0068_C02_fm Page 1 Monday, August 19, 2002 12:28 PM

2
Earthquake Risk
Management:
An Overview

2.1 Introduction
2.2 Overview of Earthquake Risk
2.3 Identifying the Assets at Risk
2.4 Earthquake Hazard
2.5 Earthquake Damage and Loss
Infrastructure · Loss
2.6 Mitigation Alternatives
2.7 Earthquake Risk Management Decision-Making
Maximization (Benefit–Cost) or Minimization (Total Cost)
· Maximin and Maximax · Minimum Regret · Minimum
Uncertainty · Satisficing · Next Steps
2.8 Earthquake Risk Management Program
2.9 Summary
Charles Scawthorn Defining Terms
Consulting Engineer References
Berkeley, CA Further Reading

2.1 Introduction
Earthquakes represent a risk in many parts of the world, particularly western North and South America,
Japan, China, the Philippines, New Zealand, and the lands surrounding the Mediterranean, to name only
some of the higher seismic risk regions. This is well known. What to do about earthquake risk, however,
and how, is not so clear. To address these questions, this chapter provides an overview of the earthquake
risk management process, as a sort of road map for reducing earthquake risk. That is, the reduction of
earthquake risk involves the use of knowledge, methods, and data from disparate fields, including the
geosciences, engineering, emergency planning and disaster response, insurance, and economics. These
are the tools that engineers, planners, and other professionals have for the job of reducing, or managing,
earthquake risk. This chapter provides an overview on how to use these tools — that is, on assessing the
risk of earthquakes, determining mitigation alternatives, making a decision about what to do, and doing
it. While each succeeding chapter in this handbook provides detail on one or another of the tools, it is
this chapter that provides the framework within which these tools are employed.

© 2003 by CRC Press LLC

0068_C02_fm Page 2 Tuesday, August 13, 2002 7:21 AM

2-2 Earthquake Engineering Handbook

2.2 Overview of Earthquake Risk

The first step in understanding earthquake risk is to dissect the earthquake risk or loss process into its
constituent steps (see Figure 2.2). Earthquake risk begins with the occurrence of the earthquake, which
results in a number of earthquake hazards. The most fundamental of these hazards is faulting, that is,
the surface expression of the differential movement of blocks of the Earth’s crust. Faulting can be a simple
“mole-track” lateral movement, or a major vertical scarp, or may not even be visible. In most cases,
faulting is typically a long narrow feature, and therefore affects a relatively small fraction of the total
affected structures and persons. Affecting a much greater number of structures and persons is shaking,
which is typically the primary hazard due to earthquakes. Depending on the earthquake, liquefaction,
other forms of ground failure, tsunamis, or other types of hazards may be significant agents of damage.
For various reasons, many buildings, portions of the infrastructure, and other structures cannot fully
resist these hazards, and sustain some degree of damage. Primary damage can vary from minor cracking
to total collapse. Some building types are more vulnerable than others, but even when a building sustains
no structural damage, the contents of the building may be severely damaged. For certain occupancies,
such as hospitals or emergency services dispatch centers, this damage to contents (laboratories, specialized
machinery, communication equipment, etc.) can be very important. Additionally, these various kinds of
primary damage can lead to other secondary forms of hazard and damage, such as releases of hazardous
materials, major fires, or flooding.
Damage results in loss. Primary loss can take many forms — life loss or injury is the primary concern,
but financial loss and loss of function are also of major concern. The likelihood of sustaining a loss is
termed risk. Primary losses lead to secondary forms of loss, such as loss of revenues resulting from
business interruption and loss of market share and/or reputation.

2.3 Identifying the Assets at Risk

Earthquake risk management starts by defining the problem — what is at risk? As a general rule of thumb,
anything that is potentially subject to an earthquake hazard is at risk from earthquakes. It is of value to
catalog the types of assets potentially at risk, as a sort of checklist. Table 2.1 provides a sample catalog
of the typical assets that are at risk from earthquakes — it focuses not only on the assets at risk, but also
the threats that earthquakes pose to these assets and, generally but not exhaustively, the mitigation
alternatives available for countering each of these threats. In general, three types of things of value (assets)
are at risk:
• People can be killed or injured by earthquake effects in a number of ways — most commonly by
building collapse, but also by tsunami, fire, landslide, etc.
• Money can be lost, both in a relatively direct manner, such as having to pay to repair or rebuild
a building, and in less direct but no less real ways, such as loss of employment when the factory
is damaged, or due to short- to long-term shifts in economic conditions caused by earthquake
damage.
• Function can be impaired or lost, which eventually translates to human injury or financial loss.
Examples of functional impairment include 911 services not being able to function, and business
interruption due to damage to a factory or store.
In summary, if one is attempting to manage earthquake risk, the starting point is to define what one
might possibly lose due to an earthquake. In general, for an individual person or a large corporation,
losses can occur in all three asset classes (people, money, function). For example, an engineer may examine
the earthquake safety of a school building. The benefit of strengthening the building will be not only
fewer deaths or injuries, but also reduced repair costs for the building and contents, and continued
maintenance of function (i.e., another building does not have to be found and school can stay in session).

