environment • Exposure environment is moderate 10 L.5 Rate of De-icing Chemical Application • Rate of de-icing chemical application is high (100 times per year) 20 C.5 Maintenance Cycle • Maintenance cycle is at least limited 10 Corrosion Profile point total 86 out of 140

Illustrative Examples 115 Corrosion Damage—Piers and Abutments, Substructure Attribute Score Corrosion Profile score 86 C.1 Current Condition Rating • Current substructure condition rating is six 5 C.4 Joint Condition • Joints are significantly leaking onto substructure 20 C.8 Corrosion-Induced Cracking • Moderate corrosion-induced cracking noted 10 C.9 General Cracking • Presence of minor general cracking 5 C.10 Delaminations • Minor localized delaminations on footings 5 C.11 Presence of Repaired Areas • No repaired areas present 0 C.12 Presence of Spalling • Significant spalling with exposed reinforcement present on piers 20 C.13 Efflorescence/Staining • Moderate efflorescence without rust staining 10 Substructure Elements point total 161 out of 290 Substructure Elements ranking 2.22 Moderate Based on the attribute scoring, the RAP estimated the OF was Moderate (OF = 3) for corrosion damage for the piers and abutments. A considerable amount of damage has already accumulated in the form of spalling with exposed reinforcement and moderate cracking. F 3.3 Consequence Assessment Since the bridge carries a state highway over a non-navigable river, key damage to the bridge deck is likely to be in the form of spalling on the riding surface of the bridge deck. The most likely consequence of severe damage to the deck is Moderate (CF = 2) because there may be some disruption of service or reduction in posted speed. The bridge is a four girder bridge with typical girder spacing, such that even a through-thickness punch-through is likely to be local in nature and not represent a high consequence. The assignment of a moderate consequence is based on common experience with bridge decks of similar design characteristics. The consequence of the asphalt overlay failing was determined to be Low, because failure of the asphalt overlay was a maintenance need and would not necessitate increased inspection or monitoring. The superstructure consists of four steel girders with diaphragms spaced at 20 to 25 feet. Although fatigue damage is the most likely damage mode, the worst outcome associated with fatigue would be the fracture of one of the girders. Hence, the consequence scenario evaluated was the fracture of one of the girders. Note that this analysis does not depend on the damage failure mode, thus, failure could also be due to corrosion. As stated, the cross section is made up of four identical built-up members. In evaluating the most likely consequence, the RAP identified several similar designs where full-depth fractures of steel girders occurred. These bridges had spans greater

116 Proposed Guideline for Reliability-Based Bridge Inspection Practices than or equal to this bridge, had similar skew, had similar girder spacing, and had a non-composite deck. In all cases, none of the bridges collapsed, though some displayed minor sagging. The bridges carried full service load up until the time that fracture was detected in later inspections. Hence, the RAP determined that the consequence associated with fracture of one of the girders should be set as High (CF = 3) based on the following rationale: • The bridge is redundant, based on AASHTO definitions; • The bridge is very similar to other bridges where full-depth girder fractures occurred, but did not result in collapse of the bridge or excessive deflection; • The bridge meets required Inventory and Operating ratings; • Fracture in a member will have a major impact on travel, since the member failure would result in a lane closure; and • The bridge is located over a non-navigable river. Thus, the risks to people or property under the bridge are minimal. The RAP’s consequence assessment will be included in the bridge file along with appropriate references to the other bridges cited in the consequence scenario evaluation. Engineering calculations showing that the effects of a girder fracture would result in a Mod- erate consequence (CF = 2) would be required to reduce the consequence category for this sce- nario. Based on the above and the fact that the bridge is not fracture-critical, the consequence category of Severe was not considered a plausible outcome for girder fracture. For the substructure, the scenario considered for damage to the piers and abutments of the bridge was severe corrosion damage and spalling. The most likely consequence of this scenario is a Low consequence (OF = 1), because severe corrosion damage of this type would typically require monitoring and assessment, but would not affect the serviceability of the bridge. The summary of the RAP assessment is shown in Table F3. Based on this assessment, the maxi- mum inspection interval for this bridge is 24 months, due to the likelihood and high consequence associated with the development of fatigue cracking. This is due to in part to the fact that the bridge has fatigue-prone details (category D), the bridge was constructed before modern fracture control requirements were in place, and there is truck traffic on the bridge. Even though the bridge has not developed any fatigue cracks in more than 50 years of service, the rational assessment performed by the RAP indicates that the potential for cracking exists, and should be treated appropriately. Addi- tionally, the bridge is susceptible to serious corrosion damage, because its current condition includes active corrosion, the applications of de-icing chemical are high, the members are built up, and the joints are leaking. As such, the required maximum interval for an RBI is 24 months. F 3.4 Scoring Summary The scoring summary for this bridge is shown in Table F3. Based on the reliability assessment, the maximum inspection interval was determined to be 24 months. Element Damage Occurrence Factor (OF) Consequence Factor (CF) Interval OF x CF (IPN) Deck Corrosion Damage Low (2) Moderate (2) 72 months 4 Overlay Debonding/Spalling High (4) Low (1) 48 months 4 Steel Girders Fatigue Moderate (3) High (3) 24 months 9 Corrosion Moderate (3) High (3) 24 months 9 Fracture Low (2) High (3) 48 months 6 Substructure Corrosion Damage Moderate (3) Low (1) 48 months 3 Table F3. Reliability assessment scoring summary for Example Bridge 2.

