Practices for Ensuring the Smoothness of Concrete Bridge Decks (2022)

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

8 This chapter summarizes the results from the literature review for the following topics: • Federal reporting requirements for bridges. • Bridge management systems. • Bridge preservation. • Differential settlement at the bridge-road interface. • Equipment for measuring smoothness of bridges for construction acceptance. • Ride quality indices. • Past studies that have been conducted in the area of bridge roughness. • Resources related to smoothness. Federal Reporting Requirements for Bridges State DOTs are required to perform biennial inspection of highway bridges and submit the data annually to the FHWA in a suitable format to be entered into the NBI database. The FHWA has developed a guide for recording and coding data elements that are contained in the NBI database (FHWA 1995). This guide includes instructions for recording 125 entries associated with a bridge, together with coding forms. Bridge condition ratings for the deck, superstructure, and substructure are assigned on a scale of 0 to 9, where 0 represents a failed condition and 9 rep- resents an excellent condition. A rating between 7 and 9 is considered as good, a rating between 5 and 6 is considered as fair, and a rating between 0 to 4 is considered as poor. If all the metrics fall into the good condition, the bridge is rated as good. If any of the metrics is rated to be in a poor condition, the bridge is rated as poor. The ride quality of the bridge deck is not a factor that is considered when rating the bridge deck. The FHWA published a final rule establishing performance measures for state DOTs to use in managing pavement and bridge performance on the NHS as rule 82 FR 5886 in the Fed- eral Register in January 2017 (Federal Register 2017). This rule, titled “National Performance Management Measures: Assessing Pavement Condition for the National Highway Performance Program and Bridge Condition for the National Highway Performance Program,” addressed requirements established by the Moving Ahead for Progress in the 21st Century Act and the Fixing America’s Surface Transportation Act. The rule became effective on May 20, 2017. The FHWA has set the upper limit for the percentage of all NHS bridges in a state classified to be in a poor condition at 10%. This percentage is computed by dividing the deck area of NHS bridges that are in a poor condition by the total deck area of all NHS bridges in the state, and then expressing this value as a percentage. C H A P T E R 2 Literature Review

Literature Review 9   Bridge Management Systems All state DOTs currently have a BMS (Markow and Hyman 2009). A BMS has bridge-related information, including NBIS data and ratings, but can also contain more detailed element-level data (Markow and Hyman 2009). The information in a BMS can be used by state DOTs to select bridges to receive maintenance or preservation treatments. Bridge Preservation The primary source of federal funding for bridges over the past several decades was the FHWA Highway Bridge Replacement and Rehabilitation Program, which was later known as the High- way Bridge Program (FHWA 2018). This program offered state DOTs the ability to use the funds that are provided not only for bridge rehabilitation and replacement, but also for many preven- tive maintenance activities. Cost-effective bridge preservation activities are eligible for federal funding, but not routine maintenance activities. An example of a routine maintenance activity is asphalt patching on a concrete bridge deck without applying a membrane on the deck. Typical rehabilitation activities performed on a bridge can include partial or complete deck replacement, superstructure replacement, substructure strengthening, and partial or full replacement of the substructure (FHWA 2018). Adopting a worst-first approach for managing bridges by focusing only on poor bridges while not addressing the maintenance needs of bridges that are in a good or fair condition is inefficient and cost-prohibitive in the long term (FHWA 2018). The FHWA has developed a guide that can be utilized by bridge owners to establish a bridge preservation program if they do not have one, or to improve an existing bridge preservation program (FHWA 2018). The FHWA bridge preservation guide defines bridge preservation as “actions or strategies that prevent, delay, or reduce deterioration of bridges or bridge elements; restore the function of existing bridges; keep bridges in good or fair condition; and extend their service life” (FHWA 2018). The FHWA guide describes common bridge preservation activities. Performing bridge preservation activities when the bridge is still in a good or fair condition and not exhibiting serious deterioration can delay performing rehabilitation or replacement of the bridge or com- ponents of the bridge. The preservation activities can be performed cyclically, or they can be performed based on existing conditions (FHWA 2018). Examples of cyclical maintenance activities that are performed on a bridge deck include (FHWA 2018) • Cleaning and washing deck, • Cleaning and flushing drains on the deck, • Cleaning joints on the deck, • Sealing crack on the deck, and • Sealing concrete surface on the deck. Condition-based maintenance activities are carried out on bridge components that show defects that are identified by inspection. These components may be in the substructure, super- structure, or the deck. Some examples of condition-based maintenance activities that are per- formed on the bridge deck and/or approach slabs include • Repairing or replacing drains, • Replacing joint sealant,

