ArticlePDF Available

A composite indicator model to assess natural disaster risks in industry on a spatial level

Authors:

Abstract and Figures

In the event of natural disasters, industrial production sites can be affected by both direct physical damage and indirect damage. The indirect damage, which often exceeds the direct ones in value, mainly arises from business interruptions resulting from the impairment of information and material flows as well as from domino effects in interlaced supply chains. The importance of industry for society and the domino effects often result in severe economic, social, and environmental consequences of industrial disasters making industrial risk management an important task for risk managers at the administrative level (e.g. civil protection authorities). Since the possible industrial disaster damage depends not only on hazard and exposure but also on the vulnerability of a system, an effective and efficient industrial risk management requires information about the system's regionalized vulnerability. This paper presents a new methodology for structural industrial vulnerability assessment based on production factors that enables to assess the regional industrial disaster vulnerability. In order to capture industry-specific vulnerability factors and to account for the processes underlying regional industrial vulnerability, a two-stage approach is developed. This approach combines a composite indicator model to assess sector-specific vulnerability indices (V-s) with a new regionalization method. The composite indicator model is based on methodologies from the field of multicriteria decision analysis (MultiAttribute Value Theory) and the Decision-Making Trial and Evaluation Laboratory Method is applied to correct the (V-s) for interdependencies among the indicators. Finally, the developed approach is applied to an exemplar case study and the industrial vulnerability of 44 administrative districts in the German federal state of Baden-Wuerttemberg is assessed.
Content may be subject to copyright.
This article was downloaded by: [Universitetsbiblioteket i Agder]
On: 18 April 2014, At: 01:48
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Risk Research
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/rjrr20
A composite indicator model to assess
natural disaster risks in industry on a
spatial level
Mirjam Merza, Michael Hieteb, Tina Comesa & Frank Schultmanna
a Institute for Industrial Production (IIP), Karlsruhe Institute for
Technology (KIT), Karlsruhe, Germany.
b Center for Environmental Systems Research (CESR), University of
Kassel, Kassel, Germany.
Published online: 20 May 2013.
To cite this article: Mirjam Merz, Michael Hiete, Tina Comes & Frank Schultmann (2013) A
composite indicator model to assess natural disaster risks in industry on a spatial level, Journal of
Risk Research, 16:9, 1077-1099, DOI: 10.1080/13669877.2012.737820
To link to this article: http://dx.doi.org/10.1080/13669877.2012.737820
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions
A composite indicator model to assess natural disaster risks in
industry on a spatial level
Mirjam Merz
a
, Michael Hiete
b
*, Tina Comes
a
and Frank Schultmann
a
a
Institute for Industrial Production (IIP), Karlsruhe Institute for Technology (KIT),
Karlsruhe, Germany;
b
Center for Environmental Systems Research (CESR), University of
Kassel, Kassel, Germany
(Received 24 August 2011; nal version received 24 August 2012)
In the event of natural disasters, industrial production sites can be affected by
both direct physical damage and indirect damage. The indirect damage, which
often exceeds the direct ones in value, mainly arises from business interruptions
resulting from the impairment of information and material ows as well as from
domino effects in interlaced supply chains. The importance of industry for soci-
ety and the domino effects often result in severe economic, social, and environ-
mental consequences of industrial disasters making industrial risk management
an important task for risk managers at the administrative level (e.g. civil protec-
tion authorities). Since the possible industrial disaster damage depends not only
on hazard and exposure but also on the vulnerability of a system, an effective
and efcient industrial risk management requires information about the systems
regionalized vulnerability. This paper presents a new methodology for structural
industrial vulnerability assessment based on production factors that enables to
assess the regional industrial disaster vulnerability. In order to capture industry-
specic vulnerability factors and to account for the processes underlying regio-
nal industrial vulnerability, a two-stage approach is developed. This approach
combines a composite indicator model to assess sector-specic vulnerability
indices (V
s
) with a new regionalization method. The composite indicator model
is based on methodologies from the eld of multicriteria decision analysis (Mul-
tiAttribute Value Theory) and the Decision-Making Trial and Evaluation Labora-
tory Method is applied to correct the (V
s
) for interdependencies among the
indicators. Finally, the developed approach is applied to an exemplar case study
and the industrial vulnerability of 44 administrative districts in the German fed-
eral state of Baden-Wuerttemberg is assessed.
