Standard Definitions for Common Types of Pavement Cracking (2020)

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.

Standard Definitions for Common Types of Pavement Cracking 4 C H A P T E R 2 Literature Review and Agency Survey Introduction This chapter documents a comprehensive literature review regarding pavement cracking data collection and definitions and survey results from current SHAs’ practices. Seven national cracking protocols and 34 state protocols were reviewed and summarized. Example cracking protocols at the national and SHA level are provided as representations of manual and automated cracking surveys. Subsequently, an online survey of 50 SHAs was performed to obtain current practices on cracking data collection, processing, cracking definitions, wheel path definitions, and also expectations and comments for the next generation of cracking protocols for fully automated systems. The survey results from SHAs were summarized to determine the state-of- the-practice in cracking data collection and management. Cracking Data Collection Methods The traditional manual survey and recently developed automated surveys are predominant ways for different SHAs to obtain pavement cracking data, and introduced as follows. Manual Survey There are two basic methods for conducting manual pavement cracking data collection, walking survey and windshield survey (Timm and McQueen 2004). Walking surveys are completed by a rater trained according to an agency’s distress identification protocol (Timm and McQueen, 2004). The rater walks along a pavement section and fills out a pavement cracking survey form that describes the amount, extent, and severity of each cracking category visually present on the pavement surface. Some agencies conduct a 100% survey, while others conduct surveys on a sample of the road network (Henning and Morrow 2017). Walking surveys can generate precise cracking data, and hence have been applied for obtaining ground truth of pavement cracking data (Timm and McQueen 2004). Windshield surveys are completed by driving along the road or on the shoulder of the road at low speed and estimating the extent and severity of the various distress (Sivaneswaran et al. 2004). This method allows for a greater amount of coverage in less time; however, the quality of the pavement distress data can be compromised. Manual surveys are slow, pose potential safety problems, may require traffic control, and can result in subjective data which may be inconsistent from year to year and from rater to rater.

Standard Definitions for Common Types of Pavement Cracking 5 Automated Survey With the increasing traffic levels on the highways and the need to use network level pavement cracking data, SHAs are moving toward automated surveys to record pavement surface characteristics via digital images. Digital imaging-based automated cracking surveys capture pavement surface images as streams of electronic bits and store them on an electronic medium for pavement cracking condition interpretation (McGhee 2004). In recent years, a common data collection method uses a combination of lasers and digital cameras to capture both two- dimensional (2D) and 3D images. 3D laser scanning and imaging systems, which typically contain data acquisition, data storage, data display, and processing subsystems, have rapidly gained popularity (Wang and Smadi 2011). The 3D laser triangulation imaging technology has been widely applied for the inspection of manufactured products for decades (Figure 1). Since pavement surface defects have unique 3D characteristics, laser triangulation techniques can be used to collect pavement surface images and evaluate surface defects, such as cracking. Figure 1 3D Laser triangulation imaging technology. Currently, several vendors have implemented 3D laser triangulation techniques for pavement surface data collection with height information at pixel points. The Laser Crack Measurement System™ (LCMS™) by INO of Quebec, Canada, has been used for a few years by Applus, Australian Road Research Board (ARRB) Group, Dynatest, Fugro Roadware, International Cybernetics, and Mandli for pavement image data collection. Other 3D imaging systems include the PaveVision3D system by the WayLink Systems Corporation, the Road excellent automated logging system (Real) from the PASCO Corporation, the 3D system for texture (VTexture) and rutting (VRUT) developed by the Texas DOT, and Pathway Services’ 3D pavement system (http://www.ino.ca/en/, Wang and Smadi 2011, Attoh-Okine and Adarkwa 2013, Huang et al. 2013, www.pathwayservices.com). Agencies either purchase these automated systems to collect and process crack data in-house or rely on outside contractors or vendors for this work. Automated systems are able to acquire pavement surface imaging data at highway speeds without requiring traffic control. These systems measure pavement conditions and distresses

Standard Definitions for Common Types of Pavement Cracking 6 including cracking, roughness, rutting, faulting, texture, and geometric characteristics. Automated surveys can greatly reduce errors associated with transferring data from paper forms used in the field to computer systems for analysis. Other benefits of automated cracking surveys include improved safety for survey crews, faster data collection and processing speed, and more objective and comparable survey results from year to year (McGhee 2004, Timm and McQueen 2004, Wang and Gong 2006, Wang et al. 2011, Attoh-Okine and Adarkwa 2013). Cracking Data Protocols To gain a thorough understanding of current cracking data collection practices, processing, and reporting, an extensive literature review of pavement crack protocols was performed. Cracking protocols developed by AASHTO, ASTM, LTPP, National Park Service (NPS), and the Surface Condition Assessment for the National Network of Roads (SCANNER) were reviewed. Subsequently, pavement crack protocols for 34 SHAs were explored and summarized. Protocols at the National Level Pavement distress protocols were developed at the national level for various purposes. Currently, seven cracking protocols are widely used, as summarized in Table 1. The AASHTO, NPS, and SCANNER protocols can be implemented for automated cracking data collection while the ASTM Standards and the LTPP Distress Manual were developed for traditional manual or windshield surveys. The cracking data defined in these protocols are not consistent in terms of categories, definitions, or severity levels. The protocols suitable for automated technology specify fewer cracking categories whereas the protocols for manual survey report more cracking types for different pavement surface types, including asphalt pavement, portland cement concrete pavement (PCC), jointed plain concrete pavements (JPCP), jointed reinforced concrete pavement (JRCP), and continuously reinforced concrete pavements (CRCP). Detailed summaries of these cracking protocols are provided in Appendix A. The LTPP Distress Manual and the AASHTO R 85-18 are the most widely implemented protocols for manual and automated cracking surveys. Therefore, detailed crack information for these protocols are summarized in Tables 2 and 3. Distress Identification Manual for the Long-Term Pavement Performance Program The LTPP Distress Manual illustrates distress types and measurements for asphalt pavements, JPCP, and CRCP (FHWA 2014). The LTPP Distress Manual is widely used in the U.S. and worldwide. Some SHAs have adopted the LTPP Distress Manual in its entirety, while others have modified it for inclusion in state-specific pavement distress manuals. As noted previously, the crack definitions and evaluations covered in the LTPP Distress Manual were developed for manual surveys.