10 Practices for Ensuring the Smoothness of Concrete Bridge Decks • Repairing joints, • Repairing concrete deck, • Placing overlays on deck (e.g., thin polymer epoxy, AC, concrete), and • Repairing or replacing approach slabs (FHWA 2018). NCHRP Research Report 950: Proposed AASHTO Guides for Bridge Preservation Actions developed two guides for the preservation of highway bridges (Hearn 2020). One is a general guide for preservation of bridges and is referred to as the Bridge Guide, and the other guide is for preservation of bridge decks and is referred to as the Deck Guide. State DOTs can identify bridges for performing preservation treatments from the informa- tion contained in their BMS. They can also track the effectiveness of various preservation treat- ments using the data stored in the BMS. The FHWA bridge preservation guide does not mention the ride quality of the bridge deck, which is an attribute desired by highway users. However, bridge preservation activities, such as repairing joints or patching distresses on the bridge deck, can reduce the roughness felt by highway users at such features. Bridge preservation treatments that involve the placement of an overlay on a bridge deck are decided based on the distresses observed on the bridge deck and the condition of the bridge deck. Carrying out such an activity will result in a major improvement in the ride quality of the bridge deck. Repairing or replacing approach slabs that are in a poor condition will also reduce the roughness felt by highway users when entering or departing a bridge. Therefore, although bridge preservation activities performed on the bridge deck are not specifically performed to improve the smoothness of the bridge deck, as described above, some of the bridge preservation activities can result in reducing the roughness felt by highway users as they traverse the bridge. Differential Settlement at the Bridge–Road Interface Differential settlement that occurs at the interface between the bridge and the road typically results in an abrupt grade change. This phenomenon is referred to as the bump at the end of the bridge. As shown in Figure 1, the bridge is supported on an abutment, while the pavement is Figure 1. Elements of a bridge approach system (Source: Briaud et al. 1997).

Literature Review 11   supported on the approach embankment. An approach slab may be present between the pave- ment and the bridge. Figure 2 illustrates the differential settlement that occurs at the interface between the bridge and the road. This differential settlement occurs because the bridge is sup- ported by an abutment, while the pavement is supported by the embankment, and the settle- ment of the abutment and the settlement at the top of the embankment are different. It has been reported this problem affects 25% of the bridges in the United States (Briaud et al. 1997). State DOTs have conducted numerous research projects over the years to investigate this issue in order to find a solution (Phares et al. 2011, Reza 2013, Abu-Farsakh and Chen 2014). Bridge approach slabs have been used to keep the differential settlement within a tolerable limit, but this strategy does not necessarily prevent a bump from occurring (Hoppe 1999). Hoppe (1999) reported that in many cases the use of an approach slab has resulted in moving the bump from the end of the bridge to the end of the approach slab. This bump causes poor ride quality and can be a potential safety issue for handling vehicles. The bump can cause damage to the bridge from snowplow operations and can also impose dynamic loads on the bridge deck that can result in accelerated deterioration of the bridge deck (Briaud et al. 1997). This bump becomes a maintenance issue to highway agencies as they must perform repairs to improve the ride quality. The formation of the bump is a complex issue and there are many factors that contribute to the formation of the bump. Based on the results of a survey of state highway agencies, Briaud et al. (1997) reported the cause of the bump is related to • Compression of the fill material, • Settlement of the natural soil under the embankment, • Poor construction practices, • High traffic loads, • Poor drainage, Figure 2. Illustration of differential settlement at the bridge–road interface (Source: Hoppe 1999).

12 Practices for Ensuring the Smoothness of Concrete Bridge Decks • Poor fill material, • Loss of fill by erosion, and • Temperature cycles. The results from the same survey indicated the following items are associated with minimiz- ing the bump: • Abutment and embankment on strong natural soil. • An approach slab that is long enough and strong enough. • Well compacted fills or stabilized fills. • Good fill material. • Good drainage. • Low embankments. • An adequate time period between fill placement and paving. • Good construction practices and inspection. • Low truck traffic. The settlement of the embankment will occur over time and the bump will develop over time and will result in a poor ride quality to the road users. Such locations will have a high roughness, resulting in high IRI values. Although bridge approaches and departures are typically locations of high roughness, highway agencies do not typically measure the roughness at these locations to quantify the roughness (Nicks 2020). Nicks (2020) indicated that in order to implement an efficient BMS, it is important to measure and quantify roughness at bridge approaches. If such roughness measurements are available, highway agencies can evaluate these measurements and determine if repairs are needed at such locations to improve the ride quality. The FHWA has developed guidelines to collect profile data at bridge approaches and depar- tures using an inertial profiler (Henderson et al. 2016). These data can be used to compute the MIRI at the bridge approaches and departures, and be used to assess if repairs are needed to improve the ride quality at such locations based on a specified MIRI threshold or a localized roughness value. The profile data can also be analyzed to identify bumps or dips that are outside a specified tolerance. Equipment for Measuring Smoothness of Bridges Types of Equipment This section presents a description of the different types of equipment that are used to deter- mine the smoothness of bridge decks for construction acceptance. The equipment used to determine the smoothness of bridge decks for construction acceptance includes • Straightedge, • Rolling straightedge, • Profilograph, • Walking profiler, and • Inertial profiler. Of this equipment, only walking profilers and inertial profilers can measure the true profile of the pavement, which affects the ride quality. A walking profiler can obtain the actual elevation pro- file of the pavement, while the profile obtained by inertial profilers does not contain any informa- tion about the grade of the road, and only contains features that affect ride quality. A description of the equipment is presented in this section.