Keywords: industrial disaster risk; indirect losses; vulnerability assessment;
composite indicator; multicriteria decision analysis
1. Introduction
Natural and man-made hazards such as earthquakes, storms, oods, and terrorist
attacks constitute a major risk to industrial productions sites (Kleindorfer and Saad
2005). Recent events, such as the Great East Japan Earthquake and Tsunami 2011
and the Hurricane Katrina 2004, stressed the fact that natural disasters can cause
devastating damage to industrial production systems. The characteristics and extent
of the damage vary with the type of hazard and industry affected (Hiete and Merz
*Corresponding author. Email: hiete@uni-kassel.de
Journal of Risk Research, 2013
Vol. 16, No. 9, 10771099, http://dx.doi.org/10.1080/13669877.2012.737820
Ó2013 Taylor & Francis
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
2009). In general, industrial production sites are affected by both direct and indirect
disaster impacts (Okuyama 2007) (see Figure 1).
Direct disaster impacts comprise all physical damage to buildings, materials, and
production equipment, the obstruction of workers as well as the interruption of
critical infrastructure systems (e.g. electricity supply and water supply) (Messner
and Meyer 2005). The so-called primary direct disaster impacts notably physical
damage to production facilities and infrastructure systems may also cause second-
ary direct impacts. Though rare, Natural-Disaster Triggered Technological Accidents
(NATECHs) play an important role due to their potentially devastating environmen-
tal and social consequences (Cruz and Okada 2008; Steinberg, Sengul, and Cruz
2008). An example for a NATECH is the release of hazardous materials from a
chemical plant destroyed by an earthquake.
Indirect disaster impacts are mainly caused by business interruptions resulting
from the impairment of material and information ows (Messner and Green 2007).
Due to the integration of industrial companies within complex and often globally
interlaced supply networks, domino effects are present, and business interruptions
propagate into companies that have not been affected by the original event. This
leads to a spatial and temporal extension of the negative disaster impacts within
industrial production systems (Green and van der Veen 2007). Therefore, indirect
disaster losses may exceed direct disaster losses economically by multiple times
(Kleindorfer and Saad 2005).
In order to reduce the industrial losses caused by natural disasters, well-structured
disaster risk management procedures comprising all phases of the risk management
cycle (preparedness, prevention or mitigation, response, and rehabilitation or recon-
struction) need to be installed (Sauerwein and Thurner 1998). To prevent business
interruptions, it is, for example, mandatory to protect infrastructure systems and to
allocate emergency resources (e.g. emergency power supply) according to the
specic risk level of the individual regions. Due to the environmental, social, and
economic impacts of industrial disasters, risk management strategies (e.g. measures
for prevention or mitigation) should not only be implemented on the level of
Natural hazard
physical damage to
buildings, material,
production equipment
Interruption of
critical infrastructures
Primary business interruptions
Obstruction
of staff
Supply chain
disruption Primary business interruptions
Affected production site
supply chain members
primary
secondary
NATECH:
e.g. explosion/
fire, hazmat release
direct effects
indirect effects
Figure 1. Impacts of natural disasters on industrial production systems.
1078 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
industrial companies but also on an administrative (spatial) level (Dastous, Nikiema,
and Megaloconomos 2008; Gheorghe, Mock, and Kröger 2000).
An effective and efcient risk management requires detailed knowledge about
possible damage, which depends not only on hazard and the exposure but also on
the vulnerability of the system. Due to the economic importance of indirect disaster
impacts, this paper analyzes, in particular, the vulnerability against business inter-
ruptions caused by natural disasters. Within the existing literature, the assessment of
the industrial vulnerability is very scarce. Here, mainly the social, economic, and
ecologic vulnerability of society is assessed (Birkmann 2006b).
