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.
54 This chapter addresses the state of the art versus the state of the practice for automated pavement distress collection tech- niques. The distinction is important because of the real and perceived differences in what is achievable and what is being achieved in the areas within the scope of this synthesis. Clearly, many of those differences are because the applicable technologies are rapidly evolving and will continue to do so for the foreseeable future. This chapter attempts to capture the sense of how that evolution is taking place and how it will continue. Several conclusions point out technical areas where it might be advantageous for user agencies or vendors to move closer to the state of the art. PAVEMENT IMAGING It is likely that the area of pavement imaging and the interpre- tation of those images present the largest gap between art and practice. It is clear that much of the work going on in research studies show the great potential of automated real-time collec- tion of pavement surface distress (cracking, etc.) data. How- ever, many of the agencies using the applicable technologies at the time the project questionnaire was circulated did not have great confidence in the quality of the data collected, pro- cessed, and delivered to them by either vendors or by agency personnel. Therefore, it is safe to say that there is a relatively wide gap between art and practice for image-related data. One of the reasons for an apparent reluctance to accept data at face value may be the relative dearth of workable and accepted QA procedures. Although such procedures were dis- cussed earlier and some are in use, it is evident that much more needs to be done. The research must be conducted before it will be possible for data collection and processing protocols to contain the appropriate QA requirements and guidelines. Until then, there may be little hope of significantly narrowing the gap between art and practice for those data. Another area requiring further development is in the achievement of truly acceptable pavement images from equip- ment operating at highway speeds. Although some vendors and some agency-owned equipment appear to deliver adequate images under some conditions, it can be noted that those con- ditions too often do not prevail, such that the images delivered are at best borderline in quality. The result is that users again do not have confidence in the data derived from those images. One limitation of the current state of digital imaging and automated processing of distresses is in the area of shadows on images. In the absence of special lighting to reduce shadows, it is very difficult to obtain usable images under many condi- tions, such as shady roadways and in the presence of vehicle- cast shadows. Another limitation is with coarse-textured surfaces such as chip-sealed surfaces, tined surfaces, and sealed cracks. Although researchers are confident that high- resolution images will overcome this limitation, the art has not yet been developed. The whole area of laser imaging, essentially 3-D profil- ing of the pavement surface, is an emerging technology that will require some maturing before it comes into practical use. At present, the limitation appears to be in the resolu- tion that is reported by some to be too low for effective crack detection (20). SENSOR-RELATED DATA There is a wide variation in how art and practice relate with regard to sensor data, depending on the data item considered. The separate items are discussed briefly here. Roughness For roughness, there appears to be little gap between art and practice. No doubt this relates to an emphasis placed on IRI determination by the HPMS program, requiring agencies to test the National Highway System at least every other year. That emphasis means that all agencies are doing at least longitudinal profile testing. The need for more equipment and faster deliv- ery of IRI data has caused equipment manufacturers and ven- dors to place greater emphasis on updating their equipment. As a result, almost all are using the fastest sensors (high-speed lasers) available and the supporting hardware and software. The absence of agency resources to do roughness monitoring with older equipment has led almost all agencies to either pur- chase or hire the latest equipment as it becomes available. Rut-Depth Measurement An entirely different matter prevails with rut-depth measure- ment, where there is a broad gap between the practice and the art as propounded by vendors. As discussed in chapters two and six, there is a wide variation in how agencies are speci- fying rut-depth measurement. Protocols in use provide for from 3 to 37 points of measurement along the transverse pro- file and a wide variation of how the data are to be used to determine rut depth. One researcher (38) makes a case for using at least nine sensors to determine rutting, whereas at least two vendors are promoting transverse scanning lasers CHAPTER NINE ART VERSUS PRACTICE