Illustrative Examples 117 F 3.5 Inspection Data Table F4 summarizes the information from the RAP analysis to be included in the RBI proce- dure to be used for this bridge. The identified screening criteria for fatigue cracking due to pri- mary stresses and significant section loss are included as special emphasis items. The data in the table also indicates that fatigue cracking and corrosion damage are priority items for inspection of the steel girders, based on their IPN of 9. Because of the high IPN for corrosion damage, the RAP recommends utilizing an ultrasonic thickness gauge (UT-T) to assess the areas of section loss in the steel girder. This will ensure accurate reporting of the remaining section and mitigate the risks associated with severe section loss, which was identified through the RAP analysis as being moderately likely to occur, and resulting in a high consequence. For fatigue cracking, the high IPN number will prioritize fatigue cracking for the inspection team conducting the RBI on the bridge. The IPN of 9 indicates to the inspector that the bridge has the potential for fatigue cracking, and the consequences of that cracking are potentially high were it to go undetected. The RAP could recommend NDE, such as Magnetic Particle Testing (MT) or Dye Penetrant Testing (PT), be applied to a sampling of locations during periodic inspections to ensure that fatigue cracking is detected and enhance the reliability of the inspection. F 4 Example 3: Reinforced Concrete Bridge F 4.1 Bridge Profile F 4.1.1 Overview This example bridge is a typical, simply-supported three-span reinforced concrete bridge with a bare cast-in-place deck. The bridge owner’s inventory includes more than 100 bridges of simi- lar span length and design characteristics, and, as such, is developing the RAP analysis for appli- cation to a family of bridges, using this bridge as an example of the family. The specific bridge was constructed in 1963 and carries highway traffic over a local road. The estimated ADT on the bridge is 22,000 vehicles, while the ADT on the local road under the bridge is 60 vehicles. Both the rate of salt application and the surrounding environment are considered to be moderate. A photograph of the bridge is shown in Figure F3. F 4.1.2 Concrete Bridge Deck For this bridge, the deck was constructed with normal grade cast-in-place concrete and uncoated mild steel reinforcement. The asphalt has been removed from the top of the deck and a water proof sealant has been applied. Hairline to 1⁄16-inch cracks have been observed on the top of the deck near the abutments. Hairline diagonal cracks with efflorescence stains have been observed on the soffit of the deck near the abutments. No delaminations or spalling are noted on the deck. Maximum Inspection Interval: 24 Months Special Emphasis Items S.7 Active Fatigue Cracks due to Primary Stress Ranges S.9 Significant Level of Active Corrosion or Section Loss RBI Damage Modes Element Damage Mode IPN Deck Corrosion Damage 4 Steel Girder Fatigue Cracking 9 Corrosion Damage 9 Fracture 6 Substructure Corrosion Damage 3 Table F4. Table of information to be included in the RBI.