Literature Review 13   Straightedge A straightedge consists of a metal beam. Smoothness specifications that are based on a straight- edge indicate the maximum permissible deviation between the bottom of the straightedge and the top of the paved surface when the straightedge is placed on the paved surface. A photograph of a straightedge is shown in Figure 3. The length of a straightedge that is used for construction acceptance can vary based on the specification, but straightedges that are 10 or 12 ft. long are commonly used. A common straightedge-based specification indicates when the straightedge is placed on the paved surface, the maximum deviation from the bottom of the straightedge to the pavement surface should be less than ⅛ in. Rolling Straightedge A rolling straightedge consists of two wheels at the end of a beam, with another wheel at the center. A photograph of a rolling straightedge is shown in Figure 4. The spacing between the wheels at the two ends of a rolling straightedge is typically 10 ft. As the rolling straightedge is pushed along a paved surface, a pointer attached to the center wheel shows the deviation at the center of the straightedge from the datum established by the two wheels at the ends. Smoothness specifications that are based on a rolling straightedge specify a maximum per- missible deviation at the center of the rolling straightedge with respect to the datum established Figure 4. Rolling straightedge (Source: R.W. Perera). Figure 3. Straightedge (Source: R.W. Perera).

14 Practices for Ensuring the Smoothness of Concrete Bridge Decks by the wheels at the ends. For example, the specification could say that the maximum permis- sible deviation at the center of the rolling straightedge from the top of the paved surface should be less than ⅛ in. Some rolling straightedges can be set up so that a can of paint is attached to the device, and locations that do not meet the specified smoothness specification are painted on the paved surface. A rolling straightedge is more efficient than a straightedge as the device can be pushed along the surface at walking speed, and locations that do not meet the specification can be detected for the entire travel path. Profilograph A profilograph consists of a rigid beam or frame with a set of wheels at each end to support it and a center wheel (Scofield 1992). Profilographs can be categorized into either California pro- filographs or Rainhart profilographs, based on the configuration of the support wheels, with the California profilograph being the most common profilograph that is used (Scofield 1992). The California-type profilograph can be a truss type or a beam type. A beam type profilograph is com- monly referred to as an Ames profilograph, as it is manufactured by Ames Engineering (Scofield 1992). Figure 5 shows a photograph of a truss-type California profilograph, while Figure 6 shows a photograph of a beam type (i.e., Ames) profilograph. A set of six wheels is located at each end of the profilograph, with four wheels on one side of the device and the other two wheels on the other side of the device. The support wheels that are at the ends establish a datum, and the wheel at the center measures the deviations at the center of the truss or beam from a datum that is established by the support wheels at the ends. The movements of the center wheel are recorded on a strip chart recorder or in a computer. Profilographs that record the readings on a strip chart recorder are referred to as manual profilo- graphs, while those that record the data in a computer are referred to as computerized profilo- graphs. The profilograph is pushed along the pavement by an operator to obtain measurements and about 2 to 3 miles of pavement can be measured in an hour. The wheel support system in a Rainhart profilograph has both similarities and differences when compared to the California profilograph. Both types of profilograph use a total of 12 sup- port wheels; however, the configurations of the wheels are different for the two profilographs (Scofield 1992). Figure 5. Truss-type California profilograph (Source: Surface Systems and Instruments).

Literature Review 15   The measurements obtained by a profilograph are used to compute a smoothness index called the PI. The measurements obtained by a profilograph can also be analyzed to detect bumps or dips that exceed a specified tolerance. ASTM E1274-18, Standard Test Method for Measuring Pavement Roughness Using a Profilograph, describes the procedures to follow when obtaining measurements with a profilograph. Walking Profiler Walking profiler refers to devices that are pushed along the pavement to measure the eleva- tion profile of a pavement. These devices are pushed along the pavement at a walking speed and can collect profile data at 1-in. intervals. The walking profilers that are commonly used in the United States are manufactured by International Cybernetics Corporation and Surface Systems and Instruments (SSI). The device manufactured by International Cybernetics Corporation is marketed as the SurPRO, while the device manufactured by Surface Systems and Instruments is marketed as the SSI Walking Profiler. A photograph of the SurPRO is shown in Figure 7; this device has two wheels. Figure 8 shows a photograph of the SSI Walking Profiler. This device has three wheels on one side and two wheels on the other side. Both devices have a computer to record the collected data. The data collected by a walking profiler can be used to compute the IRI. The data can also be used to simulate a rolling straightedge or a profilograph on the data using a computer program to produce the outputs that are obtained by an actual rolling straightedge or a profilograph. Inertial Profiler Inertial profilers record the true profile of a pavement surface that affects ride quality. The prin- cipal components of an inertial profiler are the height sensors, accelerometers, the distance mea- suring instrument (DMI), and a computer. The height sensor records the height to the pave ment surface from the sensor. A profiler typically has two height sensors mounted on the vehicle such that each sensor collects data along a wheelpath. An accelerometer is located on top of each height sensor to record the vertical acceleration. A computer program is used to convert the acceleration to vertical displacement, and then the data from the height sensor are combined with the vertical Figure 6. Beam-type California profilograph (Source: Ames Engineering).