Standard Definitions for Common Types of Pavement Cracking 7 Table 1 Summary of Crack Protocols Protocols Automation Pavement and Cracking Types AASHTO R 85-18 Yes Asphalt: longitudinal, transverse, pattern, and other cracking. AASHTO R 55-10 Yes Asphalt: wheel path and non-wheel path cracking. ASTM D6433-16 for Highway No  Asphalt: alligator, block, edge, joint reflection, longitudinal/transverse, and slippage cracking.  PCC: corner break, “D”, linear on non-reinforced slabs, linear on reinforced slabs, shrinkage cracking, and divided slab. ASTM D5340-12 for Airfield No  Asphalt: alligator, block, joint reflection or longitudinal/transverse, and slippage cracking.  PCC: corner break, cracks (longitudinal, transverse, and diagonal), “D”, shattered slab/intersecting cracks, and shrinkage cracking. FHWA LTPP No  Asphalt: fatigue, block, edge, reflection, longitudinal, and transverse cracking.  JPCP: corner break, “D”, longitudinal, and transverse cracking.  CRCP: “D”, longitudinal, and transverse cracking. FHWA NPS Yes  Asphalt: alligator, longitudinal, and transverse cracking.  PCC: longitudinal and transverse cracking. SCANNER Yes Any Pavement Type: carriageway cracking intensity and intensity of wheel track cracking. Table 2 FHWA LTPP Cracking Protocol Crack Type Severity Levels Low Medium High Asphalt Pavement Fatigue Cracking Cracks with little connection are not spalled Interconnected cracks may be slightly spalled Cracks are severely spalled interconnected Block Cracking Mean width ≤ 1/4 in. (6 mm) 1/4 in. (6 mm) < Mean width ≤ 3/4 in. (19 mm) Mean width > 3/4 in. (19 mm) Edge Cracking Cracking with no breakup Cracking with some breakup Considerable breakup Longitudinal Cracking Mean width ≤ 1/4 in. (6 mm) 1/4 in. (6 mm) < Mean width ≤ 3/4 in. (19 mm) Mean width > 3/4 in. (19 mm) Reflection Cracking Mean width ≤ 1/4 in. (6 mm) 1/4 in. (6 mm) < Mean width ≤ 3/4 in. (19 mm) Mean width > 3/4 in. (19 mm) Transverse Cracking Mean width ≤ 1/4 in. (6 mm) 1/4 in. (6 mm) < Mean width ≤ 3/4 in. (19 mm) Mean width > 3/4 in. (19 mm)

Standard Definitions for Common Types of Pavement Cracking 8 Table 2 FHWA LTPP Cracking Protocol (continued) Crack Type Severity Levels Low Medium High JPCP Pavement Corner Breaks Spalled crack < 10% crack length; no faulting Spalled crack > 10% crack length; or faulting ≤ 1/2 in. (13 mm) Spalled crack > 10% crack length; or Faulting > 1/2 in. (13 mm) “D” Cracking Tight cracks with no loose pieces Well-defined cracks with some loose pieces Well-developed cracks with significant loose pieces Longitudinal Cracking Mean width ≤ 1/8 in. (3 mm) with no spalling 1/8 in. (3 mm) < Mean width ≤ 1/2 in. (13 mm); or spalling ≤ 3 in. (75 mm); or faulting ≤ 1/2 in. (13 mm) Mean width > 1/2 in. (13 mm); or spalling > 3 in. (75 mm); or faulting > 1/2 in. (13 mm) Transverse Cracking Mean width ≤ 1/8 in. (3 mm) with no spalling 1/8 in. (3 mm) < Mean width ≤ 1/4 in. (6 mm); or spalling ≤ 3 in. (75 mm); or faulting ≤ 1/4 in. (6 mm) Mean width > 1/4 in. (6 mm); or spalling > 3 in. (75 mm); or faulting > 1/4 in. (6 mm) CRCP Pavement “D” Cracking Tight cracks with no loose pieces Well-defined cracks with some loose pieces Well-developed cracks with significant loose pieces Longitudinal Cracking Mean width ≤ 1/8 in. (3 mm) with no spalling 1/8 in. (3 mm) < Mean width ≤ 1/2 in. (13 mm); or spalling ≤ 3 in. (75 mm); or faulting ≤ 1/2 in. (13 mm) Mean width > 1/2 in. (13 mm); or spalling > 3 in. (75 mm); or faulting > 1/2 in. (13 mm) Transverse Cracking Crack spalling ≤ 10% of crack length Crack spalling > 10% and ≤ 50% of crack length Crack spalling > 50% of crack length Table 3 AASHTO R 85-18 Cracking Protocol Crack Type Definitions Longitudinal Crack A crack at least 12 in. (0.3 m) long and with a crack orientation between and including +20 and -20 degrees relative to the lane centerline. Transverse Crack A crack at least 12 in. (0.3 m) long and with a crack orientation between and including 70 and 110 degrees relative to the lane centerline. Pattern Crack A crack that is part of a network of cracks that form an identifiable area or grouping of cracks. Other Crack All cracks not identified as longitudinal, pattern, or transverse.