118 Proposed Guideline for Reliability-Based Bridge Inspection Practices From the design plans, the minimum cover was determined to be 1-13⁄16 inches. Based on the most recent inspection report, the deck is considered to be in CS 6-Satisfactory. This deck con- tains concrete edge joints with silicon sealant. The seals are considered to be in good condition but are leaking water. No other ponding or drainage issues are noted. F 4.1.3 Reinforced Concrete Girders The superstructure for this bridge consists of seven reinforced concrete girders that are constructed from normal grade concrete and uncoated mild steel reinforcement. Each girder, per span, has hairline vertical flexure cracking. The right exterior girder has a spall on the bottom end which measures 12 inches tall by 3 inches wide by 5 inches deep due to impact. One of the exterior girders has an 8-inch diameter spall resulting from an over-height vehicle collision. Girders five and six also have scrapes and spalls from an over-height vehicle collision. The superstructure is considered to be in CS 5-Fair. From the design plans, the minimum con- crete cover is 3-5⁄8 inches. F 4.1.4 Substructure The substructure for this bridge is also constructed of normal grade concrete with uncoated mild steel reinforcement. From the design plans, the minimum cover was deter- mined to be 2-½ inches. The columns have random hairline cracks and the top of column four has an area of delamination that is 29 inches tall by 21 inches wide. Both abutments have hairline to 1⁄16-inch vertical cracks and spalling with exposed reinforcement on their right sides. All bents have water staining resulting from leaking joints. Bent cap one, span one, has hori- zontal cracks with delamination in the bottom left corner. Bent cap two, span two, has an area of cracking and delamination that is 16 inches wide by 8 inches tall near girder six. Bent cap two, span three, also has an area of cracking and delamination that is 27 inches wide by 4 inches tall near girder six. The substructure has neoprene pad bearings which have curled on the ends but are still in satisfactory condition. The overall condition rating for the substructure is “5-Fair.” There are no signs of settlement or rotation and the substructure itself is founded on rock. Figure F3. Elevation view of Example Bridge 3.

Illustrative Examples 119 F 4.2 Assessment The primary elements of this bridge are a concrete bridge deck, reinforced concrete gird- ers, and piers and abutments. For the concrete bridge deck element, the typical damage mode identified was corrosion damage (concrete cracking and spalling). The same corrosion profile developed for the previous examples was also used for this deck. For the reinforced concrete girders, the damage modes considered were: • Bearing Area Damage, • Corrosion Between Beam Ends, and • Flexural and Shear Cracking. Based on the owner’s inventory data and experience, there has been no occurrences of significant shear cracking in bridges of similar design to the one being analyzed. However, there have been isolated cases of cracking due to flexural stresses, possibly resulting from overloaded trucks. Based on this experience, the RAP determines that flexural cracking is an important damage mode, while the likelihood of shear cracking is more remote, generally. To provide focus on the flexural cracking experience in this particular inventory, the RAP determines that shear cracking and flexural crack- ing should be separated into distinct damage modes. Additionally, the RAP determined through consensus that the likelihood of overload would have the greatest influence on the likelihood of flexural cracking progressing; existing flexural cracking had moderate effect, and the fact that bridge may be load posted has only a small effect. As such the RAP assigns 20 points to L.4, Likelihood of Overload, only 10 points to D.2, Load Posting and 15 points to C.14, Flexural Cracking. The key attributes for flexural cracking were therefore determined by the RAP to be as follows: • S.4 Flexural Cracking (screening criteria), • D.2 Load Posting, • L.4 Likelihood of Overload, and • C.14 Flexural Cracking. The screening criteria for Flexural Cracking (S.4) was also utilized to identify bridges with significant flexural cracking, which may require individual engineering assessment. For shear cracking, the relevant attributes identified by the RAP were: • S.5 Shear Cracking (screening), • D.2 Load Posting, • L.4 Likelihood of Overload, and • C.15 Shear Cracking. Again, the screening attribute S.5 for unresolved shear cracking is utilized to identify any bridges with shear cracking that may require engineering assessment. For the substructure, the damage mode considered was: • Corrosion Damage (cracking and spalling due to the effects of corrosion). F 4.2.1 Concrete Bridge Deck The concrete deck was assessed for the damage mode of corrosion damage, using the corro- sion profile for concrete elements and attributes identified for the deck, as shown below Corrosion Profile, Concrete Bridge Deck Attribute Score D.4 Poor Deck Drainage and Ponding • No drainage problems noted 0