© 2003 by CRC Press LLC

0068_C02_fm Page 3 Tuesday, July 30, 2002 8:28 AM

Earthquake Risk Management: An Overview 2-3

TABLE 2.1 Assets at Risk

Asset: Loss Earthquake Threat Mitigation Alternatives

People: death and injury Building damage/collapse, Investigate site for potential faulting
via fault rupture, shaking, Strengthen the building
ground failure, etc. Base isolate the building
Provide supplemental damping
Provide ground or foundation improvements, if ground
failure is the issue
Replace the building (i.e., move, or new construction)
Building contents damage Inventory all contents and brace or otherwise reduce damage
Modify building motions via base isolation, supplemental
damping etc.
Equipment malfunction Identify and review critical equipment for continuity of
functionality during and after and earthquake (e.g., check for
relay chatter, backup power, water, fuel, etc.)
Assure equipment will not be damaged (i.e., brace, etc.)
Provide redundant equipment
Develop emergency plans and procedures for equipment
malfunction
Offsite threats Identify and review neighborhood for earthquake hazards
(e.g., tsunami, landslide) and threats (e.g., nearby hazardous
operations, such as a chemical process plant)
Develop Emergency plans and procedures, including possible
warning mechanisms
Build protective barriers
Acquire protective equipment and training (e.g., fire brigades)
Modify offsite threat (e.g., earthmoving, for a landslide; or buy
out nearby hazardous operation; or move)
Property: financial loss Same as above Same as above, plus
Inventory Emergency plans and procedures to minimize damage (e.g.,
recovery of inventory, quick shut-down of broken sprinklers)
Earthquake insurance
Function: business Same as above plus loss of Contingency planning for loss/replacement or recovery of
interruption, revenue, infrastructure (e.g., facilities (e.g., backup sites or suppliers, rapid recovery via
market share transportation), loss of pre-arranged inspection and repair contractors)
vendors Financial planning for loss of revenue
Earthquake/loss of profits insurance
Planning for alternative production/transportation to
maintain market share

2.4 Earthquake Hazard

Hazard and risk can be expressed and measured in a variety of ways. Prior to the invention of modern
scientific instruments, earthquakes were measured qualitatively by their effect or intensity, which varied
from point to point. With the deployment of seismometers, instrumental quantification of the entire
earthquake event — the unique size or magnitude of the event — became possible. A number of different
scales for magnitude and intensity have been developed, which are sometimes confused.1
In the United States, intensity is measured qualitatively using the Modified Mercalli Intensity (MMI)
scale, while engineers and scientists quantify seismic intensity in terms of peak ground acceleration (PGA)
and other measures. PGA is often expressed in terms of a fraction of the acceleration of gravity (g), such

1 Earthquake magnitude and intensity are analogous to a lightbulb and the light it emits. A particular lightbulb has

only one energy level, or wattage (e.g., 100 watts, analogous to an earthquake’s magnitude). Near the lightbulb, the light
intensity is very bright (perhaps 100 foot-candles, analogous to MMI IX), while farther away the intensity decreases (e.g.,
10 foot-candles, MMI V). A particular earthquake has only one magnitude value, whereas it has many intensity values.

© 2003 by CRC Press LLC

0068_C02_fm Page 4 Tuesday, July 30, 2002 8:28 AM

2-4 Earthquake Engineering Handbook

Trinet Rapid Instrumental Intensity Map for Northridge

JAN 17 1994 04:30:55 PST M6.7 N34.213 W118.5357 (site corrected)

34° 30'N

BLUE AREA
MMI < V

RED AREA
MMI > IX

34° 00'N

km
0 10 20 30

119° 00'W 118° 30'W 118° 00'W

FIGURE 2.1 MMI shaking intensities, 1994 Northridge earthquake. (Courtesy Trinet)

as 0.5 g or 50% of the acceleration of gravity. Figure 2.1 shows the MMI distribution for the 1994
Northridge, California earthquake, for example.
The identification and determination of earthquake hazards for a site or community are critical
elements in the earthquake risk management process. Earthquake hazard is a quantification of the various
ground effects at a specific site produced by earthquakes, and the likelihood that these effects will exceed
certain levels. More typically, it is a representation of how strongly the ground will shake and how often
it is likely to do so. Earthquake hazards are site specific, that is, they are different at each individual site,
depending on the site’s location and the properties of the ground beneath the site.
Earthquake shaking is usually very strong on or very close to the fault, and decreases or attenuates
with distance from the fault. This attenuation depends on the magnitude of the earthquake, and the
geology of the region. Soils, especially soft soils such as old filled-in marshes, can greatly increase or
amplify the ground motion. The effect of soils is a primary factor in the intensity of the shaking so that,
even though shaking generally attenuates with distance from the fault, shaking can still be very strong
at a large distance, if a building is on poor soils.

2.5 Earthquake Damage and Loss

As noted above, many buildings, elements of the infrastructure, and structures other than buildings
cannot fully resist earthquake effects. This section provides a summary overview of the kinds of damage
and loss that earthquakes can cause. Subsequent chapters provide more detailed information on the
earthquake performance of specific types of structures.
Buildings can fail or be damaged in any of several ways in an earthquake:
• Exterior walls fall away — this is the primary deficiency of unreinforced masonry (URM) and
older tilt-up buildings, and can result in partial or even total collapse of the roof system, posing
a significant life hazard and, typically, a major financial loss.
• Concrete columns can shear — dropping the building in a “pancake” mode of collapse. This mode
of failure poses a great life safety hazard and is common in older reinforced concrete frame buildings.