Standard Definitions for Common Types of Pavement Cracking 9 AASHTO R 85-18 Protocol The AASHTO R 85-18 outlines the procedures for quantifying cracking at the network level for asphalt pavement surfaces utilizing automated methods (AASHTO 2018). Only cracks detected in pavement images with dimensions wider than 0.04 in. (1 mm) and 1 in. (25 mm) long are reported in this protocol. Five measurement zones across a full lane are included in the wheel path definition, as shown in Figure 2. The sum of the length in feet (meters) and the average width in inches (millimeters) for each cracking category are summarized and reported as the cracking extent and severity for each measurement zone and summary section (0.01 mi [0.015 km] or less). The extent and severity of cracking in Zones 2 and 4 typically reflect the impact of traffic loading. Figure 2 AASHTO R 85-18 wheel path definition (AASHTO 2018). SHA Protocols A comprehensive literature review and SHA survey were performed. Table 4 summarizes the SHA pavement cracking protocols by pavement type. In particular, nine SHAs have developed protocols for current or future applications of automated pavement condition survey technologies, including Alabama, California, Delaware, Florida, Idaho, Maryland, Pennsylvania, Vermont, and Virginia. Detailed summaries of SHA cracking protocols are provided in Appendix A.

Standard Definitions for Common Types of Pavement Cracking 10 Table 4 Summary of Crack Protocols by SHAs Agency Automated Pavement and Cracking Types Alabama Yes Project Level  Asphalt: fatigue, block, edge, longitudinal, and transverse cracking.  JPCP & CRCP: corner break, “D”, longitudinal, and transverse cracking. Network Level  All Pavements: transverse, load associated, and non- load associated cracking. California Yes  Asphalt: alligator A, alligator B, block, edge, longitudinal, and transverse cracking.  JPCP: corner break, longitudinal, transverse, and third stage cracking.  CRCP: longitudinal and transverse cracking. Colorado No  Asphalt: fatigue, block, edge, longitudinal, reflective, and transverse cracking.  PCC: corner break, “D”, longitudinal, and transverse cracking. Delaware Yes  Asphalt: fatigue, block, non-wheel path, and transverse cracking.  Composite: fatigue, block, joint reflective, and non- wheel path longitudinal cracking.  PCC: map and slab cracking.  Surface treated: block, edge, fatigue, non-wheel path longitudinal, and transverse cracking. Florida Yes  Asphalt: class IB, class II, and class III cracking.  PCC: transverse, longitudinal, and corner cracking. Georgia No  Asphalt: load, block/transverse, and reflection cracking. Idaho Yes  Asphalt: longitudinal and transverse, alligator, block, and edge cracking.  PCC: transverse slab, corner break, and meander cracking. Illinois No  Asphalt: alligator, block, transverse/reflective cracking, longitudinal, and edge cracking.  PCC: “D”, transverse, longitudinal, map cracking, and scaling. Indiana No  Asphalt: fatigue, reflection, longitudinal, thermal/transverse, block, and edge cracking.  PCC: corner breaks, transverse, longitudinal, spalling of joints and cracks, and “D” cracking.

Standard Definitions for Common Types of Pavement Cracking 11 Table 4 Summary of Crack Protocols by SHAs (continued) Agency Automated Pavement and Cracking Types Kansas No  Asphalt: fatigue, transverse, and block cracking.  PCC: joint distress. Kentucky No  Asphalt: fatigue and other cracking.  PCC: other cracking. Louisiana No  Asphalt: alligator, longitudinal, and transverse cracking.  JPCP: longitudinal and transverse cracking.  CRCP: longitudinal, transverse, and map cracking. Maryland Yes  All: longitudinal and transverse cracking. Massachusetts No  All: fatigue, longitudinal, and transverse cracking. Michigan No  Asphalt: fatigue, block, longitudinal, and transverse cracking.  Composite: reflective cracking and joints.  PCC: longitudinal and transverse cracking. Minnesota No  Asphalt: alligator, multiple, longitudinal joint, longitudinal, and transverse cracking.  JPCP: “D”, transverse, and longitudinal joint cracking.  CRCP: “D” and transverse cracking. Montana No  Asphalt: alligator and miscellaneous cracking.  PCC: miscellaneous cracking. Nebraska No  Asphalt: alligator, edge, longitudinal, grid block, and transverse cracking.  PCC: slab cracking I and slab cracking II. New Jersey No  Asphalt: multiple/alligator, longitudinal, and transverse cracking.  PCC: multiple/alligator cracking. New Mexico No  Asphalt: alligator/fatigue, longitudinal, and transverse cracking. New York No  Asphalt: alligator cracking.  APC: alligator cracking. North Carolina No  Asphalt: alligator and transverse cracking. Ohio No  Asphalt: wheel track, block & transverse, longitudinal, edge, and thermal cracking.  Composite: transverse (unjointed base), joint reflection (jointed base), intermediate transverse (jointed base), and longitudinal cracking.  JPCP: transverse, longitudinal, and corner breaks.  CRCP: transverse and longitudinal cracking.