For both risk and vulnerability assessment, various methodologies can be
applied. Due to the complex and abstract concept of vulnerability, the vulnerability
of a system or a region cannot be measured directly (Gall 2007). Therefore, the pro-
cess of operationalization aims at dening and expressing vulnerability in terms of
related observations or measurements. One possibility to operationalize vulnerability
is the development of (composite) indicator models (Birkmann 2007; Merz, Hiete,
and Schultmann 2010). Within the 2005 Hyogo Framework of Action, the UN
explicitly calls for the development of composite indicators for risk and vulnerabil-
ity assessment (UN/ISDR 2005), as the results of an indicator-based vulnerability
assessment constitute important information for risk mitigation and resource alloca-
tion. Therefore, within this paper, a composite indicator model to assess the indus-
trial vulnerability against indirect disaster losses and its regionalization are
presented. Its application is exemplarily shown in a case study for the German
federal state of Baden-Wuerttemberg.
The remainder of the paper is organized as follows. The next section highlights
the relevance of vulnerability assessment in industrial risk management. Section 3
describes the use of composite indicators for vulnerability assessment. In Section 4,
the methodological approach used to develop the composite indicator model is spec-
ied. Here, the operationalization of the industrial disaster vulnerability on the level
of industrial sectors is presented, and it is shown how sector specic results can be
assigned to a spatial level of analysis. In Section 5, the developed indicator model
is applied to an exemplar case study. Here, results of the vulnerability assessment
for 16 industrial sectors as well as for the regional vulnerability in the German
federal state of Baden-Wuerttemberg are shown. Finally, in Section 6, the main
ndings are summarized in the conclusions.
2. Vulnerability assessment for risk management
An effective and efcient risk management requires the knowledge and prioritiza-
tion of risks. According to ISO 31000:2009 Risk Management Standard, the risk
management process can be split up into four iterative steps (Purdy 2010): risk
identication, risk analysis, risk evaluation, and risk treatment. The rst three steps
are aimed at an assessment of the risks as a prerequisite for implementing risk
mitigation measures (Baas et al. 2008).
In general, risk R
h
is dened as the product of the probability of occurrence p
h
of a hazard h and the consequences C
h
of this hazard (Leitch 2010):
Rh¼phChð1Þ
Journal of Risk Research 1079
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
In the analysis of natural disaster risk, the risk Ris often expressed as a function
of hazard (characterized by the associated intensity and probability) and vulnerabil-
ity (UNDP 2004).
R¼fðHazard;VulnerabilityÞð2Þ
Here, the vulnerability describes the characteristics and circumstances of a
community, system or asset that make it susceptible to the damaging effects of a
hazard(UN/ISDR 2004).
Since neither the intensity nor the probability of a natural hazard can be
controlled, risk management focuses on the vulnerability as the internal side of risk.
Vulnerability assessment, which plays an important role within the disaster risk
assessment procedure (Birkmann and Wisner 2006), should address the following
components (European Commission [EC] 2010): assessment of likely impacts, iden-
tication of vulnerability factors, identication of coping capacities and resilience
factors, and identication of elements potentially at risk (on a spatial level).
