Performance-Based Seismic Bridge Design (2013)

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.

9 CHAPTER TWO PUBLIC AND ENGINEERING EXPECTATIONS OF SEISMIC DESIGN AND THE ASSOCIATED REGULATORY FRAMEWORK the public’s mind may be a function of previous experi- ence. Kawashima (2004) cites survey data of engineers following the 1995 Kobe earthquake, where those who had firsthand experience with that event preferred higher per- formance objectives than those who had not experienced it. His survey systematically indicated similar trends for repair time, for which those who had experienced the Kobe earthquake gave higher and more realistic estimates. Thus, experience with actual earthquakes is an important param- eter in setting realistic goals and expectations, yet many stakeholders in seismic regions across the United States do not have such experience. For the most part, informed decision making has taken a back seat to “risk and safety as by-products of design” (May 2001). May argues that safety and risk must be treated as explicit considerations and not the products of other choices. The other choices he refers to are the engineering decisions that are made largely out of the public view and that are related to satisfying and choosing among prescrip- tive requirements of design specifications. The public is not equipped to participate in decision-making discussions that focus on prescriptive engineering requirements. The decisions made in structural seismic design are almost all based on satisfying requirements related to resistance and structural behavior. May makes the case, and PBSD indeed requires, that more choices be considered in the process, and these choices must be put into terms that are meaning- ful to the public. For instance, choices about seismic design might be considered relative to such alternatives as purchas- ing insurance to mitigate the risk of loss as the result of an earthquake or use of alternative facilities. “As with any cli- ent, the engineering profession should seek to inform, rather than make, collective decisions about minimum standards of performance for different situations or classes of facilities” (May 2001). However, with our current prescriptive design requirements, we are operating more on the “making deci- sions” level than on the “seeking to inform” level. In some sense, the public relies on the regulatory com- munity to consider alternatives and set appropriate require- ments. In so doing, the public does not consider or understand the choices that its building officials are making for it. This may have elements of representative democracy, but it does not always bring the public into the decision-making process The public expects that structures, including bridges, are designed to resist earthquakes. Beyond that simple state- ment, it is not abundantly clear what the public really expects, as few surveys of the general public have been con- ducted and published. The University of Delaware Disas- ter Research Center used mail surveys and focus groups of Alameda County, California, residents to determine “per- ceptions of acceptable levels of performance of different elements in the built environment in the event of a major earthquake” (Argothy 2003). Within the portion of the sur- vey and focus group discussions on transportation systems, strong and varied views were expressed, with one respon- dent stating “there’s just not a perfect world, but you don’t expect the bridge to fall down when you’re driving across it” (Argothy 2003). This clearly reinforces the impression that the public is expecting at least life-safety or no-collapse performance under even the most severe earthquakes. Addi- tionally, some respondents said that closures of important bridges, such as the 1-month closure of the east portion of the San Francisco–Oakland Bay Bridge after the 1989 Loma Prieta earthquake, were unacceptable. This response shows that the general public expects enhanced performance objec- tives along essential corridors and for signature structures. Although not specifically addressed in the University of Delaware study, some of the public will undoubtedly surmise that if a new structure is designed for earthquake loading or an existing structure is retrofitted for the same loading, then the structure is “earthquake-proof.” Engineers know better, but sometimes only marginally so. We know from experi- ence and from the design codes’ general language that life safety must be assured and that damage may be significant for the majority of structures. If a structure is designed with better performance in mind, then we may expect more from the structure, bordering on the earthquake-proof designa- tion. In all, there is a wide range of expectations and a gen- eral lack of public consensus as to what to expect following an earthquake. The engineering community does not help the matter because we generally are not adept at articulat- ing how structures will behave under earthquake loading. This situation must be changed if PBSD is ever to take hold because owners, and often the public, must have input into the project’s performance objectives. Additionally, the perception of acceptable performance in both the engineering community’s mind and likely in