16 Practices for Ensuring the Smoothness of Concrete Bridge Decks displacement computed from the accelerometer data to determine the distance to the pavement surface relative to an inertial reference frame. The DMI consists of an encoder fitted to the rear wheel of the vehicle and keeps track of the distance with respect to a reference starting point. A computer program computes the profile at each data recording point using the data recorded by the height sensor and the accelerometer. The computed profile data are recorded in the computer and can be used to compute various roughness indices, including the IRI. Figure 7. SurPRO (Source: R.W. Perera). Figure 8. SSI Walking Profiler (Source: Surface Systems and Instruments).

Literature Review 17   AASHTO has published three standards that address inertial profilers. These standards include the following: • M 328-14, Standard Specification for Inertial Profiler: This standard defines the attributes required for an inertial profiling system. • R 57-14, Standard Practice for Operating Inertial Profiling Systems: This standard describes the procedure for operating and verifying the calibration of an inertial profiling system. • R 56-14, Standard Practice for Certification of Inertial Profiling System: This standard describes a certification procedure for inertial profilers and provides guidance on procedures to certify profiler operators. The data collected by an inertial profiler are used to compute the IRI for each wheelpath. The data can also be used to simulate a rolling straightedge or a profilograph on the data, using a computer program to produce the outputs that are obtained by an actual rolling straightedge or a profilograph. High-Speed Inertial Profilers An inertial profiler that collects data at highway speeds is referred to as a high-speed profiler, where the profiling system is housed on a van or a truck. High-speed profilers can be classified as profilers where the equipment is permanently fixed to the vehicle or as portable profilers. Figure 9 shows a high-speed profiler, where the profiling equipment is permanently fixed to the vehicle. In this profiler, the sensors that collect profile data are housed inside a sensor bar that is attached to the front of the van. The portable profiling equipment consists of a bar that houses the height sensors, accelerometers, and a signal processing unit. The portable system can be mounted on the trailer hitch of a van or a truck, or in the front of a vehicle that is specially con- figured to mount the system. The portable system eliminates the need to have a dedicated vehicle to house the profiler, as this system can be installed in any suitable vehicle. Portable profiling systems have been popular with contractors, as they do not need a dedicated vehicle to house the profiler. When not needed, the profiling system can be taken off the vehicle, and the vehicle can be used for other purposes. Figure 9. High-speed profiler (Source: R.W. Perera).

18 Practices for Ensuring the Smoothness of Concrete Bridge Decks Lightweight Inertial Profiler The term lightweight profiler is used to refer to a light utility vehicle that houses a profiling system. Figure 10 shows a photograph of a lightweight profiler. Typically, the total weight of a lightweight profiler is about 950 lb. without the operator. The primary reason why lightweight profilers were developed was because such equipment could collect profile data on concrete pavements as soon as the pavement could support the weight of the utility vehicle and the opera- tor. A lightweight profiler has all the functionality of a high-speed profiler except that the high- est speed that it can collect data is limited to the maximum speed of the utility vehicle, which is typically 15 to 20 mph. Ride Quality Indices This section describes two ride quality indices. They are the IRI and PI. International Roughness Index The IRI was developed based on the work performed from an experiment in Brazil that was sponsored by the World Bank (Sayers et al. 1986). The true profile of the pavement that includes all the features that impact the ride quality is needed in order to compute the IRI. An inertial profiler or a walking profiler can collect the data that satisfy this requirement. The computation of IRI is based on a mathematical model called a quarter-car model. The quarter car is simulated on the measured profile using a computer program to calculate the suspension deflection (Sayers 1995). The simulated suspension motion is accumulated and then divided by the distance traveled to provide the IRI that has units of slope (e.g., in./mi.). The procedure for calculating the IRI from a longitudinal profile is described in ASTM E1926-08, Standard Practice for Computing International Roughness Index of Roads from Longitudinal Profile Measurements. The IRI is computed for a single wheelpath. As profile data are collected for the left and the right wheelpaths in a lane, the IRI of each wheelpath is computed. The left and the right wheel- path IRI values are averaged to obtain the MIRI that is used to represent the roughness of a roadway and has a strong correlation to the ride quality experienced by highway users (Sayers and Karamihas 1998). Figure 10. Lightweight profiler (Source: Pennsylvania DOT).

