Assessing the Long-Term Performance of Mechanically Stabilized Earth Walls (2012)

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.

20 After wall condition and performance data have been col- lected, assessments can be performed to determine how well MSE walls are meeting their performance objective(s). Assess- ments can also be performed to prioritize maintenance and replacement functions. [As a reference, FHWA (1999) has developed a basic primer regarding assessment manage- ment concepts while Bernhardt et al. (2003) have discussed application of these concepts to “geotechnical infrastructure” assets.] Such assessments commonly involve some type of numerical scale or standard set of terms. These scales or terms can be used in quantitative rating algorithms and/or more sub- jective, qualitative expressions of wall performance. Ideally, these scales ultimately link current wall performance with the wall’s position within its design life cycle. This chapter will discuss how wall performance data are assessed and then used for asset management. ASSESSMENT AND INTERPRETATION OF DATA Referring again to the established and tested FHWA’s WIP, the wall element and performance data collected (as discussed in the previous chapter) are combined with factors measuring the relative importance of each element to establish a final overall wall condition rating, which ranges from 5 to 100. Conver- sion of this numeric rating to a qualitative description can be approximately achieved by dividing the rating by 10 and com- paring it to the element and wall performance rating definitions shown in Table 6 and Table 7, respectively. Although their origin is not explicitly stated, it appears that the weighting factors used in the WIP were established by some type of consensus of experienced persons. The pro- cedure manual states, “these element weightings have been determined to sufficiently discern element impacts on wall performance. However, as more wall inventory data are col- lected, weightings will be re-evaluated for appropriateness, and altered as needed to provide meaningful and consistent wall condition ratings.” The FHWA WIP wall condition rating was also cited by Brutus and Tauber (2009) in their consideration of how to quantify wall performance. They also provided the five-point rating scale in Table 14 as another possible sample rating system. In some numeric schemes, adverse performance is indicated by a low rating, whereas in others a low score is desirable. Some MSE wall assessments do not incorporate a type of condition rating, numeric or otherwise. For example, state agencies in Utah and Ohio currently document only the existence of certain adverse conditions. As part of this synthesis, 44 survey participants provided feedback regarding how important they thought particular wall features and conditions are in assessing the long-term perfor- mance of MSE walls. These beliefs are in large measure repre- sentative of the relative importance of specific wall condition data and might function similarly to the FHWA WIP weighting factors in a current assessment or prediction of future wall per- formance. In the survey, relative importance was distinguished using a numerical rating scale where 1 = not important, 2 = mildly important, 3 = moderately important, 4 = very impor- tant, and 5 = most important. The results in terms of average rating are shown in Table 15. Also shown is the variance for each feature from the overall mean rating, helping indicate each feature’s perceived importance relative to the others. As can be seen in the table, features associated with drain- age (both external and internal) typically are considered to be among the most important. Changes to wall geometry resulting from excavation or addition of surcharge load that would affect global stability are also viewed as being rela- tively important. Most important, however, is corrosion and degradation of internal reinforcement. This result appears to be consistent with the impetus for the initial establishment of many existing MSE wall inventories—concerns relative to, or premature failures stemming from, corrosion of MSE wall reinforcement. Interestingly, a small panel of MSE wall experts convened by the Utah DOT judged that drainage issues are the most significant issues during the first 15 years or so of wall life, after which corrosion issues become the most important (Bay et al. 2009). Perhaps most surprisingly, the survey indicated that wall height is considered among the least important—surprising because this parameter is among the more frequently included parameters in wall inventories. This also appears inconsistent with the assessment of Brutus and Tauber (2009) that the most important component contributing to risk stemming from wall failure is the height of the wall. Also surprising is that wall age (as implied from date constructed) is rated as being below aver- age in importance because internal corrosion (the most impor- tant factor) is itself a function of age. chapter four ASSESSMENT AND USE OF MECHANICALLY STABILIZED EARTH WALL DATA