Standard Definitions for Common Types of Pavement Cracking 12 Table 4 Summary of Crack Protocols by SHAs (continued) Agency Automated Pavement and Cracking Types Oklahoma No  Asphalt: fatigue, transverse, and miscellaneous cracking.  JPCP: corner Break, “D”, longitudinal or transverse cracking.  CRCP: longitudinal cracking. Oregon No  Asphalt: fatigue, transverse and block, and longitudinal cracking.  JPCP: corner break, corner crack, longitudinal, and transverse cracking.  CRCP: longitudinal and transverse cracking. Pennsylvania Yes  Asphalt: fatigue, transverse, and miscellaneous cracking.  PCC: longitudinal and transverse cracking. South Dakota No  Asphalt: fatigue, block, and transverse cracking.  PCC: corner break, “D”, CRCP block, and longitudinal cracking. South Carolina No  Asphalt: fatigue, longitudinal, and transverse cracking.  PCC: spalled cracks, shattered slabs, and slabs with longitudinal cracking. Texas No  Asphalt: alligator, block, longitudinal and transverse cracking.  JPCP: slabs with longitudinal cracks, shattered slabs, failed joints, and cracks.  CRCP: spalled cracks and transverse cracking. Utah No  Asphalt: wheel path, longitudinal (non-wheel path), transverse, and block cracking.  PCC: longitudinal, transverse, and corner breaks. Vermont Yes  All: transverse, wheel path, and miscellaneous non-wheel path cracking. Virginia Yes  Asphalt: transverse, longitudinal, longitudinal lane joint, and alligator cracking.  APC: reflective cracking.  JPCP: corner breaks, transverse, and longitudinal cracking.  CRCP: transverse, cluster, and longitudinal cracking. Washington No  Asphalt: alligator, longitudinal, transverse, and block cracking.  PCC: cracking. Wisconsin No  Asphalt: transverse, longitudinal, reflection, block, and alligator cracking.  Composite: reflective cracking.  PCC: transverse slab, “D”, corner breaking, and meander cracking.

Standard Definitions for Common Types of Pavement Cracking 13 Distress rating protocols from Oregon DOT and Vermont Agency of Transportation (VTrans) are selected as the example protocols to illustrate manual and automated cracking surveys. The cracking information defined in these two protocols are summarized in Tables 5 and 6. Vermont Agency of Transportation The VTrans Asset Management and Performance Bureau Roadway Condition Assessment Collection Scope of Work describes the condition assessment details used to analyze the pavement condition and provide input into the agency’s PMS for condition forecasting, budgeting, and project programming (VTrans 2017). Pavement cracking, roughness, and rutting are collected and reported. Cracking distress is collected in terms of severity and extent and aggregated into 0.05 mi. (0.08 km) sections. Automated identification and classification of cracks are performed with the collected pavement images, with a minimum crack detection resolution of 0.04 in. (1 mm). Distress types include transverse cracking, longer than 1 ft. (0.3 m), for each severity level, wheel path cracking (left and right wheel paths), and miscellaneous non-wheel path cracking (left edge, center, and right edge of the lane). Table 5 Cracking Protocol for VTrans Crack Type Severity Levels Low Medium High Transverse Cracking Mean width ≤ 1/8 in. (3 mm). 1/8 in. (3 mm) < Mean width ≤ 1/4 in. (6 mm). Mean width > 1/4 in. (6 mm). Wheel Path Cracking Mean width ≤ 1/8 in. (3 mm) or a few interconnected cracks with no developed pattern. 1/8 in. (3 mm) < Mean width ≤ 1/4 in. (6 mm) or interconnected cracks forming a pattern < 50% of the wheel path width. Mean width > 1/4 in. (6 mm) or interconnected cracks forming a pattern > 50% of the wheel path width. Miscellaneous Non-Wheel Path Cracking Mean width ≤ 1/8 in. (3 mm) or a few interconnected cracks with no developed pattern. 1/8 in. (3 mm) < Mean width ≤ 1/4 in. (6 mm) or interconnected cracks forming a pattern < 50% of the appropriate edge or lane center width. Mean width > 1/4 in. (6 mm) or interconnected cracks forming a pattern > 50% of the appropriate edge or lane center width. Oregon Department of Transportation The Oregon DOT Pavement Distress Survey Manual identifies and quantifies the amount and severity of surface distress for rating asphalt pavements, JPCP, and CRCP (ORDOT 2010). Two- person trained crews conduct condition surveys according to the distress identification procedures in a slow-moving vehicle operating on the adjacent shoulder.