© 2003 by CRC Press LLC

0068_C02_fm Page 5 Tuesday, July 30, 2002 8:28 AM

Earthquake Risk Management: An Overview 2-5

• Certain kinds of steel beam-column connections can crack — this problem has only recently emerged
and was observed in significant numbers following the 1994 Northridge, California earthquake,
and is the subject of current research. None of the buildings that sustained this damage collapsed,
and the hazard to life safety is still unclear, but the financial consequences can be significant.
• Wood-frame buildings, for example, can slide off their foundations — this is the primary mode of
failure for older wood single-family dwellings with a crawl space beneath the first floor. The crawl
space can collapse when the so-called cripple walls supporting the first floor fall over. This failure mode
usually does not result in great loss of life, but can result in major or total financial loss of the building.
These are only the more common modes of building failure, and can all occur in a partial manner or to
a lesser degree — in such cases, the building will require inspection, perhaps vacating, and structural
and/or nonstructural repairs. As a rule of thumb, certain eras of building types have higher vulnerability:
• Masonry buildings may be unreinforced2 and have a high likelihood of exterior wall and parapet
collapse in even moderate shaking.
• Multistory concrete frames from the 1920s to 1970s often rely on their columns for seismic
resistance (termed moment frames). However, the columns often do not have the shearing resis-
tance we now understand is needed, so that this era of unreinforced concrete buildings has the
potential for a pancake type of collapse, with great life loss.
• Steel moment frame buildings of the 1970s and 1980s rely on the beam–column connection for
seismic resistance and, as noted above, certain kinds of these connections have been observed to
crack in recent earthquakes.
• Pre-1991 tilt-up buildings (a common commercial and factory building in California) may have
a particularly weak wall–roof connection, which has resulted in many instances of damage. Because
this weakness was observed in the 1971 San Fernando earthquake, newer tilt-ups may have better
connections, and many of the older buildings have been retrofitted.
From the above, it can be seen that, in earthquake terms, one or two decades may make a building
seismically “old.”
Nonstructural damage and repairs, from an occupancy viewpoint, can be as significant as structural
damage. Tumbled and broken furniture or equipment, fallen ceilings, cracked floors, broken water,
sprinkler, or gas lines, broken windows, extensive cracking of nonstructural interior walls, broken heating
or air conditioning equipment, spilled noxious materials, and other types of nonstructural damage can
all result in major costs, extended vacancies, and loss of productive use of the building.
For certain occupancies, such as hospitals or emergency service dispatch centers, this nonstructural
damage can be catastrophic, since these facilities are most needed following an earthquake. However,
even nonessential occupancies, such as small businesses, can suffer severe financial hardships as a result
of this type of disruption.

2.5.1 Infrastructure
When buildings sustain significant damage, infrastructure in the same neighborhood is typically also
heavily impacted. Examples include:
• Freeway collapse, such as the I-880 Cypress Structure in the 1989 Loma Prieta earthquake, the
collapsed bridges and interchanges on the various Los Angeles area freeways in the 1994 Northridge
earthquake, and the collapsed Hanshin Expressway in the 1995 Kobe earthquake.
• Underground piping breaks, especially in areas of poor soil. The combination of broken gas and
water lines both provides the fuel for fires and takes away the firefighter’s best weapon. More
noxious materials, such as sewage or crude or refined oil, may also be spilled, causing major
occupancy and clean-up problems.

2 Except more modern masonry buildings in high seismic areas.

© 2003 by CRC Press LLC
0068_C02_fm Page 6 Tuesday, July 30, 2002 8:28 AM

2-6 Earthquake Engineering Handbook

• Water and wastewater treatment plants can also be damaged, resulting in prolonged periods of
boil-water orders and spilling of raw sewage into bays or rivers.
• Electric power and telecommunications may suffer widespread short-term outages but, except in
the hardest hit areas, have typically performed reasonably well and been restored within a few
days. For certain types of businesses or services, however, even a few days disruption of these vital
services can be a major problem.
These various kinds of primary damage can lead to other secondary forms of damage, such as releases
of hazardous materials, major fires, or flooding. Major fires occur and spread due to the many ignitions
occurring in an earthquake. These numerous ignitions, combined with delayed response by the fire
department (since communications have been impaired), and lack of water for firefighting (due to
damaged infrastructure), can lead to a catastrophic situation.

2.5.2 Loss
Damage results in loss. Primary loss can take many forms:
• Life loss or injury is the primary concern. With the exception of a rare heart attack, earthquakes
do not kill people — rather, falling buildings or other structures injure and kill people.3 Thus, if
building collapses, major releases of hazardous materials, and/or fires can be prevented, numbers
of lives lost and serious injuries may be at levels which are as low as reasonably possible.
• Major damage to property can occur, even if life loss or damage are minimal. The cost to repair
or replace damaged buildings, contents, equipment, and other infrastructure is a direct financial
loss. Additional losses are incurred as a result of disruption of use.
• Loss of function can also be a major loss. If schools are closed for an extended period of time, for
example, the postponement of education is clearly a loss. Loss of hospitals, city offices, emergency
communications, and other important public functions are other clear examples.
• Primary losses lead to other forms of loss, such as loss of revenues resulting from business
interruption, loss of market share and/or reputation. These forms of loss can impact both the
public and private sectors.
• In the private sector, business interruption (BI) is a serious matter that some companies insure
against. Recent earthquakes in California, for example, have seen a number of small- and medium-
sized businesses fail, due to BI following earthquakes. These failed businesses and reduced activity
of the rest of the business community result in reductions in sales, property, and other taxes,
adversely affecting a local government’s finances.
• The public sector can have BI analogous to the private sector. Besides the obvious disruption to
public services, many local governments rely on revenue from ports, airports, or other special
facilities. If these are disrupted, finances are strained.
• If disruption to a community is prolonged, permanent loss of businesses or factories, with their
jobs, may occur. This is analogous to loss of market share in the private sector. If the jurisdiction
relies on revenue from a special facility, and the operation of that facility is disrupted for a
prolonged period, that facility’s customers may be forced to go elsewhere and may never return.
A prime example of this was the 1995 Kobe earthquake in Japan, where the world’s largest container
port permanently lost several major shipping lines to other ports it had been in competition with
prior to the earthquake. The overall disruption of the Port of Kobe in that case was less than a
year, but the customers are lost for many years, if not forever.