21 Rating Description Excellent No significant indication of distress or deterioration. Good Some indications of distress or deterioration, but wall is performing as designed. Fair Moderate or multiple indications of distress or deterioration affecting wall performance. Poor Significant distress or deterioration with potential for wall failure. Critical Severe distress or deterioration. Indications of imminent wall failure. Source: Brutus and Tauber (2009). TABLE 14 SAMPLE RATING SYSTEM FOR WALL PERFORMANCE Response Mean Variance Internal corrosion/degradation of reinforcement 4.4 +0.9 Internal erosion of backfill 4.1 +0.7 Wall geometry changes (e.g., excavation at toe, added surcharge load) 4.1 +0.6 Functionality of internal drainage features (e.g., weepholes and piping) 4.0 +0.6 Drainage conditions 4.0 +0.6 Proximity of external water sources (e.g., river, sprinklers, etc.) 3.9 +0.4 Distress in ground or pavement behind wall 3.8 +0.4 Functionality of drainage/catch basins 3.8 +0.3 Bulging or distortion of wall facing 3.7 +0.2 Maximum wall height 3.7 +0.2 Cracking of facing elements 3.6 +0.2 Settlement behind wall 3.6 +0.2 Reinforcement type 3.6 +0.1 Location and condition of drainage discharge points 3.5 +0.1 Rust stains or other external evidence of corrosion 3.5 +0.1 Distress in ground or pavement in front of wall 3.5 +0.1 External erosion 3.5 +0.1 Embedment of wall 3.5 +0.0 Post-construction modifications 3.5 +0.0 Settlement along line of wall 3.5 +0.0 Slope behind wall 3.4 +0.0 Damage from vehicular impact 3.4 +0.0 Slope in front of wall 3.3 –0.1 Alignment and spacing of joints between facing elements 3.3 –0.1 Wall plumbness 3.3 –0.2 Wall type 3.3 –0.2 Damage to corners of facing elements 3.2 –0.3 Presence of bench at toe of wall founded on slope 3.2 –0.3 Road/traffic offset 3.1 –0.3 Displacement of coping or parapet 3.0 –0.4 Date constructed 3.0 –0.5 Manufacturer 2.7 –0.7 Vegetation growth 2.7 –0.7 Average wall height 2.6 –0.8 Wall length 2.3 –1.2 TABLE 15 RELATIVE IMPORTANCE OF WALL FEATURES/CONDITIONS IN ASSESSING THE LONG-TERM PERFORMANCE OF MSE WALLS

22 In addition to the relative importance of certain wall features and conditions, survey participants were also asked to rate how significant they thought certain potential failure/distress modes were relative to the long-term performance of MSE walls. The failure/distress modes were those typically considered in wall design procedures. Significance was rated on a scale of 1 = not significant, 2 = mildly significant, 3 = moderately significant, 4 = very significant, and 5 = most significant. The results, shown in Table 16, indicate that most agencies believe that global stability and reinforcement rupture are the most likely failure modes for MSE walls in the long term. The term “reinforce- ment rupture” was not specifically defined, but is believed to have been interpreted to include failures resulting from both section loss and subsequent overstressing as well as overstress- ing of the initial section. The data also suggest that agencies believe overturning and facing failure are the least likely failure modes. This information is important in that these beliefs con- stitute a type of expert opinion that can be used in MSE wall service life prediction methods as well as in wall failure risk assessments. Both of these activities currently appear to be in their naissance, as discussed later in this chapter. USE OF PERFORMANCE ASSESSMENTS IN DECISION MAKING Once wall conditions are assessed and its condition quanti- fied on some basis (such as the FHWA WIP wall condition rating), the assigned rating can be used in more than one way for programming decisions. In some systems, the numeric value can be directly related to a specified action level (e.g., walls rated below 40 must be repaired). In other systems, the numeric value is used for ranking, and resources for items such as maintenance or repair are allocated accordingly (e.g., there is $100,000 in the budget for repairs, which walls do we start with?). In yet other systems, such as the FHWA WIP, the final overall rating is only one of several factors used to make pro- gramming decisions. The rating by itself is not directly related to a particular action. Rather, four additional items/questions are considered in the FHWA WIP: (1) are additional investiga- tions required (how reliable is our assessment); (2) what design criteria may have been used in planning the structure (was the structure engineered); (3) what aspects of the wall structure are historic or contribute to the cultural context of the road asset; and (4) what are the consequences of wall failure. These items are subjectively assessed by the person rating the wall with few objectively defined criteria; hence, programming decisions, to which wall condition ratings only partially contribute, are largely subjective in the FHWA WIP. As stated previously, some MSE wall assessments do not incorporate any condition ratings; therefore, some alternate means of decision making is required. On a comparative wall- to-wall basis, one can tally the number of adverse occurrences per wall and then rank the tallies to establish a type of prior- ity list. Swenson (2010) used the Utah wall inventory data and attempted to improve the ranking processes by associating particular conditions/issues with particular failure modes and then assigning weights to indicate criticality. Unfortunately, the expert input/consensus usually required to link conditions, failure modes, and consequences was limited. When asked about a specific methodology for assessing long-term performance of existing MSE walls, no survey respondent answered affirmatively beyond citing regular inspections or several corrosion assessment studies. These items appear to be contributing components to a methodology, but no fully developed methods were identified. From the responses gathered and review of available literature, it does appear that some agencies may rely largely on pre-approval product processes and compliance with Highway Innovative Technology Center criteria (see Highway Innovative Tech- nology Center 1998) for assurance that MSE walls will per- form adequately. Although such measures should improve the likelihood of good, long-term performance, failure case histories suggest that they are not failsafe. Estimation of Service Life In their study, Brutus and Tauber (2009) concluded that “there is no data available in technical literature on the estimate of designed service life or on construction or maintenance oper- ations on old retaining walls built somewhere between 50 to 100 years ago.” MSE walls in the United States are newer than this, yet this statement also appears to apply to those Response Mean Variance Global stability 4.3 +0.4 Reinforcement rupture 4.3 +0.4 Reinforcement pullout 4.2 +0.3 Loss of foundation support due to erosion 4.0 +0.1 Loss of foundation support to bearing capacity failure 4.0 +0.1 Excessive settlement 3.8 –0.1 Sliding 3.6 –0.3 Overturning 3.5 –0.4 Facing failure 3.3 –0.6 TABLE 16 RELATIVE SIGNIFICANCE OF POTENTIAL FAILURE/DISTRESS MODES IN LONG-TERM PERFORMANCE OF MSE WALLS