Literature Review 19   State DOTs collect data on their highway networks using an inertial profiler, and the MIRI computed from data are stored in their PMS. These data provide information about the rough- ness of the highway network and can be used to select pavement segments for rehabilitation. Many state DOTs use the MIRI in their smoothness specifications for construction acceptance of reconstructed or rehabilitated pavements for both AC and concrete pavements. The average IRI value along a wheelpath of a paved segment provides an overall IRI value for that segment but does not provide any information on how the roughness is distributed within a section. For example, consider a smooth paved segment that has a bump within the segment that causes poor ride quality at the bump. When computing the overall IRI of the paved segment, the high IRI value at this location will be averaged with other areas within the segment that are smooth. Therefore, when looking at this averaged overall IRI value, one will not know that a very rough spot is located within the paved segment. The distribution of IRI within a paved segment can be represented by a continuous IRI plot. The IRI shown at a specific location in a continuous IRI plot is the average IRI over a specified base length centered at that location. Typically, a continuous IRI plot using a 25-ft. base length is used to show the distribution of IRI within a paved segment, and to locate areas of high IRI within a paved segment. Figure 11 shows an example of a continuous IRI plot that is based on a 25-ft. base length that was prepared by the author of this synthesis using profile data collected on a paved surface. As the base length used for the continuous IRI plot is 25 ft., the IRI shown in the plot at a dis- tance of 100 ft. is the average IRI from 87.5 ft. (12.5 ft. before 100 ft.) to 112.5 ft. (12.5 ft. after 100 ft.), where the base length of 25 ft. was used to average the IRI from 87.5 ft. to 112.5 ft. The overall IRI value for the 0.1-mi. long segment shown in Figure 11 is 75 in./mi. The continuous IRI plot shows that the highest IRI within the segment based on a 25-ft. base length occurs at a distance between 110 to 130 ft. and has a value of approximately 140 in./mi. This value is approximately 90% higher than the overall IRI of the segment and results in a poor ride quality at this location. When IRI is used as the ride quality index, localized roughness is typically defined based on continuous IRI using a 25-ft. base length. The continuous IRI is sometimes referred to as moving average IRI. Locations in the plot shown in Figure 11 that are above a specified IRI value are referred to as localized roughness. For example, a specification could indicate that localized rough- ness based on a 25-ft. moving average IRI that has a value of over 120 in./mi. must be corrected. Figure 11. Continuous IRI plot based on a 25-ft. base length.

20 Practices for Ensuring the Smoothness of Concrete Bridge Decks In Figure 11, this specification will mean the localized roughness between 105 and 130 ft. and between 148 and 160 ft. will have to be corrected to a value of less than 120 in./mi. In this synthesis, localized roughness that is based on a 25-ft. continuous IRI will be referred to as 25-ft. moving average IRI. Profile Index The readings obtained by a profilograph are used to compute an index called the PI, which is a measure of roughness. Manual profilographs record the data in a strip chart, and these data must be reduced manually to compute the PI. Computerized profilographs record the data electroni- cally, and the PI is computed from these data using a computer program. To compute the PI, a blanking band is placed on the profile trace obtained from a profilo- graph. When computing the PI values manually, a plastic scale that is 1.7 in. high and 21.12 in. long is used. The length of 21.12 in. represents 528 ft., as the horizontal scale of the profilograph trace is 1:25. At the center of the scale, there is an opaque band whose width varies based on the state DOT specification. Parallel to the opaque band on both sides are five scribed lines that are at 0.1-in. intervals. The blanking band is placed over the profile trace so that the blanking band blanks out as much of the profile as possible. In this position, the deviations above and below the opaque band will be approximately balanced. Thereafter, excursions above and below the blanking band that are called scallops are identified and summed to compute the PI, which is expressed in terms of inches per mile. A computer program performs these computations when the data are recorded in a computer. A PI value is computed for each wheelpath and then averaged to obtain a mean PI value, which is used to judge the ride quality of the paved surface. The placement of a blanking band on the profilograph trace can cover features that affect the ride quality. Therefore, some state DOTs are using what is termed as a zero-blanking band. This means instead of placing a blanking band to cover as much of the profile as possible, a reference line is selected on the profile that will balance deviations above and below this line. Thereafter, the scallops are counted based on this reference line. The PI can also be computed from the data collected by an inertial profiler or a walking profiler. A profilograph simulation is performed on the collected data first using a computer program to provide an output, which is similar to the profile that would have been obtained if an actual profilograph was used to collect the data. Thereafter, the PI is computed for the simulated profilograph trace. ASTM E2955, Standard Practice for Simulating Profilograph Response to Longitudinal Profiles of Traveled Surfaces, describes the procedure to perform a simulation of a profilograph on data collected by an inertial profiler to compute the PI. Locations of bumps are also detected from the profilograph trace. When manually locating bumps from a profilograph trace, a small plastic template with a 1-in. line scribed parallel to the edge and located 0.3 in. to 0.6 in. away, depending on the agency criteria, is used. The 1-in. on the bump template represents a distance of 25 ft., as the horizontal scale of the profilograph printout is 1:25. At high points or peaks in the profile, the template is placed such that the small holes at each end of the line intersect the profile trace to form a chord across the base of the peak. Any portion of the trace extending above the upper edge of the template will indicate a bump. A horizontal line is then drawn at the top of the bump template to denote the bump. In a computerized profilograph, or when a profilograph is simulated on inertial profiler or walking profiler data, this process is performed by a computer program. An example of a document that describes computation of PI and detection of bumps from the data collected by a profilograph is Iowa Test Method 341 (Iowa DOT 2018).

