WBI Risk Analysis

1
Instruments and Approaches in Risk Analysis
Session 3 Dr. Bijan Khazai
Risk Analysis Fundamentals of Risk Analysis
2
Learning objectives

• Learn
• Probabilistic concept of risk.
• Concepts of return period, loss-frequency curve,
  average annual loss.
• Sources of uncertainty in risk analysis.
• Understand
• Difference between probabilistic and
  deterministic risk analysis.
• Outcomes of a scenario analysis.
• Depiction and ranking of risk through risk matrix
  and risk indexing.
• Components of a loss estimation model.
• Methodology of a cost-benefit analysis.
• Uses of a participatory risk analysis.

3
Risk

• There are two approaches to defining risk
• Probabilistic Approach
• Risk is defined as the likelihood (i.e.,
  probability) of sustaining a certain level of
  loss during a given time period.
• Risk Probability of an event occurring x impact
  of the event
• Deterministic Approach
• The geographical distribution of the severity of
  loss due to the occurrence of a postulated event
  (i.e., Scenario).

4
Probabilistic Risk Analysis

• Probabilistic Risk Assessment answers three basic
  questions
• What and where are the initiators or initiating
  events?
• What and how severe are the potential
  consequences?
• How likely will these consequences occur
  (probability or frequency)?
• In a PRA, risk is characterized by two
  quantities
• The magnitude (severity) of possible adverse
  consequences
• The likelihood (probability) of occurrence of
  each consequence

Value and Fragility of Exposed Assets
x

1
5
10
Probability earthquake intensity will be exceeded
during a particular time period (T 200 yrs)
Probable losses (Billions Euros)
5
Uncertainty

• Estimating risk is fraught with lots of
  uncertainty.
• Dealing with uncertainty is the essence of risk
  analysis..
• Uncertainty is the state of having limited
  knowledge.
• In probabilistic risk analysis, we can account
  for uncertainty.
• Uncertainties are inherent (for example) in

6
Return Period

• Return period also known as a recurrence interval
  is an estimate of the interval of time between
  events.
• An event with a 100-year recurrence interval will
  on average only occur once every 100 years.

• Return period (RP) is the inverse of the
  probability that the event will be exceeded in
  any one year (Exceedance Probability, EP).

Return Period, in years (RP) Probability of occurrence in any given year Annual Probability of Exceedance (EP) Probability of Exceedance in 50 years
500 1 in 500 0.2 10
200 1 in 200 0.5 25
100 1 in 100 1 50
50 1 in 50 2 100
10 1 in 10 10 500
7
Exceedance Probability Curve

• The key relationship in managing risk is the
  exceedance probability (loss frequency) curve,
  which represents the relationship between
  frequency (Exceedence Probability) and severity
  (amount of losses).
• Area under curve Average Annual Losses (AAL)

AAL ? probability that event occurs loss
associated with event
10
8
6
rare events, apply with caution!
Annual Probability of Exceedance ()
100 year loss (1 EP) US 125 million
4
2
0
50
100
150
200
250
300
0
Loss Amount (US Million)
8
Multiple Risk Mapping - AAL

• Average Annual Loss (AAL) presents a rational way
  to integrate risk associated with various
  hazards.
• An AAL value is calculated in each cell for any
  desired factor (e.g., Loss of housing units,
  economic loss, casualty, etc.).

Cell Hazard Hazard Hazard Hazard Hazard Hazard Hazard Hazard Hazard
Cell Earthquake (AAL) Earthquake (AAL) Earthquake (AAL) Flood (AAL) Flood (AAL) Flood (AAL) Earthquake Flood (AAL) Earthquake Flood (AAL) Earthquake Flood (AAL)
Cell Housing Units Economic Loss (1000) Casualties Housing Economic Loss (1000) Casualties Housing Units Economic Loss (1000) Casualties
1 10 150 5 5 90 3 15 240 8
2 15 300 14 25 50 15 40 350 29
3 30 130 20 5 70 6 35 200 26
4 1 20 5 10 20 8 11 40 13
5 0 35 0 27 14 15 27 49 15
6 5 12 2 4 10 2 9 22 4
7 8 50 1 7 35 1 15 85 2
Total 69 697 47 83 289 50 152 986 97
9
Risk Matrix Analysis

• Risk Matrix presents a visual two-dimensional
  display of the ranking of the risk for a
  region
• frequency and severity scale that is relevant to
  the region of interest .
• The scale will help in interpreting historical
  experience and translating expert opinion in a
  consistent manner.
• It is a simple approach for setting priorities.