Starting point of the vulnerability assessment procedure is the development of a
theoretical vulnerability framework (Birkmann 2006b). Due to the multifaceted and
complex nature of vulnerability, the denition of the conceptual framework is a
challenging task (Twigg 2001), and the components and dimensions in the frame-
work depend on the specic objectives and desired level of assessment (Thywissen
2006; Villagrán de Léon 2006b). In most vulnerability frameworks, fragility and
resilience are central components (e.g. Alexander 2000; Birkmann and Wisner
2006; Dilley et al. 2005; Hahn 2003; and Turner et al. 2003). While fragility
describes the susceptibility of a system, resilience represents the coping capacity
(Hossini 2008). Therefore, fragility enhances and resilience reduces vulnerability. If
the vulnerability of a region or a community is assessed, the exposure must be inte-
grated as a further component into the vulnerability framework. Here, exposure
describes the number and the value of the considered elements at risk. Thus, the
exposure represents the spatial component in the concept of vulnerability. If the
industrial vulnerability (V
r
) of different regions r is assessed, the following
denition of vulnerability can be used:
Vr¼fðFragility;Resilience;ExposurerÞð3Þ
3. Composite indicator models for vulnerability assessment
In general, indicator systems are management tools which describe and operational-
ize complex system characteristics in a transparent way. The operationalization of
vulnerability using composite indicators bridges the gap between the theoretical vul-
nerability framework and decision-making within disaster risk management.
A composite indicator model is the mathematical combination of a thematic set
of indicators that represents different dimensions of a concept, which cannot be
operationalized by a single indicator (Nardo et al. 2005a). Within a composite indi-
cator model, each individual indicator is a quantitative or qualitative measure
derived from observed facts depicting the reality of a complex situation (Cutter,
Burton, and Emrich 2010; Freudenberg 2003). Composite indicator models do not
result in absolute measures. They rather reveal the relative position of a system (or
a region) with regard to the operationalized characteristic, here, its vulnerability
1080 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
(Nardo et al. 2005b). Composite indicator models are recognized as a useful tool
for decision-making and public communication. The main benet of a composite
indicator model is the transparent and well-structured processing of multidimen-
sional characteristics. This leads to a reduced complexity and an enhanced
understanding of the situation (Belton and Stewart 2002).
In different disciplines, numerous approaches using composite indicators have
emerged (Cutter, Burton, and Emrich 2010). In the eld of disaster risk manage-
ment, indicator systems are regarded as a base for monitoring and increasing the
capacity in risk management and achieving societal consensus on disaster risk
reduction measures (cf. Cutter, Burton, and Emrich 2010; UN/ISDR 2005). Here,
mainly, approaches to assess the social vulnerability of different regions are
developed at both global and local scales (e.g. Birkmann 2006a; Cutter, Boruff, and
Shirley 2003; Fekete 2009; Gall 2007).
The Social Vulnerability Index (SoVI) by Cutter, Boruff, and Shirley (2003) is
considered to be the most popular composite indicator within disaster risk manage-
ment (cf. Gall 2007). The hierarchically structured SoVI evaluates the social vulnera-
bility of the US counties taking into account fragility and resilience factors. The
Disaster Risk Index developed by the United Nations Environment Programme
(UNEP) and the United Nations Development Programme (UNDP 2004) and the
Prevalent Vulnerability Index developed for the InterAmerican Development Bank
by Cardona (2004) are constructed to assess the social vulnerability on a national
level. Adger (2006) includes various different indicators in a model depicting the
national vulnerability against the impacts potentially arising in the course of climate
change. There is a lack of published work in the area of industrial production
vulnerability assessment. Only Villagrán de Léon (2006a) considers the susceptibility
of the industrial sector against physical damage (direct losses) within his Sector
Indicator Approach. However, this assessment is focused on the vulnerability of sin-
gle industrial objects and not on the industrial vulnerability of a region as a whole.
4. A composite indicator model for structural industrial vulnerability
In order to capture both fragility and resilience factors and to account for the pro-
cesses underlying industrial vulnerability, a two-stage approach to assess the struc-
tural industrial vulnerability against indirect disaster effects on a spatial level is
developed. We dene the structural industrial disaster vulnerability as the suscepti-
bility of industrial production to damage from disasters. From the structural per-
spective, industries are considered as components (or sub-systems) of the larger
socioeconomic environment, which is itself understood as a set of interrelated sub-
systems. Therefore, the vulnerability assessment focuses on the identication of the
most relevant impact factors in each of the sub-systems and investigates the type of
inuence they exert on the industrial production without explicitly modeling each
sub-system in detail. The structural susceptibility to damage of productions depends,
therefore, on the impact on production factors like production equipment, work-
force, required infrastructures, and the supply chain structure. This structural indus-
trial vulnerability analyzes the state of the industrial sectors in certain regions. It
does not account for (potential) risk management measures enhancing the resilience.