Literature Review 21   Bridge Roughness Studies Evaluation of Roughness of Bridges in Ohio Schleppi (2003) performed a study to investigate the roughness level of bridges in Ohio’s Interstate system. This section presents a summary of the methods used by Schleppi for analyses and the findings that were reported. Schleppi (2003) indicated that Ohio DOT collects profile data on its Interstate system annu- ally and roughness statistics computed from these data for pavements are stored in Ohio DOT’s PMS and submitted to the FHWA to meet HPMS reporting requirements. The HPMS reporting requirements at the time this study was conducted indicated that the submitted data should not include the roughness data for bridge segments. The roughness data stored in Ohio DOT’s pavement management system in 2003 did not include the roughness of bridge segments. Schleppi (2003) indicated when Ohio DOT collects profile data on its highway system using an inertial profiler, the data are marked to indicate the start and the end location of a bridge. When processing the data to compute roughness indices, the data within the marked limit are ignored, so that only the roughness values corresponding to the pavement are computed. How- ever, as profile data collected on the bridges were still available, these data were processed to obtain roughness statistics for the bridges. Two roughness statistics were used for analysis, which were the Half-Car Roughness Index (HRI) and the ride number. Only the results that were obtained for the HRI are presented in this section. HRI has a very high correlation to the MIRI. Sayers and Karamihas (1998) reported that HRI = 0.89 MIRI. Schleppi (2003) indicated that Ohio DOT’s Interstate system consists of 1,330 centerline miles, which translates into 2,660 directional miles. When driving through these 2,660 direc- tional miles, 1,910 bridges would be encountered. Of these encounters, 260 bridges carry traffic in both directions (i.e., a single bridge carries traffic in both directions), while 1,390 bridges carry traffic in only one direction (i.e., separate bridges are provided to carry traffic in each travel direction). The total length of the encountered bridges was 94 mi., with the average length of the bridge being 261 ft. This represented 3.5% of the length of the Interstate system in Ohio. Schleppi (2003) found the average HRI for the Interstate system when the bridges were excluded was 66 in./mi. The average HRI of the bridges on the Interstate system was 162 in./mi. Therefore, on average, bridges were 2.5 times rougher than the pavement. The average HRI of the Interstate system when bridges were also included was 71 in./mi. This study showed if the bridges were also included, the average HRI of Ohio’s Interstate system would increase by 7.6%. Schleppi (2003) reported a Geographical Information System was used to link HRI data on bridges with the information in Ohio DOT’s BMS. With this system, it was possible to click on a location where a bridge was located and see the roughness level of the bridge as well as attributes of the bridge that were stored in the BMS. Schleppi (2003) indicated the procedure used in Ohio DOT’s bridge smoothness specifica- tions in 2003 did not properly account for ride quality. In 2003, Ohio DOT’s smoothness speci- fication was based on using a 10-ft. rolling straightedge, which cannot account for wavelengths in surface profile much longer than its base length of 10 ft. Schleppi also indicated that IRI, which is a commonly used roughness metric that is used to measure ride quality, is sensitive to much longer wavelengths, and therefore the rolling straightedge is accounting for only a small percentage of the wavelengths that are critical to ride quality. Schleppi (2003) concluded that ride quality should be considered in bridge design, bridge construction, and bridge maintenance. Schleppi indicated the development of a new ride-quality- based smoothness specification for bridges, which is different from the method used by Ohio DOT in 2003, should improve the ride quality experienced on Ohio DOT’s highway system.