55 capable of providing thousands of incremental measurements on that transverse profile (26,40). Presumably, that wealth of data would permit an extremely accurate determination of rut depth at almost any longitudinal frequency desired. How- ever, the protocols to do that type of measurement still need to be developed. The gap here will remain broad until such protocols are available. Joint-Faulting Measurements The automation of joint-faulting measurements has attracted so little attention that there is virtually no agreement on what is required. Although there is a provisional AASHTO stan- dard, very few agencies report its use, because most simply take what vendors provide or use protocols suggested by ven- dors. The result is that there really is little definition of either art or practice. That situation will prevail until someone under- takes to clarify what is really needed to do automated joint- faulting measurements. Slab Warping and Curling Areas related to joint faulting are warping and curling of concrete pavement slabs. Although no agency mentioned these concerns in survey responses, the FHWA has a project underway to apply inertial profiling as the measuring tool in a major warping and curling research study (86). The FHWA cites recent advances in inertial profiling technology devel- oped at FHWAâs TurnerâFairbank Highway Research Center and advances in computer technology as making it possible to reliably measure the shape of very large numbers of PCC pavement slabs over a short period of time and to perform analyses of these data. Products of the research would include written and computer-based guidelines that will focus on the effects that design and construction decisions will have on slab curvature and ultimate long-term performance. Successful completion of such a project would no doubt stimulate agency interest in applying similar technologies. AASHTO PAVEMENT DATA COLLECTION PROVISIONAL STANDARDS Many respondents to the questionnaire professed to no knowl- edge of the AASHTO provisional standards. Several even asked for copies or a web address where the standards might be accessed. It is not clear whether this situation reflects breakdowns in internal agency communications or a problem at the national level. On the other hand, discussion revealed that at least the roughness standard is being widely used, but it is often given some other designation such as SHRP, LTPP, or ASTM. Only four agencies reported using the AASHTO crack- ing standard or some modification thereof (see Table B1 in Appendix B). The reason most often given for not using it relates to banks of legacy data that agencies fear might no longer be useful if too many changes in the collection stan- dards are adopted. C. Grogan, pavement management engi- neer, MDOT (personal communication, August 2003), pro- vided this perspective: We are not using the AASHTO provisional protocols at this time since we have been adhering to the SHRP Manual and feel that this meets our needs. Also, there is the problem of converting legacy data, at which task we spent years getting it into our pres- ent information system, TMIS (Transportation Management Information System), which is a custom-built data warehouse application that includes a pavement management subsystem. We are also trying to get our Pavement Analysis Package (PAP), which is an optimization program, up and running. More histori- cal data conversion would involve huge coding changes to TMIS, and would keep us from moving forward toward more actual use of the data. We also like the specific nature of the SHRP Manual interpretation of distress. Some agencies reported using AASHTO provisional standards with local modifications. For example, Wang et al. (87) described a crack detection protocol developed for the Arkansas State Highway and Transportation Department based primarily on the AASHTO provisional standard. Changes were in the definitions of transverse and longitudinal cracking to meet Arkansas needs. Again, Maryland (78) described its crack detection process as not totally compliant with but âwell within the spiritâ of the AASHTO provisional standard On the other hand, many agencies appear to use the rough- ness standard or something very closely related (see Table B2 in Appendix B). Twelve agencies listed the AASHTO stan- dard, while 19 cited either ASTM, FHWA, or LTPP guide- lines. Although these results suggest that there is not a firm understanding of the standard and its origins, again they prob- ably reflect a natural outgrowth of HPMS requirements and the accompanying ASTM standards. As discussed, there is little real agreement on what is really required for either rut-depth or joint-faulting measurements. Although several agencies listed the AASHTO standard for rut-depth determinations, most either left the question blank or cited a vendor or agency protocol for joint faulting (see Appendix B, Tables B3 and B4). It is unlikely that AASHTO provisional standards will be put to much use until a better understanding of the automation issues has evolved. SUMMARY The issue of state of the art versus state of the practice for automated pavement condition data collection and processing is cause for concern. Although there appears to be little gap in roughness measurement, there is a significant one for both cracking and for rut-depth measurement. For joint-faulting measurements, the automation issues are not well enough defined to determine either the state of the art or state of the practice. It is unlikely that the AASHTO provisional standards will be put to widespread use for other than roughness monitor- ing until some of the issues discussed in art versus practice are resolved.