23 • Disruption of highway operations, including full or par- tial closure of the roadway, or appurtenant facilities; • Disruption of adjacent utility lines, such as water mains or electrical conduits; • Environmental consequences, such as damage to a sig- nificant wildlife habitat or blockage of a watercourse; and • Damage to cultural assets or sensitive land uses. Again, as outlined by Brutus and Tauber, the consequences of adverse wall performance or failure can be affected by: • The volume of earth retained by and otherwise contained in the wall, which in turn is most frequently reflected by the height of the wall; • The proximity of the wall ERS to the roadway or other potentially affected facilities or structures; • The intensity of usage of potentially affected facilities, such as traffic volume on a roadway or occupancy of a building; • The structural robustness of adjacent buildings and facilities; and • The vulnerability of occupants and/or users. Often the consequence of failure (either functional or structural) is also quantified or expressed in terms of some type of scale. Possible metrics include monetary losses, inju- ries or fatalities, and/or decrease in levels of serviceability. Brutus and Tauber suggest use of a three-level rating system such as that shown in Table 17. Performance of risk assessments for MSE walls at pres- ent appears to be problematic. Risk assessments (particularly probabilistic ones) typically require the use of “expert opin- ion” or “expert consensus”; however, being expert requires being experienced. As agencies continue to monitor wall performance, they will gain further experience, and with this increased experience, their ability to assess risk will improve; hence it is in this manner that methods for risk assessment are likely to evolve. Wall function as reflected in inventory inclusion criteria such as that shown in Table 2 would be of particular importance when executing risk assessments. newer MSE walls that have intended design lives of 75 to 100 years. As reported in the previous section, none of the agencies surveyed had a specific methodology for assessing long-term performance of existing MSE walls, let alone a method for estimating design life. Brutus and Tauber do however suggest two approaches that might be used to estimate the remaining service life of walls. One approach is to perform repeated inspections and “chart escalating maintenance and repair costs to project a remaining service life . . . using some criterion such as when the repair and maintenance costs exceed more than 50% of the replacement cost.” The other approach is to assess the performance of simi- lar walls (e.g., same construction standards) built over a long period of time and use the observed performance to forecast the performance of newer walls. However, care must be taken when interpreting adverse performance of walls constructed using different, older design methods that may not be represen- tative of newer walls. Elements of these approaches are now beginning to be implemented with the development of MSE wall inventories and the collection of data as described in the previous chapters. As pointed out previously, the development of initial inventories appears to be progressing much more rap- idly than regular ongoing performance data collection. Risk Assessment Tied closely to the assessment of wall performance is the assessment of risk. Sometimes, risk assessment is not explic- itly undertaken, particularly if wall performance appears more than adequate. Ultimately however, it is questions of risk and consequence of adverse performance that drive many asset management activities. Potential consequences of failure that are considered in the performance of risk assessments include (Brutus and Tauber 2009): • Death or injury to persons, including facility users and those on adjacent properties or facilities; • Damage to property including vehicles, highway prop- erty or facilities, and adjacent property or facilities; Rating Description Severe High likelihood of injuries or death fro m debris falling on a heavily traveled roadway, on other heavily used adjacent areas, or from collapse of structures near top of wall. High likelihood of extensive or total-loss damage to vehicles or structures. Complete closure of a heavily traveled roadway requiring lengthy detours. Significant Low probability of injury to persons but likelihood of any of the following: (a) substantial property damage, (b) interruption of water or other utility service to a large area, (c) lengthy blockage of access to business properties or public facilities, (d) long - term damage to environmental or cultural re sources, (e) closure of two or more lanes of a heavily traveled roadway, (f) full closure of any roadway with no alternative access or requiring lengthy detours. Minor Low probability of injury to persons or of damage to vehicles or non-highway property or facilities. Full roadway closures where alternative access is available. Closure of a single lane on a heavily traveled roadway. Source : Brutus and Tauber (2009). TABLE 17 SAMPLE RATING SYSTEM FOR CONSEQUENCES OF FAILURE