Standard Definitions for Common Types of Pavement Cracking 14 Table 6 Cracking Protocol for Oregon DOT Crack Type Severity Levels Low Medium High Asphalt Pavements Fatigue Cracking Cracks with little connection are not spalled. Interconnected cracks may be slightly spalled. Cracks are severely spalled and interconnected. Transverse and Block Cracking Mean width ≤ 1/4 in. (6 mm) or sealed crack in good condition. 1/4 in. (6 mm) < Mean width ≤ 3/4 in. (19 mm) and adjacent low severity random cracking. Mean width > 3/4 in. (19 mm) and moderate to high severity random cracking. Longitudinal Cracking Mean width ≤ 1/4 in. (6 mm) or sealed crack in good condition. 1/4 in. (6 mm) < Mean width ≤ 3/4 in. (19 mm) and adjacent low severity random cracking. Mean width > 3/4 in. (19 mm) and moderate to high severity random cracking. JPCP Pavements Corner Break Mean width ≤ 1/8 in. (3 mm) with no spalling. 1/8 in. (3 mm) < Mean width ≤ 1/2 in. (13 mm); or spalling ≤ 3 in. (75 mm); or faulting ≤ 1/2 in. (13 mm). Mean width > 1/2 in. (13 mm); or spalling > 3 in. (75 mm); or faulting > 1/2 in. (13 mm). Corner Crack Tight cracks with no loose pieces. Well-defined cracks with some loose pieces. Well-developed cracks with significant loose pieces. Longitudinal Cracking Mean width ≤ 1/8 in. (3 mm) with no spalling. 1/8 in. (3 mm) < Mean width ≤ 1/2 in. (13 mm); or spalling ≤ 3 in. (75 mm); or faulting ≤ 1/2 in. (13 mm). Mean width > 1/2 in. (13 mm); or spalling > 3 in. (75 mm); or faulting > 1/2 in. (13 mm). Transverse Cracking Mean width ≤ 1/8 in. (3 mm) with no spalling. 1/8 in. (3 mm) < Mean width ≤ 1/4 in. (6 mm); or spalling ≤ 3 in. (75 mm); or faulting ≤ 1/4 in. (6 mm). Mean width > 1/4 in. (6 mm); or spalling > 3 in. (75 mm); or faulting > 1/4 in. (6 mm). CRCP Pavements Longitudinal Cracking Mean width ≤ 1/8 in. (3 mm) with no spalling. 1/8 in. (3 mm) < Mean width ≤ 1/2 in. (13 mm); or spalling ≤ 3 in. (75 mm); or faulting ≤ 1/2 in. (13 mm). Mean width > 1/2 in. (13 mm); or spalling > 3 in. (75 mm); or faulting > 1/2 in. (13 mm). Transverse Cracking Crack spalling ≤ 10% of crack length. Crack spalling > 10% and ≤ 50% of crack length. Crack spalling > 50% of crack length.

Standard Definitions for Common Types of Pavement Cracking 15 Based on the comprehensive literature review, the cracking protocols used by SHAs can be highly diversified and, at times, are not comparable from agency to agency in terms of cracking categories, definitions, severity levels, etc. For example, block cracking is included in the LTPP Distress Manual and Oregon DOT surveys, but excluded in AASHTO R 85-18 and VTrans’ protocol. In addition, the crack width thresholds to define the severity levels are not consistent between protocols or manuals. For instance, VTrans defines low, medium, and high severity transverse cracking as mean width < 1/8 in. (3 mm), 1/8 to 1/4 in. (3 to 6 mm), and > 1/4 in. (6 mm), respectively. However, the Oregon DOT defines it as mean width < 1/4 in. (6 mm), 1/4 to 3/4 in. (6 to 19 mm), and > 3/4 in. (19 mm), respectively. Survey of State Highway Agencies An online survey of SHAs was conducted to obtain information and insight regarding cracking data collection practices, crack data processing and reporting, desired crack data usage, data quality and variations, and desired improvement of cracking data and data collection practices. The final questionnaire was prepared using SurveyMonkey® and the survey link was sent to 50 SHA representatives in pavement management via email in April 2018, many of whom are members of the Technical Section 5a of the AASHTO Committee on Materials and Pavements. The survey questionnaire was organized into five broad areas (Appendix B):  Part I: Cracking data collection, processing, and common issues.  Part II: Cracking definitions including transverse, longitudinal, alligator/fatigue, block, edge, and “D” cracking, corner breaks, and other cracking data.  Part III: Wheel path definitions.  Part IV: AASHTO R 85 applications.  Part V: General comments. By May 30, 2018, a total of 38 SHA responses were received from Alaska, California, Colorado, Connecticut, Delaware, Florida, Illinois, Iowa, Kansas, Kentucky, Louisiana, Maine, Maryland, Massachusetts, Michigan, Mississippi, Missouri, Nevada, New Hampshire, New Jersey, New Mexico, New York, North Carolina, North Dakota, Ohio, Oregon, Pennsylvania, Rhode Island, South Carolina, South Dakota, Tennessee, Texas, Utah, Vermont, Virginia, Washington, West Virginia, and Wyoming (Figure 3). Due to some incomplete survey answers of certain questions, the total number of survey respondents may not be consistent among the different categories presented in the figures below.