120 Proposed Guideline for Reliability-Based Bridge Inspection Practices Corrosion Profile, Concrete Bridge Deck Attribute Score D.6 Year of Construction • Bridge constructed in 1963 6 D.7 Application of Protective Systems • Waterproof penetrating sealer applied, frequency unknown 5 D.8 Concrete Mix Design • Constructed of normal grade concrete, no admixtures 15 D.11 Minimum Concrete Cover • Design cover is between 1.5 inches and 2.5 inches 10 D.12 Reinforcement Type • Uncoated carbon steel reinforcement 15 L.3 Exposure Environment • Deck environment is moderate 10 L.5 Rate of De-icing Chemical Application • Rate of de-icing chemical application is moderate 15 C.5 Maintenance Cycle • Maintenance cycle is at least limited 10 Corrosion Profile score 86 out of 140 Corrosion Damage, Concrete Bridge Deck Attribute Score S.1 Current Condition Rating • Current deck condition rating is greater than four Pass S.2 Fire Damage • No fire damage in the past 12 months Pass Corrosion Profile score 86 L.1 ADTT • ADTT is high (5,500 vehicles) 20 C.1 Current Condition Rating • Current deck condition rating is 6 5 C.8 Corrosion-Induced Cracking • Moderate corrosion-induced cracking noted 10 C.9 General Cracking • Moderate general cracking observed 10 C.10 Delaminations • No delaminations noted 0 C.11 Presence of Repaired Areas • No repaired areas 0 C.12 Presence of Spalling • No spalling noted 0

Illustrative Examples 121 Corrosion Damage, Concrete Bridge Deck Attribute Score C.13 Efflorescence/Staining • Minor efflorescence without rust observed 5 Corrosion Damage total 136 out of 290 Corrosion Damage ranking 1.88 Low Based on the attributes identified by the RAP, the OF for corrosion damage in the deck was estimated as Low (OF = 2). F 4.2.2 Reinforced Concrete Girders The reinforced concrete girders were assessed for the damage modes of bearing area damage, corrosion between the beam ends, and flexural and shear cracking. Corrosion Profile, Reinforced Concrete Girder Attribute Score D.4 Poor Deck Drainage and Ponding • No drainage problems noted. 0 D.6 Year of Construction • Bridge constructed in 1963 6 D.7 Application of Protective Systems • Protective systems never applied 10 D.8 Concrete Mix Design • Constructed of normal grade concrete 15 D.11 Minimum Concrete Cover • Minimum concrete cover is greater than 2.5 inches 0 D.12 Reinforcement Type • Reinforcement is uncoated mild steel 15 L.3 Exposure Environment • Superstructure environment is moderate 10 L.5 Rate of De-icing Chemical Application • Rate of salt application is moderate 15 C.5 Maintenance Cycle • Bridge maintenance is at least limited 10 Corrosion Profile point total 81 out of 140 Bearing Area Damage, Reinforced Concrete Girder Attribute Score Corrosion Profile score 81 D.1 Joint Type • Bridge has closed joints 0

122 Proposed Guideline for Reliability-Based Bridge Inspection Practices Bearing Area Damage, Reinforced Concrete Girder Attribute Score C.4 Joint Condition • Joints are leaking but sealant is still in fair condition 15 C.8 Corrosion-Induced Cracking • No corrosion-induced cracking noted 0 C.9 General Cracking • No general cracking observed 0 C.11 Presence of Repaired Areas • No repaired areas 0 C.12 Presence of Spalling • Moderate spalling in several locations, no exposed reinforcement noted. 15 Bearing Area Damage point total 111 out of 240 Bearing Area Damage ranking 1.85 Low Corrosion Between Beam Ends, Reinforced Concrete Girder Attribute Score Corrosion Profile score 81 C.1 Current Condition Rating • Current condition rating is 5 20 C.6 Previously Impacted 20 C.8 Corrosion-Induced Cracking • No corrosion-induced cracking noted 0 C.9 General Cracking • No general cracking observed 0 C.10 Delaminations • Unknown 20 C.11 Presence of Repaired Areas • No repaired areas 0 C.12 Presence of Spalling • Moderate spalling in several locations (due to impact), no exposed reinforcement noted 15 C.13 Efflorescence/Staining • No signs of efflorescence 0 Corrosion Between Beam Ends point total 156 out of 290 Corrosion Between Beam Ends ranking 2.15 Moderate Flexural Cracking, Reinforced Concrete Girder Attribute Score S.4 Flexural Cracking • Hairline flexural cracking noted, determined to be benign Pass