10
Scenario Analysis

• A deterministic risk analysis or a scenario
  analysis is the process of analyzing the
  consequences - damages, losses or impacts - from
  a single postulated hazard event, e.g. a
  scenario.
• Choice between a probabilistic risk analysis and
  scenario analysis depends on the aims of the
  study.
• Scenarios Analysis (scenarios such as the worst
  case scenario) can be used in establishing a
  disaster risk management plan.
• Scenario events are not meant predict the next
  event.
• Different scenarios should be simulated to
  develop a comprehensive understanding of the
  potential impacts.

11
Loss Estimation Models
Loss estimation software packages can be used to
identify the risk profiles and risk parameters of
a region under the simulated events and display
them in maps.
1. POTENTIAL HAZARDS
2. INVENTORY DATA
3. DIRECT DAMAGE
4. INDUCED DAMAGE
5. SOCIAL LOSSES
6. ECONOMIC LOSSES
7. INDIRECT LOSSES
12
Cost-Benefit Analysis

• Cost-Benefit analysis (CBA) is a systematic
  procedure for evaluating decisions that have an
  impact on society.
• CBA provides comparison between the upfront
  investment costs of mitigation and the benefits
  of mitigation.
• CBA uses results of risk analysis to ensure
  effectiveness of protective measures by
• Balancing the various interests of the
  stakeholders,
• Considering the reasonability (or utility) of
  measures, and
• Enabling consensus on strategies and measures to
  reduce risks
• Clarify objectives
• At outset, check objectives of conducting a CBA.
• Combination of scoping exercises, shared learning
  dialogues and more qualitative.

13
Cost-Benefit Analysis Methodology

• Cost-Benefit Analysis in the context of DRM
  requires
• Assessment of risk, and
• Assessment of avoided risks.

• Hazard
• Intensity
• Recurrence

Analysis of risk reduction potential impacts
with risk management
Risk Analysis potential impacts without risk
management

• Vulnerability
• Exposure
• Fragility
• Capacities

Net benefits Reduction of the potential
impacts less Cost of risk reduction
Source Mechler, Reihnard
14
Participative Risk Analysis

• A Participatory Risk Analysis is carried out with
  the participation of an affected target groups in
  cooperation with the experts and decision makers.
• Participatory Appraisal sociological approach
  of rapid, action-oriented assessment of local
  knowledge, needs, and potentials.

Source GTZ, Reconstruction after Hurricane Stan

• PRA can be used in
• development of disaster emergency plans,
• arrangement and introduction of early warning
  systems for flooding
• implementation of disaster prevention exercises,
• development of prevention-based planning methods,
  in which risk analysis is part of the planning
  process.

15
Summary

• Probabilistic concepts
• Uncertainty, Return Period, Exceedance
  Probability, Loss-Frequency Curve, AAL.
• Probabilistic Risk Analysis
• All potential impacts corresponding to a specific
  return period is assessed.
• Scenario (or Deterministic) Analysis/Loss
  Estimation Models
• Impact assessed for a postulated event over a
  defined spatial area.
• Risk Indexing/Mapping
• Key indicators of risk are aggregated in order to
  set priorities.
• Risk Matrix Analysis
• Visual two-dimensional display of the ranking
  of the risk.
• Cost Benefit Analysis
• Upfront investment costs of mitigation vs. the
  benefits of mitigation.
• Participative Risk Analysis