Standard Definitions for Common Types of Pavement Cracking 16 Figure 3 Map of survey respondents with states in red as responding states. Part I: Cracking Data Collection, Processing, and Common Issues As shown in Figure 4a, 13, 10, 9, and 30 of the 39 responses, respectively, use manual surveys (including windshield surveys), semi-automated (data collection at highway speed, substantial manual editing of automated results) methodologies, 2D and 3D image-based automated technologies for cracking data collection and processing (note: some SHAs conduct multiple survey types). In summary, 2D and 3D image-based automated technologies are becoming the dominant approach for cracking data collection and processing. Twenty-four of the responding SHAs collect and process crack data in-house while the remaining SHAs rely on contractors (Figure 4b). SHAs have implemented AASHTO R 85, ASTM D6433, LTPP Distress Manual, HPMS Field Manual, and State-Specific cracking protocols, 17, 2, 13, 20, and 22, respectively, to define, analyze, and report pavement cracking data (Figure 4c). Figure 4d illustrates 25 SHAs conduct quality assurance/quality control (QA/QC) on automated cracking analysis. Almost all the respondents collect transverse, longitudinal, and alligator/fatigue cracking (Figure 4e). Crack severity levels are rated based on the computed average crack width, the highest severity within that length, and the predominant crack width, 16, 6, and 7, respectively. In addition, 9 of the respondent SHAs do not depend on crack width to rate crack severity levels.

Standard Definitions for Common Types of Pavement Cracking 17 (a) Data collection and processing methods. (b) Data collection and processing personnel. (c) Data collection and processing protocols. (d) QA/QC on automated cracking analysis results. (e) Collection of different cracking data. (f) Cracking severity levels. Figure 4 Pavement cracking data collection and processing. Part II: Cracking Definitions In this section, cracking definitions are summarized for transverse, longitudinal, alligator/fatigue, block, edge, “D” cracking, corner breaks, and other cracking data. The survey results are summarized to demonstrate the state-of-the-practice for each crack type. 13 10 9 30 Manual (including windshield survey) Semi‐automated Automated: 2D Images Automated: 3D Images 2424 Agency personnel Outside contractor 17 2 13 20 22 AASHTO PP 67 ASTM Designation D6433 LTPP Distress Identification Manual HPMS Field Manual State Specific 25 12 Yes No 0% 20% 40% 60% 80% 100% 16 6 7 9 Per average crack width Per the highest severity Per predominant crack width Others

Standard Definitions for Common Types of Pavement Cracking 18 Transverse Cracking All responding SHAs collect transverse cracking data, 33 measure transverse cracking over the full width of a travel lane, and 33 use crack width to classify the severity levels. As shown in Figure 5a, 8, 3, and 6 SHAs apply 45 ̊ to 90 ̊, 80 ̊ to 100 ̊, and 90 ̊ of angle orientation to classify and distinguish transverse cracking from other cracking types, respectively. As depicted in Figure 5b, 22, 23, 19, and 11 of the SHAs respondents report linear length, number of transverse cracks, number of slabs affected (JPCP only), and other criteria to quantify the extent of transverse cracking, respectively. Figure 5c reveals that transverse cracking with a minimum length of 1 ft. (0.3 m) is required for reporting in 13 of SHAs. As observed in Figure 5d, 11 SHAs use crack width thresholds of 1/4 in. (6 mm) to 1/2 in. (13 mm) to define low, medium, and high severity levels for transverse cracking. (a) Angle orientation. (b) Extent evaluation factors. (c) Minimum length. (d) Crack width thresholds. Figure 5 Transverse cracking. Longitudinal Cracking Of the responding SHAs, 34 collect longitudinal cracking data, 30 distinguish longitudinal cracking based on location (inside or outside wheel path(s)), and 30 use crack width to classify severity levels for longitudinal cracking. As shown in Figure 6a, 20 of the respondents use angular orientation between crack and centerline to classify and distinguish longitudinal cracking from other cracking types. SHAs use a 0 ̊ to 45 ̊ orientation angle and parallel to centerline, 10 14 8 3 6 5 No Yes (45  ͦ ‐ 90  ͦ) Yes (80  ͦ ‐ 100  ͦ) Yes (90  ͦ) Yes (Don't know) 22 23 19 11 Linear length Number of transverse cracks Number of slabs affected (Jointed PCC only) Other 6 13 3 3 2 4 5 No minimum length 1 ft 2 ft 3 ft 4 ft Half lane width 11 9 3 3 7 1/4"~1/2" 1/4"~3/4" 1/4"~1" 1/8"~1/4" Others