3 Earthquake-related natural hazards, such as tsunami or landslides, have killed hundreds or thousands of people.

© 2003 by CRC Press LLC

0068_C02_fm Page 7 Tuesday, July 30, 2002 8:28 AM

Earthquake Risk Management: An Overview 2-7

2.6 Mitigation Alternatives

The previous section described the nature and variety of damages and loss that can result from earth-
quakes. However, damage and loss can be reduced, or mitigated, in a number of ways. Figure 2.3 is a
modified form of Figure 2.2, and indicates how, at each step of the earthquake risk process, mitigation
of damage and loss is possible. That is, just as Figure 2.2 lays out the chain of causation of earthquake
damage, Figure 2.3 shows that breaking that chain anywhere reduces or eliminates the loss — and that
typically the earlier (higher) in the process the chain is broken, the more effective is the mitigation.
Each of the mitigation approaches indicated in the spectrum of earthquake mitigation alternatives is
a proven technology:
• Hazards such as faulting and shaking can be mapped. In many cases faults have been mapped in
detail, with improved and more detailed mapping going on all the time. For example, the California
Geological Survey over the last two decades has mapped the most active faults in California
(Figure 2.4). Shaking intensity maps are also readily available, even through the Internet (Figure
2.1). This type of mapping identifies the existence of the specific hazard. If that hazard is poor
soil, where liquefaction or ground failure might occur, then ground remediation may be required.
Recent decades have seen the development of a variety of new methods for soil improvement and
liquefaction mitigation, including soil mixing, stone columns, soil wicks, and chemical and pres-
sure grouting, to name only a few.
• Primary damage mitigation is the purview of design professionals, who have developed a large
toolkit for bracing, strengthening, or otherwise improving the earthquake performance of buildings
and other structures, nonstructural elements, equipment, and contents. These techniques are
discussed in subsequent chapters.
• Secondary damage is typically due to the interaction of several problems, and can be a very complex
issue. It is therefore best mitigated prior to the earthquake, through better handling of materials
and improving of infrastructure. Since this cannot be done everywhere, it must also be mitigated

HAZARD
EARTHQUAKE OCCURS

PRIMARY HAZARDS:
Faulting, Shaking, Liquefaction,
Landsliding, Tsunami…

PRIMARY DAMAGE:
Building / Structural
DAMAGE OR Nonstructural / Equipment
VULNERABILITY

SECONDARY HAZARD /
DAMAGE:
Fire, Hazmat, Flooding…

PRIMARY LOSS:
Life / Injury, Repair Costs, Function,
Communications/Control…

LOSS or RISK SECONDARY LOSS:

Business / Operations Interruption
Market Share, Reputation…

FIGURE 2.2 Earthquake loss process.

© 2003 by CRC Press LLC

0068_C02_fm Page 8 Tuesday, July 30, 2002 8:28 AM

2-8 Earthquake Engineering Handbook

EARTHQUAKE OCCURS

MITIGATION RESULT

Hazard mapping; ground PRIMARY HAZARDS:

remediation; tsunami walls… Demand (hazard)
Faulting, Shaking, Liquefaction,
eliminated or reduced
Landsliding, Tsunami…

Bracing and strengthening,

reduction of mass, base
PRIMARY DAMAGE: Capacity
isolation, structural control… Building / Structural (strength…)
Nonstructural / Equipment increased

Improved storage/infrastructure,
SECONDARY HAZARD / DAMAGE: Secondary demands
better emergency response… Fire, Hazmat, Flooding… eliminated or reduced

PRIMARY LOSS:
Life / Injury, Repair Costs, Function,
Communications/Control…
Improved emergency planning Loss
and response; insurance… avoided or shared
SECONDARY LOSS:
Business / Operations Interruption
Market Share, Reputation…

FIGURE 2.3 Mitigation of earthquake damage.

Fault Trace

Special Studies
Fault Zone

FIGURE 2.4 Typical special studies fault zone map. (Courtesy California Geological Survey)

© 2003 by CRC Press LLC

0068_C02_fm Page 9 Tuesday, August 13, 2002 7:21 AM

Earthquake Risk Management: An Overview 2-9

via improved emergency response, that is, via analysis of the problem, acquisition of the necessary
equipment, and ongoing training and exercises.
• Loss is mitigated via damage control, that is, via improved emergency planning and response.
Since the damage has not been prevented, coping with the damage so as to minimize loss is
necessary. The other dimension of loss mitigation at this stage is financial, that is, earthquake
insurance. Earthquake insurance can be effective in selected circumstances, but it does nothing
for life loss or injury, and typically only partially offsets primary financial loss.
The farther down in the damage process one goes, the more difficult is mitigation. Secondary losses,
such as business interruption, often cannot be fully mitigated.