Illustrative Examples 123 Flexural Cracking, Reinforced Concrete Girder Attribute Score D.2 Load Posting • Bridge is not load posted 0 L.4 Likelihood of Overload • Likelihood of overload is moderate 10 C.14 Flexural Cracking • Hairline flexural cracking noted 15 Flexural Cracking point total 25 out of 45 Flexural Cracking ranking 2.22 Moderate Shear Cracking, Reinforced Concrete Girder Attribute Score S.5 Shear Cracking • No shear cracking present Pass D.2 Load Posting • Bridge is not load posted 0 L.4 Likelihood of Overload • Likelihood of overload is moderate 10 C.15 Shear Cracking • No shear cracking 0 Shear Cracking point total 10 out of 45 Shear Cracking ranking 0.88 Remote The attribute scoring indicated an OF of Moderate (OF = 3) for corrosion between beam ends and flexural cracking, an OF of Low (OF = 2) for bearing area damage, and an OF of Remote (OF = 1) for shear cracking. F 4.2.3 Substructure The substructure was assessed for the damage mode of corrosion damage, using the corrosion profile for concrete elements and attributes identified for the piers and abutments. Corrosion Profile, Substructure Attribute Score D.4 Poor Deck Drainage and Ponding • No drainage problems noted 0 D.6 Year of Construction • Bridge constructed in 1963 6 D.7 Application of Protective Systems • Protective systems have not been applied 10 D.8 Concrete Mix Design • Substructure constructed with normal grade concrete, no admixtures 15

124 Proposed Guideline for Reliability-Based Bridge Inspection Practices Corrosion Profile, Substructure Attribute Score D.11 Minimum Concrete Cover • Minimum design concrete cover is 2-½ inches 0 D.12 Reinforcement Type • Reinforcement is uncoated carbon steel 15 L.3 Exposure Environment • Exposure environment is moderate 10 L.5 Rate of De-icing Chemical Application • Rate of de-icing chemical application is moderate 15 C.5 Maintenance Cycle • Maintenance cycle is at least limited 10 Corrosion Profile point total 81 out of 140 Corrosion Damage—Piers and Abutments, Substructure Attribute Score Corrosion Profile score 81 C.1 Current Condition Rating • Current substructure condition rating is five 20 C.4 Joint Condition • Joints are moderately leaking onto substructure 15 C.8 Corrosion-Induced Cracking • Localized cracking near delaminations noted 5 C.9 General Cracking • Presence of moderate general cracking 10 C.10 Delaminations • Unknown 20 C.11 Presence of Repaired Areas • No repaired areas present 0 C.12 Presence of Spalling • Moderate spalling with exposed reinforcement present 15 C.13 Efflorescence/Staining • No efflorescence noted 0 Concrete Elements point total 166 out of 290 Concrete Elements ranking 2.28 Moderate Based on their analysis, the RAP assessed that the likelihood of failure due to corrosion damage was moderate for the pier and abutments (OF = 3). Already, a considerable amount of damage has accumulated in the form of localized delaminations and spalling resulting in exposed reinforcement. F 4.3 Consequence For the concrete bridge deck, the RAP considered the scenario that the corrosion damage in the deck resulted in spalling of either the driving surface of the deck or deck soffit. In this

Illustrative Examples 125 case, the bridge carries a high-volume highway over another, lower-volume roadway. The roadway on the bridge carries 22,000 vehicles a day, and the roadway below the bridge carries 60 vehicles a day. Based on this information, any spalling from the deck soffit has the potential to fall into the roadway below and strike a motorist. However, given the low traffic volume and speed on the roadway below, the RAP considered the likelihood of this occurring to be relatively small. Therefore, the consensus of the RAP was that the appropriate Consequence Factor was High (CF = 3). For spalling of the riding surface, the panel determined that such a scenario was likely to have an effect on serviceability of the deck, and may require a reduction in the posted traffic speed. Therefore, the consensus of the RAP was that this represented a Consequence Factor of Moderate (CF = 2). For this case, the scenario of concrete falling into the roadway below the bridge provides the Consequence Factor for corrosion damage in the bridge deck. To determine the Consequence Factor for the concrete beams, the RAP considered the scenario that one of the reinforced concrete beams lost 100% of its load carrying capac- ity due to corrosion damage between the beam ends, flexural or shear cracking, or bearing area damage. The RAP considered that the superstructure is reinforced concrete with a composite deck such that redundancy in the structure would prevent the total collapse of a girder. The RAP also reviewed data from two very similar bridges for which corro- sion damage had resulted in loss of load carrying capacity in one girder of a multi-girder, reinforced concrete bridge with a composite deck. The RAP determined that these two bridges could be considered “very similar” to the bridge being analyzed because their span lengths were within 10% of the bridge under consideration and they utilized a nearly identi- cal girder spacing and deck configuration. In both cases, the corrosion damage had reduced a single girder’s load carrying capacity effectively to zero, however, the bridge exhibited little or no additional dead load deflection and was capable of carrying normal live loads. Lane clo- sures were required on the bridges as the result of the faulted girder, resulting in a significant impact on traffic. The load rating information for the bridge was reviewed and the bridge possessed a capacity far in excess of the required Inventory and Operating ratings. However, the bridge carries a high ADT, such that a lane closure would have a major impact on traffic. Additionally, the roadway under the bridge is a low-volume road that may be impacted by the shoring required or debris. As a result, the Consequences Factor was determined to be High (CF = 3) based on the following rationale: • The bridge is redundant, based on AASHTO definitions; • The bridge is very similar to other bridges for which a member failure has occurred, but did not result in collapse of the bridge or excessive deflection; • The bridge capacity far exceeds required Inventory and Operating ratings; • The bridge has high ADT, such that there could be a major impact on traffic; and • The bridge is located over a low-volume roadway such that there would be some risks to traffic on the roadway below. For the damage mode of bearing area damage, two scenarios were considered. The first scenario considered that the bearing area damage was sufficient to result in a downward dis- placement of the bridge deck. The most likely consequence was assessed by the panel to be Moderate (CF = 2), resulting in only a minor disruption of service, since the deck is composite with the superstructure and it is a multi-girder bridge with normal beam spacing. The second scenario considered was that the bearing area damage resulted in severe cracking in the shear area of the beam, resulting in a loss of load-carry capacity. As above, the Consequence Factor of High (CF = 3) was assigned.