Standard Definitions for Common Types of Pavement Cracking 19 and 6, respectively, to define longitudinal cracking. Respondents report linear length, the number of slabs affected (JPCP only), and other criteria to quantify the extent of longitudinal cracking, 30, 15, and 10, respectively (Figure 6b). Twelve of the SHAs do not require a minimum length for longitudinal cracking, while 11 require a minimum length of 1 ft. (0.3 m) (Figure 6c). In addition, 7 and 8 of the SHAs use the crack width thresholds of 1/4 in. (6 mm) to 1/2 in. (13 mm) and 1/4 in. (6 mm) to 3/4 in. (18 mm), respectively, to define low, medium, and high severity levels for longitudinal cracking (Figure 6d). (a) Angle orientation. (b) Extent evaluation factors. (c) Minimum length. (d) Crack width thresholds. Figure 6 Longitudinal cracking. Alligator/Fatigue Cracking SHAs (34) indicated collecting alligator/fatigue cracking data. In addition, 22 prefer to retain alligator/fatigue cracking in distress protocols rather than considering those located in the wheel path(s) as load associated cracking. As shown in Figure 7a, 16 of the SHAs count the portion of transverse or longitudinal cracking in the wheel path(s) as alligator/fatigue cracking data. Moreover, 26 of the respondents quantify the extent of alligator/fatigue cracking based on the affected area (Figure 7b). Additionally, 16 of the responding SHAs report all detected alligator/fatigue cracking of any area, while 6 and 3 only record alligator/fatigue cracking with the area larger than 1 ft2 or with the length longer than 1 ft. (0.3 m), respectively (Figure 7c). 14 10 6 4 No Yes (0  ͦ ‐ 45  ͦ ) Yes (Parallel to centerline) Yes (Don't know) 30 15 10 Linear length Number of slabs affected (Jointed PCC only) Other 12 11 1 2 1 1 1 4 No minimum length 1 ft 2 ft 3 ft 4 ft 5 ft 6 ft Don't know 7 8 3 4 8 1/4"~1/2" 1/4"~3/4" 1/4"~1" 1/8"~1/4" Others

Standard Definitions for Common Types of Pavement Cracking 20 SHAs (24) still would like to maintain the interconnectivity of cracking as the criteria to rate the severity levels for alligator/fatigue cracking (Figure 7d). (a) Consideration of cracks in wheel path(s). (b) Extent evaluation factors. (c) Minimum length or area. (d) Severity evaluation factors. Figure 7 Alligator cracking. Block Cracking SHAs (15) reported collecting block cracking data (Figure 8a). Of these agencies, 13 of the 19 SHA utilize the affected area to quantify the extent of block cracking (Figure 8b). In addition, 15 SHAs indicated a preference to report block cracking separately instead of classifying as either load associated (inside wheel paths) or non-load associated (outside wheel paths) cracking. 1616 Yes No 16 26 8 Linear length Affected area Other 16 6 3 8 No minimum length or area 1 sf 1 ft Other 19 24 7 Crack width Interconnectivity of the cracks Others

Standard Definitions for Common Types of Pavement Cracking 21 (a) Block cracking collection. (b) Extent evaluation of block cracking. Figure 8 Block cracking. Edge Cracking Figure 9a illustrates that 13 of the 35 SHA respondents collect edge cracking data. Moreover, 10 of the 15 SHA respondents use the linear length to quantify the extent of edge cracking (Figure 9b). In addition, crack width (10) and the level of breakup or loss of materials within cracking (8) are the two primary factors to rate the severity levels for edge cracking. (a) Edge cracking collection. (b) Extent evaluation of edge cracking. Figure 9 Edge cracking. “D” Cracking “D” cracking data on concrete pavements is collected by 12 of 35 SHA respondents. SHAs indicated (6 of the 12 respondents) using the number of slabs affected as the indicator to quantify the extent of “D” cracking (Figure 10a). For severity level, 7 of the 12 SHA respondents use the developed patterns and amount of loose or missing materials (Figure 10b). 15 19 Yes No 9 13 2 Linear length Affected area N/A 13 22 Yes No 10 5 Linear length Others

Standard Definitions for Common Types of Pavement Cracking 22 (a) Extent evaluation factors. (b) Severity evaluation factors. Figure 10 “D” cracking. Corner Break The corner break data collection on concrete pavements is required in 18 of the 33 SHA respondents. As shown in Figure 11a, the number of corner breaks (10) and the number of slabs affected (9) are the two primary criteria to quantify the extent of corner breaks. Additionally, Figure 11b illustrates that crack width (13) and level of spalling (10) are the two primary factors to evaluate corner break severity levels. (a) Extent evaluation factors. (b) Severity evaluation factors. Figure 11 Corner break. Other Cracking Data Sealed cracking is collected by 26 of the 35 SHA respondents (Figure 12a). The predominant number of respondents (17 of the 19 respondents) indicated sealed cracks are rated as low severity cracking (Figure 12b). Furthermore, 15 of the 26 SHA respondents report sealed cracks by linear length (Figure 12c). Finally, 17 of the 31 respondents do not collect other cracking data, which was not included in the survey questionnaire (Figure 12d). 6 2 4 Number of slabs affected Number of  transverse  joints with  “D” cracking Others 4 7 4 Crack width Level of patterns developed and amount of loose or missing materials Others 9 10 3 Number of corner breaks Number of slabs affected Others 10 13 9 Crack width Level of spalling Others