22 Practices for Ensuring the Smoothness of Concrete Bridge Decks At the annual meeting of the Road Profiler User Group in 2010, Schleppi (2010) presented the data from the Ohio DOT study described in Schleppi (2003) in terms of MIRI. Figure 12 shows a cumulative distribution plot of the MIRI on Ohio’s pavements and bridges in the Inter- state system. Figure 12 was developed by the author of this synthesis using the data used by Schleppi (2010). This plot shows the dramatic difference between MIRI of bridges when com- pared to the MIRI of pavements. Study of Bridge Roughness in Virginia McGhee (2002) performed a study to evaluate the ride quality of bridges in Virginia. In this study, data collected by an inertial profiler were used to evaluate the ride quality and to assess the use of the 10-ft. straightedge in achieving a good ride quality. This section of the report presents a summary of the methods used by McGhee to collect and analyze the data and presents findings from the study. McGhee (2002) indicated bridges located on several major highway corridors were selected for this study, and data were collected at 289 bridges, with 228 of the bridges located on Inter- state highways and 61 bridges being on the primary system. The bridges included in this study were of various ages, of various structural types, and in various service conditions. Data collec- tion was started at least 100 ft. before the start of the bridge and continued over 100 ft. after the end of the bridge. The operator of the profiler entered event marks into the computer during data collection to note the start and the end locations of the bridge deck. MIRI values were then computed for the bridge deck, for 100 ft. before the bridge deck, and for 100 ft. after the bridge deck. At the joint at the start of the bridge deck and at the end of the bridge deck, the MIRI was computed over 10 ft. McGhee (2002) indicated that the method used to ensure smoothness of a bridge deck when constructed in Virginia at the time this study was conducted was based on using a straightedge 10 ft. long. The specification for smoothness stated that the deck surface should be tested with a 10-ft. straightedge and rescreeded as many times as is necessary to obtain a smooth riding surface. The specification stated that in the longitudinal direction, areas showing high spots or depressions more than ⅛ in. over 10 ft. should be struck off or filled with freshly mixed concrete during construction. McGhee (2002) indicated when all bridges were considered, the average MIRI of the bridge deck was 175 in./mi. The average MIRI of 500 projects that had been chosen to receive an AC Figure 12. MIRI distribution of pavements and bridges in Ohio’s Interstate system.

Literature Review 23   overlay in Virginia was 110 in./mi. This analysis showed that the MIRI of the bridges were con- siderably higher than pavements that were selected to receive an AC overlay. The MIRI of the joint at the bridge deck at the start and the end of the bridge was more than 50% higher than that of the bridge deck. McGhee (2002) indicated that in bridges, a camber that has an upward curving hump shape is provided to offset the deflection that will be caused by the addition of bridge superstructure components. The camber is designed to flatten with the additional dead load applied when the superstructure components are constructed. However, a residual camber could be present in the completed structure because of the conservative nature of bridge design. McGhee (2002) performed a theoretical analysis to investigate the effect of camber on MIRI for two cases. In the first case, a single span bridge that was 100-ft. long, with 50 ft. of flat roadway on either end, was considered. In the second case, a bridge with two 160-ft. spans, with 50 ft. of flat roadway at each end, was considered. The maximum camber in both cases was 4 in. The theoretical analysis showed that the MIRI values due to the camber alone for the single-span and two-span bridges were 150 and 108 in./mi., respectively. Therefore, McGhee concluded that the camber in bridges could have a major contribution to the MIRI. In this study, McGhee (2002) also compared the MIRI with the percentage of deck length that violated the straightedge criterion. The percentage of deck length that violated the straightedge criterion was computed using a rolling straightedge simulation on the collected profile data. A general linear relationship was noted between the MIRI and the percentage of deck length that violated the rolling straightedge criterion, with higher MIRI values resulting in higher percentage of deck length that violated the rolling straightedge criterion. McGhee (2002) concluded because of the contribution of the camber to the MIRI, MIRI might not be an appropriate metric for measuring the ride quality of newly constructed bridge decks. McGhee also concluded that the simulation of a 10-ft. straightedge on inertial profile data can be used as a tool for enforcing smoothness of bridge decks by mimicking the 10-ft. straightedge that was in the specification used in Virginia at that time. Study of Bridge Roughness in Illinois Rufino et al. (2001) reported a study that collected profile data at 20 newly reconstructed or rehabilitated bridges in and around Springfield, Illinois, to develop a preliminary smoothness specification for bridges in Illinois. Rufino et al. (2001) indicated the developed specification was a preliminary specification, was not intended for construction, and will need additional develop- ment and field testing to improve it. A summary of the methods used by Rufino et al. to collect and analyze the data and the findings from the study are presented below. Rufino et al. (2001) reported that profile data at the bridges were collected with a lightweight profiler. Data collection started approximately 150 ft. before the start of the entry-approach slab for the bridge and ended approximately 150 ft. after the exit approach slab at the other end of the bridge. The length of the approach slab at bridges in Illinois was 30 ft. Rufino et al. (2001) reported that the right wheelpath IRI was computed from the profile data for the segment encompassing the bridge deck and the approach slab at the start and the end of the bridge deck. Rufino et al. (2001) reported the average right wheelpath IRI of the bridge deck and the approaches for the evaluated bridges was 171 in./mi. This value was between 130 and 170 in./mi. for 65% of the evaluated bridges, with the rest having values higher than 180 in./mi., with 25% of evaluated bridges having a value between 190 and 200 in./mi. The statewide average IRI of pave- ments in Illinois was 100 in./mi. For pavements, the IRI range for the first quartile was between 40 and 73 in./mi., for the second quartile it was between 74 and 100 in./mi., for the third quartile