126 Proposed Guideline for Reliability-Based Bridge Inspection Practices For the reinforced concrete substructure, areas of delaminations are present in several loca- tions and both abutments have areas of spalling with exposed reinforcement. Here, the most likely damage mode will result in spalling of the concrete. The RAP considers that this bridge is located over a roadway, and the piers are immediately adjacent to the roadway such there is a chance that concrete spalling off of a pier could strike a passing motorist. Based on this factor, the consequence scenario for this damage mode was assessed to be High (CF = 3). F 4.4 Scoring Summary Table F5 shows a summary of the scoring for this bridge. Based on the likelihood of corro- sion damage between the beam ends, flexural cracking and corrosion damage to the substruc- ture, and the associated consequences, the maximum inspection interval was determined to be 24 months. F 4.5 Inspection Data Table F6 summarizes the information from the RAP analysis to be included in the RBI proce- dure for this bridge. This information includes the identified screening criteria of inspection for flexural cracking as a special emphasis item for the RBI inspection. Based on the RAP assessment, this particular bridge has high IPNs for corrosion between the beam ends (9), corrosion damage in the substructure (9), and corrosion damage in the deck (6), Element Damage Occurrence Factor (OF) Consequence Factor (CF) Maximum Interval OF x CF (IPN) Deck Corrosion Damage Low (2) High (3) 48 months 6 Reinforced Concrete Girders Bearing Area Damage Low (2) High (3) 48 months 6 Corrosion Between Beam Ends Moderate (3) High (3) 24 months 9 Flexural Cracking Moderate (3) High (3) 24 months 9 Shear Cracking Remote (1) High (3) 48 Months 3 Substructure Corrosion Damage Moderate (3) High (3) 24 months 9 Table F5. Reliability assessment scoring summary for Example Bridge 3. Maximum Inspection Interval: 24 Months Special Emphasis Items S.4 Flexural Cracking RBI Damage Modes Element Damage Mode IPN Deck Corrosion Damage 6 Reinforced Concrete Girder Bearing Area Damage 6 Corrosion Damage between the beam ends 9 Flexural Cracking 9 Shear Cracking 3 Substructure Corrosion Damage 9 Table F6. Table of information to be included in RBI procedure.

Illustrative Examples 127 each of which have the potential to result in debris falling into the roadway below the bridge. As a result, the RAP determined that enhanced inspection for corrosion damage was needed as part of the RBI procedure. Available technologies to complete the delamination survey include hammer sounding, infrared thermography (IR) and Impact Echo (IE). The RAP recommends delamination surveys be completed during the periodic inspections to mitigate the risk of debris falling into the roadway below the bridge unexpectedly. Flexural cracking also has a high IPN, indicating that this damage mode is of high importance and needs to be prioritized during subsequent RBIs for the bridge. Flexural cracking is included as a special emphasis item for subsequent inspections.