Standard Definitions for Common Types of Pavement Cracking 23 (a) Sealed cracking collection. (b) Severity levels of sealed cracks. (c) Reporting of sealed cracks. (d) Other cracking data. Figure 12 Other cracking data. Part III: Wheel Path Definitions Many protocols include a wheel path definition to distinguish load related distress from distresses associated with environmental factors (34 of the 35 SHA respondents) (Figure 13a). In addition, 20 of responding SHAs use a 39 in. (1 m) wheel path width (Figure 13b). For non- wheel path zones, there is no standard width and most of the responding SHAs indicated width varies depending on lane width. (a) Wheel path definition. (b) Wheel path width. Figure 13 Wheel path definitions. 26 9 Yes No 17 1 1 Low Medium High 15 11 Linear length Others 14 17 Yes No 34 1 Yes No 20 4 4 5 39 " 3 ft Depending on lane width Others

Standard Definitions for Common Types of Pavement Cracking 24 Part IV: AASHTO R 85-18 Applications As reported in AASHTO R 85-18, the total cracking length and average cracking width of each cracking type are reported for each of the five zones (Figure 2). SHA respondents (24 of the 33 responses) indicated AASHTO R 85-18 has not been implemented for cracking quantification. Among the 14 SHA respondents regarding the implementation and revision of AASHTO R 85- 18:  6 indicated concerns on whether or not AASHTO R 85-18 will meet the data needs for HPMS reporting, PMS, and Pavement ME Design™,  4 would like to distinguish alligator cracking and block cracking rather than simply use pattern cracking,  2 recommended using severity levels for pavement cracking data and calculating cracking density as additional indicators for pavement condition evaluation,  1 indicated possible inconsistency with the historical data collected using agency protocols for pavement management and the impact on performance prediction. Part V: General Comments Lastly, SHAs were asked to provide opinions and expectations for the development of a new cracking protocol for a fully automated system. Some key responses include (23 responses):  Develop cracking certification standard or practice similar to that for ride quality (AASHTO 2018),  Improve crack detection accuracy, especially on concrete pavements,  Provide a protocol that can keep up with the evolution of automated cracking data collection technology,  Provide real-time cracking detection and reporting. Other comments are contained in Appendix B. Summary State-of-the-practice cracking data definitions include:  AASHTO (R 85 and R 55), NPS, and SCANNER protocols were developed for automated cracking data collection, while the ASTM PCI standards and LTPP Distress Manual were developed primarily for manual surveys.  Cracking protocols developed in Alabama, California, Delaware, Florida, Idaho, Maryland, Pennsylvania, Vermont, and Virginia include current or future considerations for automated pavement condition survey technologies, but with varying levels of details.  Transverse, alligator/load associated/wheel path associated, and miscellaneous/non- load associated/non-wheel path associated cracking are the predominant cracking types in the distress protocols, especially for automated data collection techniques. The key findings of the SHA survey include:

Standard Definitions for Common Types of Pavement Cracking 25  2D and 3D image-based automated data collection is the dominant approach to cracking data collection and processing among the SHA respondents.  Agency-specific (30%), HPMS Field Manual (27%), AASHTO R 85-18 (23%), and LTPP Distress Manual (17%) are the major protocols used by SHAs.  Transverse, longitudinal, and alligator/fatigue cracking are the dominant cracking types collected by SHAs.  Majority of the SHA respondents use crack width to define severity level for transverse cracking (61%) and longitudinal cracking (60%); however, crack width thresholds and how they are reported vary by SHAs.  SHAs (50%) rely on cracking angle orientation to classify transverse cracking and longitudinal cracking, while the angle ranges vary within SHAs.  SHAs (30%) require reporting the minimum length of cracking, with 1 ft. (0.3 m) being the most widely used minimum length.  SHA respondents (71%) prefer to retain alligator/fatigue cracking separately from wheel path and non-wheel path locations. SHA respondents (50%) include the portion of transverse and longitudinal cracking in the wheel path(s) as alligator/fatigue cracks. SHA respondents (48%) believe the interconnectivity of the cracks should be used to rate the severity levels of alligator/fatigue cracking.  SHA respondents (44%) collect block cracking data, with 75% SHAs preferring to retain block cracking independent of wheel path locations.  Edge cracking, “D” cracking, and corner breaks are collected by 37%, 34%, and 55% of the SHAs, respectively.  SHA respondents (74%) require the reporting of sealed cracks, with 90% SHAs rating sealed cracks as low severity cracking.  SHA respondents (97%) define wheel path and non-wheel path zones to distinguish load related and non-load related cracking. The majority of the SHA respondents (61%) use 39 in. (1 m) wheel path width, while the width for non-wheel path zones varies with the width of the survey lane.  SHA respondents (73%) indicate AASHTO R 85-18 has not been implemented by their agency. The major reasons include (1) may not fully meet requirements for HPMS, PMS, or Pavement ME Design™, and (2) may generate inconsistent cracking results when compared to historical data. The outcome of the SHA surveys is rather diverse. In a limited number of cases, there are some commonalities among SHAs from the survey results. For example, wheel path definitions from the survey are 39 in. (1 m) as the width of a wheel path; feedback from the majority of the survey responses use 0.25 in. (6 mm) and 0.5 in. (12 mm) as thresholds of the low, medium, and high severity levels of cracking; cracking types including longitudinal cracking, transverse cracking, and alligator cracking are commonly used in many SHAs. From the results of the literature review and agency survey, the current state of diversified protocols has hindered the progress of producing consistent and comparable cracking data. Therefore, a highly focused new cracking standard needs to be developed to satisfy future pavement management and design practices, federal reporting mandates, and new automated processing technologies.