2.7 Earthquake Risk Management Decision-Making

The previous sections have provided an overview of earthquake risk, its components of hazard, damage,
and loss, and a glimpse of the broad spectrum of mitigation alternatives that a community or organization
has available for dealing with earthquakes. The fundamental problem of earthquake risk management is
not to find a solution, but rather to find which solution is best. The answer to this key problem varies
with each community or organization’s particular circumstances, and is arrived at by following the
decision-making process shown in Figure 2.5.

Estimate the PROBLEM

Risk

Define
Define Problem
Problem

Estimate
Estimate Baseline
Baseline Risk
Risk

Determine N
Determine
Further
Further Action
Action Stop
Stop
Needed
Needed
Y
Internal
Internal Policy
Policy
External
External Select
Select Basis
Basis for
for Analysis
Analysis
Constraints
Constraints

Identify
Identify Alternatives
Alternatives

Examine Mitigation Alternatives Screen

Screen Further
Further
Alternatives Eliminated Alternatives
Alternatives Analysis
Analysis

Choose
Choose Decision
Decision Method
Method

Describe
Describe Alternatives
Alternatives

Values
Values
Uncertainties Collect
Collect and
and Organize
Organize Data
Data
Uncertainties

Apply
Apply Decision
Decision Method
Method

Communicate
Communicate Results
Results

Make
Make
Decision
Decision

Make Decision
ACTION
FIGURE 2.5 Earthquake risk management decision process.

© 2003 by CRC Press LLC

0068_C02_fm Page 10 Tuesday, July 30, 2002 8:28 AM

2-10 Earthquake Engineering Handbook

Although the figure may appear dauntingly complex, the process consists of two basic steps.
1. Estimate the risk by:
• Defining the problem — that is, what are the assets at risk? What do we wish for them? The
assets at risk are often the same as, what is the community or organization responsible for?
What we wish is usually no significant life safety risk, and an acceptable maximum financial loss.
• Quantifying the baseline risk — that is, what is the likely and probable maximum number of
lives to be lost and financial loss, considering potential earthquakes and the assets at risk, under
present conditions? Simply determining the baseline risk is a significant task performed by
specialists, and is described in some detail in later chapters.
• Determining if further action is needed — in some cases, the current risk may be acceptable, or
obviously dealt with in a simple manner. Otherwise, mitigation alternatives need to be devel-
oped, and decisions made between alternatives, in order to reduce the existing risk to a level
that is acceptable.
2. Examine mitigation alternatives by:
• Selecting the basis for analysis — this refers to the basic constraints on the ability of a
community or organization to mitigate the risk. For example, mandated retrofitting of resi-
dential buildings may not be politically feasible — the public may vote out whoever attempts
to enact mandated retrofitting. In such a case, the basis for analysis would be voluntary
retrofitting, and the alternatives consist of distributing information, financial incentives, and
other more palatable measures.
• Identifying alternatives — this consists of identifying as broad a spectrum of mitigation alter-
natives as possible, given the constraints. The previous section indicated how each link in the
chain of causation of earthquake loss offers opportunities for mitigation, often multiple oppor-
tunities. The goal at this stage is creative thinking, to develop as many possible alternatives as
possible.
• Screening alternatives — having brainstormed a broad range of alternatives, some can be
eliminated for obvious reasons.
• Choosing a decision method — this consists of purposefully developing a framework and criteria
for making a decision, and is a very important step in the earthquake risk management process,
as the basis for the decision will be closely examined. There are many different methods and
criteria for deciding between alternatives, including simple scoring methods, benefit–cost
analysis, and multiattribute utility theory. The decision as to whether that risk is acceptable or
not should be based on two fundamental criteria:
• What loss would be tolerable, should it occur?
• How easy (i.e., at what cost) would it be to reduce the loss?
The decisions being considered here are financial in nature and, in general, significant risk of structural
collapse4 should not be tolerated today. In that sense, some aspects of the decision-making may be easy,
for example, if the baseline risk assessment indicates that collapse is likely or there is other significant
risk to life-safety, then that risk must be reduced. What we discuss here is, given that life-safety has been
addressed, then to what additional extent should earthquake risk be reduced, that is, what makes oper-
ational and financial sense? There are a variety of various methods available for making that decision
(see AIChE 1995), which we summarize in this section.

2.7.1 Maximization (Benefit–Cost) or Minimization (Total Cost )

The most common decision-making method is maximization, which forms the core of the well-known
benefit–cost method. In this method, the decision-maker quantifies and seeks to maximize the net benefits

4 Or other risks to life-safety, such as major releases of hazardous materials in an earthquake.

© 2003 by CRC Press LLC

0068_C02_fm Page 11 Tuesday, July 30, 2002 8:28 AM

Earthquake Risk Management: An Overview 2-11

Cost

Total Cost = CL + CM

C (satisficed)
A
Least total cost

CM =Cost of mitigation

B CL = Expected loss

Mitigation
(increasing effectiveness)

FIGURE 2.6 Typical cost of mitigation curve.