24 Practices for Ensuring the Smoothness of Concrete Bridge Decks it was between 101 and 137 in./mi., and for the fourth quartile it was between 137 and 200 in./mi. Based on the IRI values of the evaluated bridges, which were all reconstructed or rehabilitated recently, Rufino et al. (2001) reported that IRI of bridges is considerably higher than the IRI of pavements. Resources Related to Smoothness Over the past several years, the FHWA has undertaken several endeavors to improve the understanding of smoothness/roughness issues, develop documents related to pavement smooth- ness, develop software for analyzing profile data, and develop training courses for profiler opera- tors. This section describes a document, software, and training course that have been developed through FHWA initiatives. The Little Book of Profiling The Little Book of Profiling (Sayers and Karamihas 1998) presents basic information about measuring and interpreting road profiles. This document describes how profilers work, what can be done with measurements obtained by profilers, and procedures to follow to eliminate errors during profile data collection. ProVAL Software A free software called ProVAL that can be downloaded from the web is available for the evalu- ation and analysis of profile data collected by inertial profilers and walking profilers (ProVAL 2021). ProVAL was created in 2001 with funding from the FHWA as a product of the Long- Term Pavement Performance Program and is being updated continually to add new features, with funding for these efforts being provided through a pooled fund study. This software will be useful for state DOTs that plan to implement or are already implementing a ride-quality specification based on data collected by inertial profilers for bridge decks. ProVAL has a variety of features that can be used to view data collected by inertial profilers, as well as to compute roughness indices such as IRI. ProVAL can also perform a rolling straightedge simulation or a profilograph simulation on data collected by inertial profilers or walking profilers. ProVAL also has a feature to detect localized roughness when the IRI is used as the index, and to perform a grinding simulation to correct the pavement if a bump is causing the rough spot. NHI Course The FHWA has developed a course titled Pavement Smoothness: Use of Inertial Profilers for Construction Quality Control which is offered through the National Highway Institute. This course provides information on procedures to be followed for collecting accurate profile data with inertial profilers and how to analyze the collected data. Although this course is intended for pavement applications, this course would be beneficial to state DOTs that are currently using or planning to use a ride quality specification based on IRI for bridge decks. Summary State DOTs are required to perform biennial inspection of highway bridges and submit data annually to the FHWA in a suitable format to be entered into the NBI database. The bridge deck is rated on a scale from 0 to 9, where 0 represents a failed condition and 9 represents an excel- lent condition. However, the ride quality of the bridge deck is not considered when rating the bridge deck.

Literature Review 25   Cost-effective bridge preservation activities are eligible for federal funding, but not routine maintenance activities. The FHWA has developed a bridge preservation guide that describes activities that can be performed to preserve a bridge. However, this guide does not address improving or maintaining the ride quality of the bridge deck. Some activities that are performed for bridge preservation (e.g., placing an AC or a concrete overlay) would improve the ride quality of the bridge deck. Differential settlement that occurs at the interface between the bridge and the road typically results in an abrupt grade change. This phenomenon is referred to as the bump at the end of the bridge, and several research studies have been performed to investigate this issue in order to find a solution. Although bridge approaches and departures are typically locations of high rough- ness, highway agencies do not typically measure the roughness at these locations to quantify the roughness. If such roughness measurements were available, highway agencies could evaluate these measurements and determine if repairs are needed at such locations to improve the ride quality. The FHWA has developed guidelines to collect profile data at bridge approaches and departures using an inertial profiler. The data collected by an inertial profiler can be used to compute the MIRI at the bridge approaches and departures and can be used to assess if repairs are needed to improve the ride quality at such locations, based on a specified MIRI threshold or a localized roughness value. The types of equipment that are used to determine the smoothness of bridge decks for con- struction acceptance are the straightedge, the rolling straightedge, the profilograph, the walking profiler, and the inertial profiler. When using a straightedge, the maximum deviation from the bottom of the straightedge to the top of the paved surface is specified. When using a roll- ing straightedge, the maximum deviation at the center of the rolling straightedge based on the datum established by the wheels at the ends is specified. When the profilograph is used, an allowable PI value computed from the collected data and a criterion for the maximum value for bumps and dips that are detected from the data are specified. When an inertial profiler or a walking profiler is used, an allowable MIRI value and an allowable localized roughness value based on 25-ft. moving average IRI can be specified. The data from an inertial profiler or a walking profiler can also be used to perform either a profilograph simulation or a rolling straightedge simulation using a computer program to obtain outputs from the device selected for simulation. Past roughness studies in Illinois, Ohio, and Virginia have shown that bridges have a higher roughness level than pavements. A study performed on the Interstate system in Ohio showed that, on average, bridges are 2.5 times rougher than pavements.