10 as fully as may be warranted. Such situations, where public involvement is limited, are much more likely to exist for con- ventional or ordinary structures or bridges, whereas the oppo- site is generally true for larger, important projects where the public involvement process is much more overt and developed. May (2001) again argues that in the project development phases, deliberations must— • “expose the consequences of choices and their trade- offs with respect to safety/risks, benefits and costs,” which is often done for larger projects, but insufficient data typically exist to produce meaningful conversa- tion for ordinary projects. • “expose distributional aspects of choices,” which means that the implications of choices across geo- graphic and economic sectors must be understood. • “express consequences for different levels of decision making,” which simply means that different jurisdic- tions consider or perceive the consequences of decisions differently because the consequences are indeed differ- ent for each—for example, the monetary contributions that are made by each jurisdiction may be different. • “inspire confidence in the approach and conclusions,” which as May observes “may seem obvious, but it is an important lesson that has been lost in past debates over nuclear safety and high-level nuclear waste,” for example. Today, we likely do not consider the effects of each project equally and in the manner described earlier. In some cases, this is because the engineering community simply does not have the resources to make the comparisons and frame the questions that are necessary. To move beyond the existing situation and begin to more completely apply PBSD, the engineering community will need to develop a “greater societal awareness of earthquake risks and their consequences, but also transform the way that owners, financial entities and the design community think about seismic safety” (May 2007). One of the key aspects of considering such consequences is that there are trade-offs to be considered regarding resources dedicated to mitigating seismic hazards and the risks (i.e., losses) that could come from those hazards. The Christchurch earthquake sequence triggered by the September 4, 2010, Darfield earthquake is a recent example of the difficulties in allocating resources for earthquake haz- ard mitigation, where a previously unknown fault unleashed a series of shallow damaging earthquakes and aftershocks over a period of more than 18 months. To a large extent, life safety has been achieved, in that only a few buildings have collapsed. However, many structures are no longer safe to be occupied and, thus, a large fraction of the city’s build- ings were deemed unusable and need to be demolished and replaced. This tragedy has demonstrated that when a damag- ing earthquake directly strikes even a modern and well-pre- pared community, the sheer amount of short-term losses can cause serious disruption to the community and its economic viability after the event (I. G. Buckle, personal communica- tion, 2012). Other examples are cases where the resources dedicated to hazard mitigation might be somewhat out of line with the perceived benefits that might be achieved. This is an area where accelerated bridge construction (ABC) techniques may become more valuable. Implementation of ABC could lead to permitting more seismic damage than normally would be case, provided life safety is still achieved, because bridge replacement might be quick, thereby reducing delay costs compared with conventional construction. ABC con- cepts will likely change the dynamic of the decision-making process. Further information on ABC concepts for higher seismic regions can be found in Marsh et al. (2011). Cur- rently, the engineering community is not adept at making such decisions, or at framing the appropriate questions for decision makers. The engineering community will need to do a better job, but this will take time. With respect to systems where more than one facility is included in a linear network that delivers a service (e.g., highway system), the choices and trade-offs must be con- sidered in the context of the system performance rather than just the individual facility or structure performance. Such tools as Risks from Earthquake Damage to Roadway Sys- tems (REDARS), discussed in chapter seven, help provide a more complete picture of the questions so that decision mak- ers can allocate resources in a manner that is most beneficial to system performance. With respect to the potential benefits and costs of imple- menting PBSD, there may not be solely positive aspects of implementing PBSD. Some situations may lead to increased first cost relative to long-term risks. There may be costs of educating the engineering, construction, and regulatory community in terms of using, implementing, and admin- istering PBSD. There may be potential legal risks if target performance goals are not met. And there may be costs asso- ciated with inconsistency relating to ambiguous interpreta- tion of performance levels when criteria are unclear (May 2007). Considering the way public work is contracted in this country, unintentional (or less safe) interpretations of per- formance levels relating to criteria could occur, leading to future problems with facility service. An advantage of the current prescriptive-based seismic design procedures is that they are somewhat easy to enforce. With prescriptive methods, implementation of design details is binary—a detail either was or was not included. This advantage plays to our method of controlling the construc- tion process, and is evidenced in our special inspection and construction observation procedures. We prefer, and our legal

11 system encourages, that acceptance procedures be enforce- able at the time of the construction when the contractor is still on the project and before all payments are made. Prescriptive measures lend themselves to enforcement during the period of the contract; performance-based measures may not and instead rely on the use of warranties in case of future substan- dard performance. Thus, performance-based methodologies are not yet on the same easy-to-enforce, binary (yes/no) basis as conventional prescriptive seismic design. This lack of clear guidance leads to a political and legal challenge whereby the enforcement of seismic safety require- ments becomes less structured. This particularly becomes challenging when the public is involved in the decision-mak- ing process. As May (2007) articulates, on the one hand, determining levels of acceptable risk is fundamentally a value judgment that presumably requires some form of collective decision-making. On the other hand, knowledge of relevant risk considerations, technical details, and costs and benefits are important for establishing minimum standards. The first consideration argues for public processes for establishing safety goals. The second argues for deference to technical experts. Finding the appropriate middle ground is a serious challenge. May and Koski (2004) illustrate this challenge in Per- formance-Based Regulations and Regulatory Regimes. They first observe that the move toward performance-based design is related to the general modern political movement to relax regulation to foster innovation and remove barriers to economic growth. However, their treatment of the regu- latory environment points out through four different case studies how the challenge of open decision making and per- formance-based approaches to regulation may not achieve the desired outcomes. For instance, they review the introduction of perfor- mance-based code provisions for home construction in New Zealand in 1991, which coincided with popular preferences for stucco or adobe finishes on home exteriors. Problems with moisture, leakage, and so on began to emerge, and by the early 2000s a crisis was at hand. There were many inadequacies in the regulatory system (code provisions) and with the construction industry’s delivery of homes. Reforms were enacted to swing the pendulum back, to increase gov- ernment oversight, to more clearly define the performance standards, and develop mechanisms to monitor products and provide warnings about defective ones. In all, “a gen- eral tightening of the regulatory regime with emphasis on greater specification of performance standards and stronger monitoring of building inspection practices” was enacted. This case history illustrates the need for balance between performance-based objectives and oversight of the industry. An observation of modern design specifications is that many, if not most, of the provisions and prescriptive require- ments were included in the specifications to prevent some type of failure or poor service performance that actually occurred at some time and at some location. When pre- scriptive requirements are incorporated into design speci- fications, the associated details of the poor performance the provisions are intended to prevent are often lost. This loss accelerates proportionally with time from the originat- ing failure or research. The preservation of such behavioral information is one of the primary reasons for including a commentary to a design specification. Good engineering involves anticipating and preventing modes of failure, and if previous lessons learned regarding past failures are lost in a morass of prescriptive design pro- visions, then innovation is stifled and the engineering com- munity is likely destined to relive the past. Therefore, the well-crafted performance-based design specification would likely control specific modes of failure by a combination of performance requirements and a fallback to prescriptive requirements when performance objectives are unclear or ambiguous. This process might be thought of as a hybrid approach to performance-based design, and, given our struggle with balancing the challenges of regulation and the desire to innovate, such a hybrid design specification may be the most logical way forward.