(i.e., benefit minus cost associated with the alternative, where the benefit in this case is the reduced loss)
with each alternative. The corollary of this is the minimization of the total cost, where the total cost is
the expected loss plus the cost of mitigation. This is shown in Figure 2.6, where point A is the least total
cost (note that point A does not necessarily correspond to point B, the crossing of the expected loss curve
with the mitigation cost curve).
In order to apply these rules, each alternative must be associated with a single value that represents
benefits. When uncertainty is ignored, applying the maximization rule is straightforward and expresses
the decision-maker’s simple desire for more of a good thing. When considering uncertainty, the usual
procedure for applying the maximization rule is to determine the likelihood of the possible outcomes of
an alternative, and compute an expected value for each alternative. The expected value is the weighted
sum of values for each possible outcome, with the weights determined by the likelihood (probability) of
these outcomes.

2.7.2 Maximin and Maximax

Another common decision rule is the maximin rule (maximize the minimum). It can be applied in
situations where multiple outcomes are possible, but there are likelihoods that are not characterized.
Using this rule the decision-maker takes each decision and associates it with the value resulting from the
worst possible outcome that could occur under that alternative. The alternative that provides for the
highest value under its worst possible outcome is selected. The maximin rule is truly a pessimist’s rule,
for it ensures that the decision-maker avoids choosing alternatives that could lead to the worst possible
outcomes, even if these alternatives could yield attractive outcomes as well. On the contrary, the optimist’s
rule is the maximax rule (maximize the maximum). Using this rule, the decision-maker selects the
alternative whose return under the most favorable circumstances is the greatest. In this case, the decision-
maker is “going for broke” by ignoring all possible downsides to each alternative, and focusing only on
the possible benefits [AIChE, 1995].

2.7.3 Minimum Regret

A decision-maker may be criticized long after a decision has been made by others who operate with
perfect hindsight. Perfect hindsight may reveal that a different alternative than the one chosen would
have led to a better outcome under the circumstances that occurred subsequent to the decision. For
example, a plant manager chooses to delay seismic retrofit due to its high cost and a judgment that there
is an extremely small chance of a major earthquake in the next 5 years. An earthquake occurs soon after
this decision, which causes a 30-day interruption of operations. Even though the decision was in fact a
wise one, given the decision-maker’s understanding of the costs and risks, the plant manager is criticized

© 2003 by CRC Press LLC

0068_C02_fm Page 12 Tuesday, July 30, 2002 8:28 AM

2-12 Earthquake Engineering Handbook

for not choosing the alternative that would have yielded the best outcome. The regret rule chooses to
minimize the risk to the decision-maker of being adversely confronted in hindsight. For each possible
future scenario, the decision-maker assesses the difference in value between each alternative and the best
alternative given that the scenario was to occur. This difference is referred to as the regret. Regret represents
the cost of not having perfect foresight and selecting the best decision. This rule then applies a minimax
rule, selecting the decision in which the maximum regret is the least.

2.7.4 Minimum Uncertainty

Most decision-makers are risk-averse: they prefer to avoid choosing alternatives for which the outcomes
are uncertain. When outcomes are presented to a decision-maker in terms of the probability for each of
the possible values that could result, another decision rule, called minimum uncertainty, can be employed.
Minimum uncertainty may also be referred to as minimum risk, where risk refers not to physical or
financial risks specifically but to aversion to uncertainty as described above. With outcomes quantified
by a probability distribution, extremely risk-averse decision-makers may wish to select an alternative that
gives them the most certainty about future rewards by minimizing the variance on possible outcomes.
Alternatively, some decision-makers choose to account for only those outcomes associated with the most
likely future event. A minimum uncertainty rule can lead the decision-maker to reject attractive but
uncertain alternatives, and instead choose a safer alternative with smaller potential rewards.

2.7.5 Satisficing
Not all decision rules yield a single best alternative. Many organizations are concerned with meeting
many, sometimes contradictory, goals. Satisficing is a decision rule that is used to find one or more
alternatives that can satisfy all (or most) of the organization’s goals rather than determining the best
alternative. When using a satisficing decision rule, a minimum level of achievement on each of several
goals is set. For example, in a safety decision the goals may be to “reduce the risk of death below 10–7
per year” and “spend less than $100,000.” All alternatives that meet these goals would be considered
satisfactory.
An example of satisficing is shown in Figure 2.6, at point C, which is not the least cost, but a point
where the marginal cost of mitigation is a set ratio to the expected loss. Since many total cost curves are
rather flat, a near-minimum total cost can be achieved at a cost of mitigation far below the cost of
mitigation corresponding to the true minimum point A.

2.7.6 Next Steps

Any of the above methods can be employed in decision-making. After a decision has been made, the
next steps in the process are:
Describing alternatives, which consists of examining each alternative, including its impact on the
baseline risk, using the same methods as were used in determining the baseline risk, but this time
in the light of the mitigation alternative.
Collecting and organizing the data, which refers to assembling the various alternatives, their costs and
impacts on baseline risk. An important element in the various data is values, and the confidence
(or uncertainty) associated with each alternative.
Applying the decision method, to prioritize, or determine the most appropriate alternative(s).
Communicating the results to decision-makers for their action. The elements of communication here
are the basis, logic and range of alternatives examined, and the sensitivity of the findings to key
assumptions and parameters. Different stakeholders (such as tenants, owners, parents, rate-payers,
and others) will be interested in different aspects of the decision, so that their interests need to
be identified and addressed.

© 2003 by CRC Press LLC

0068_C02_fm Page 13 Tuesday, July 30, 2002 8:28 AM

Earthquake Risk Management: An Overview 2-13

Making the decision, which is not always as easy as it sounds, so that questions may arise, and further
analyses may be required. Eventually however, a set of alternatives will emerge.
Implementing the plan, which is discussed below, and more fully later in this handbook.

2.8 Earthquake Risk Management Program

Given that a decision to implement a set of alternatives has been made, the last step in earthquake risk
management is implementing the alternatives. This consists of developing a program, which typically
will consist of one or more of the following steps (shown in Figure 2.7):
• Funding — there are a wide variety of ways to pay for earthquake risk mitigation, depending on
the community or organization, including (for the public sector) from general revenues, special
assessments, bonds, user’s fees, and/or grants from state or federal programs, to name a few. In
the private sector, analogous sources of expensing, one-time charges, loans or bonds, and/or public
subsidy exist.

Decide on Mitigation Alternatives

Obtain Funding

Program Management
Dedicate staff, find/retain specialists, consultants and/or
schedule the work, communicate with neighbors, and affected parties

Implement CIP/Other Projects

Multi-year implementation

Residual Risk

Risk Transfer Emergency Plan

Develop as interim solution,
Insurance, capital markets...
and for ongoing residual risk

After the Earthquake

Maintain emergency plan, have backup sites ready,
engineers on-call…

FIGURE 2.7 Earthquake risk mitigation program.

© 2003 by CRC Press LLC

0068_C02_fm Page 14 Tuesday, July 30, 2002 8:28 AM

2-14 Earthquake Engineering Handbook

• Project management — putting the plan into action involves dedicating staff for managing the
program, bringing in specialists, consultants and/or contractors, scheduling the work, communi-
cating with neighbors and affected parties, and following up to assure goals are being met.
• Implementing the plan – performing the various projects. Major strengthening and other large
projects will usually be done as part of the capital improvement plan (CIP), while contents bracing
and smaller projects may be done as part of maintenance, or out of general revenues.
• Risk transference is a more general term for what is usually thought of as insurance, and which
really can only transfer parts of the financial risk to third parties. Recent innovations in the capital
markets have added new dimensions to this kind of mitigation alternative, which may make it
more viable than ordinary insurance previously was.
• What happens after the earthquake is a consideration that should always be borne in mind. That
is, a community or organization should always have an emergency response plan in place, which
considers all assets at risk and accounts for unknowns. The basis for this plan should be the
residual risk, that is, the portion of the baseline risk that the implemented mitigation alternatives
cannot feasibly reduce. Understanding, preparing for, and clearly communicating this residual
risk, that is, the loss that has been deemed acceptable, is necessary to avoid surprises and unnec-
essary recriminations following the earthquake. It also should form the basis for a considered
recovery plan.

2.9 Summary
This chapter has provided an overview of the earthquake risk problem and a guide to the management
of this problem. In summary, earthquake risk management consists of a series of rational steps aimed at
1. Identifying what is at risk — that is, what assets could be lost due to an earthquake
2. Assessing how the earthquake places these assets at risk
3. Determining which alternatives might reduce this risk
4. Deciding among these alternatives, which are the best for the specific situation
5. Implementing — that is, putting these alternatives into practice

Defining Terms
Attenuation — The rate at which earthquake ground motion decreases with distance.
Damage — Physical disruption, such as cracking in walls, overturning of cabinets, etc., often used
synonymously with loss.
Hazard — The potential for or occurrence of natural phenomena, such as ground shaking, liquefaction,
tsunami, etc., which might lead to loss.
Intensity — A metric of the effect, or the strength, of an earthquake hazard at a specific location,
commonly measured on qualitative scales such as MMI, MSK, and JMA.
Loss — The human or financial consequences of damage, such as human injury or cost of repairs.
Mitigation — Literally the moderating of a force or intensity of something that causes suffering; used
in earthquake engineering as synonymous with reducing earthquake risk.
Risk — The potential for loss. Risk can be expressed in absolute terms such as the risk of collapse, or
in probabilistic terms, such as “the risk per year is $1,000.”
Satisficing — A decision rule that is used to find one or more alternatives that can satisfy all (or most)
of the organization’s goals rather than determining the best alternative.

References
AIChE, 1995. Tools for Making Acute Decisions, with Chemical Process Safety Applications, Center for
Chemical Process Safety of the American Institute of Chemical Engineers, New York.

© 2003 by CRC Press LLC

0068_C02_fm Page 15 Tuesday, July 30, 2002 8:28 AM

Earthquake Risk Management: An Overview 2-15

Richter, C.F., 1957. Elementary Seismology, W.H. Freeman, San Francisco.

Seismic Safety Commission, 1999. Earthquake Risk Management — A Toolkit for Decision-Makers, SSC
report 99–04, California Seismic Safety Commission, Sacramento. (Available in PDF format at
www.seismic.ca.gov.)

Further Reading
There is a wide variety of information and resources readily available on various aspects of earthquake
risk management. The reader is referred to the following key publications:
Seismic Safety Commission, Earthquake Risk Management – A Toolkit for Decision-Makers, SSC report
99–04, California Seismic Safety Commission, Sacramento, 1999. This is an integrated group of
three documents meant to assist local governments in developing a successful earthquake risk
reduction program. It can be downloaded over the Web in PDF format at www.seismic.ca.gov.
California Governor’s Office of Emergency Services (OES), Emergency Planning Guidance for Local
Government. This document addresses emergency planning at the city and county level. It has
evolved from the insight and experience gathered from past disasters and the cooperation between
OES staff and local government emergency managers.
Additionally, there is a wealth of information available on the Internet:

Site URL Features

Applied Technology Council (ATC) www.atcouncil.org Information about the council’s

mission and organization; recent
news releases; newsletters; recent
reports, briefs, and training
manuals; databases; seminars and
workshops; and other ATC products.
Assn. Bay Area Governments (ABAG) www.abag.org Maps of seismicity for San Francisco
Bay area; information on building
vulnerability.
California Division of Mines and http://www.consrv.ca.gov/cgs/ CDMG has primary responsibility for
Geology (CDMG) work in the geosciences in
California, including hazard
mapping, earthquake scenarios, a
strong motion instrumentation
program, and policy
implementation.
California Seismic Safety www.seismic.ca.gov Provides information about the
Commission (CSSC) commission, its publications,
pending legislation relevant to
seismic hazards in California, and a
table of significant California
earthquakes.
Consortium of Universities for www.curee.org CUREE is a consortium of U.S.
Research in Earthquake Engineering universities for coordination of work
(CUREE) on earthquake research problems.
Disaster Research Center (DRC) www.udel.edu/DRC/ DRC is the first social science research
center in the world devoted to the
study of disasters.
Earthquake Engineering Research www.eerc.berkeley.edu/eea Earthquake Engineering Abstracts;
Center (EERC) covers the world literature in
earthquake engineering since 1971.
Contents include selected technical
reports, conference papers,
monographs, and journal articles.
Most materials are available for loan
from the PEER library (University of
California at Berkeley).

© 2003 by CRC Press LLC

0068_C02_fm Page 16 Tuesday, July 30, 2002 8:28 AM

2-16 Earthquake Engineering Handbook

Site URL Features

Earthquake Engineering Research www.eeri.org N\EERI is the de facto U.S. national

Institute (EERI) society for earthquake engineering
— contains useful information and
links.
EQNET www.EQNET.org Excellent database of earthquake
information sources on the Web.
Federal Emergency Management www.fema.gov Many free publications of high
Agency (FEMA) caliber, useful information.
Global Seismic Hazard Assessment http://seismo.ethz.ch/GSHAP/ A collection of seismotectonic models
Program (GSHAP) index.html and hazard maps, for various
portions of the globe. The project
terminated in 1999.
Japan NDRPI http://www.bosai.go.jp/ National Disaster Prevention
Research Institute, maintains real-
time seismic net (Kyoshin) and
other data (English and Japanese).
Mexico National Seismological http://www.ssn.unam.mx/SSN/ Maps and other information on
Service Sismos/region_sismica_mx.html seismic hazards in Mexico (in
Spanish).
Multidisciplinary Center for mceer.buffalo.edu/ MCEER bibliographies; literature and
Earthquake Engineering Research research guides, and list of FEMA
(MCEER) and NEHRP guidelines and
handbooks for earthquake resistant
construction and design. The latter
category includes: Seismic Safety of
Lifelines, Seismic Safety of Existing
Buildings, and Seismic Safety of New
Buildings.
Natural Hazards Research and www.colorado.edu/hazards/litbase/ An excellent on-line index that
Applications Information Center litindex provides bibliographic access only to
(NHRAIC) that collection of NHRAIC at the
University of Colorado at Boulder.
While HazLit is not a full-text
database, and the Hazards Center
Library does not loan its holdings to
the general public, this is an excellent
resource.
Pacific Earthquake Engineering peer.berkeley.edu PEER is a consortium of earthquake
Research (PEER) Center engineering research universities in
the western United States — the Web
site has their reports, much useful
information.
Southern California Earthquake www.scec.org SCEC is a Science and Technology
Center (SCEC) Center of the National Science
Foundation. Home Page contains
background information about
SCEC and links to its many member
academic institutions. Both the
SCEC newsletter and SCEC
publications list are available from
this site. Also, check out the
Earthquake Hazard Analysis Map —
a map of probable future Southern
California earthquakes.
Structural Engineers Association of www.seaoc.org SEAOC’s site is useful for keeping up
California (SEAOC) with research specific to structural
engineering, especially buildings,
and with the practice.

© 2003 by CRC Press LLC

0068_C02_fm Page 17 Tuesday, July 30, 2002 8:28 AM

Earthquake Risk Management: An Overview 2-17

Site URL Features

Swiss Seismological Service http://seismo.ethz.ch/ Excellent source of information on

earthquakes and other hazards,
including Global Seismic Hazard
Assessment Program (GSHAP)
Turkey, Disaster Management and http://www.metu.edu.tr/home/ Abundant information on Turkish
Implementation and Research wwwdmc/main-e.html earthquakes, at Middle East
Center Technical University (METU)
Turkey, Kandilli Observatory and http://www.koeri.boun.edu.tr/ Observatory with abundant
Earthquake Research Institute geophy/anasayfa/eanafr.html information on Turkish
earthquakes, at Bogazici University.
University of Washington Earthquake www.geophys.washington.edu/ Excellent general site, with directory
page seismosurfing to many other sites related to
earthquakes.
U.S. Geological Survey Earthquake quake.wr.usgs.gov Excellent source of information on
Information many aspects of earthquakes.

© 2003 by CRC Press LLC