First, a composite indicator model to assess the structural vulnerability of 16
industrial sectors (V
s
) is developed. Second, a regionalization method is proposed to
determine the structural industrial vulnerability of different regions (V
r
) (e.g. adminis-
Journal of Risk Research 1081
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
trative districts). Figure 2 gives an overview on the overall structure of the developed
methodology.
To ensure the applicability of the composite indicator model, some quality
requirements regarding the individual indicators and the development procedure
must be met (Nardo et al. 2005b). According to Birkmann (2007), the individual
indicators included in a composite indicator model should, for example, be avail-
able, precise, reproducible, objective, clear, coherent, understandable, and interpret-
able. Furthermore, the development process and the methodologies used for this
purpose have an important inuence on the quality of the indicator results (Zhou,
Fan, and Zhou 2010).
Developing a composite indicator model encompasses the following steps:
development of a theoretical concept,
selection and structuring of indicators,
operationalization of indicators and data gathering,
normalization, weighting, and aggregation of indicator values, and
visualization of the results and sensitivity analysis.
These steps are similar to the main phases of the Multi-Attribute Value Theory
(MAVT) often used in the eld of decision analysis. Therefore, the methodological
approaches of MAVT can be transferred to the vulnerability assessment (Hiete and
Merz 2009).
The rst two steps (i.e. the development of the theoretical concept and the selec-
tion of appropriate indicators) constitute the basis for every vulnerability assess-
ment. Furthermore, the selection of the methods used for normalization, weighting,
and aggregation has a high inuence on the indicator results. However, this step is
prone to a high degree of subjectivity (Belton and Stewart 2002). This highlights
Two-stage methodology to assess the indirect
disaster vulnerability on a spatial level
Theoretical concept for
industrial vulnerability
Selection and operationalisation
of indicators
Regionalization of VS
combining structural vulnerability SVr
and industrial Exposure Er
Normalization, weighting
and aggregation
Sensitivity analysis
1
2
3
Composite
indicator model:
Sector specific
vulnerability
index Vs
Regional
vulnerability
index Vr
5
4
Figure 2. Overall structure of the methodology developed for the assessment of the
structural indirect disaster vulnerability on a spatial level.
1082 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
the need to clearly document the development of a composite indicator model to
depict the results in a transparent way and to integrate sensitivity and robustness
analysis (Merz, Hiete, and Schultmann 2010).
4.1. Theoretical concept for structural industrial vulnerability
To derive measurable vulnerability factors, the industrial production process was
analyzed on an abstract level, and general mechanisms shaping industrial disaster vul-
nerability were identied. Indirect disaster impacts like business interruptions and
domino effects occur if input factors are not available or if transportation processes
necessary for material supply and product distribution are interrupted (Hiete and Merz
2009). Important input factors are the production equipment, personnel manpower,
materials, and operating material such as energy and water supply (Ellinger and
Haupt 1996). Industrial sectors are supposed to be more susceptible to business inter-
ruptions the more dependent they are on these factors. Thus, four dimensions inu-
encing the indirect disaster vulnerability of an industrial sector were identied:
dependency on equipment, labor, infrastructure, and the functioning of the supply
chain.
These four vulnerability dimensions are split up further into seven dependency
dimensions (e.g. labor dependency, power dependency, water dependency, etc.)
(Table 1). For the operationalization of the vulnerability by means of measurable
indicators, factors increasing and decreasing these dependency dimensions were
selected (fragility and resilience indicators, respectively) (Table 1).
4.2. Selection and operationalization of indicators
The vulnerability indicators selected for the following case study are listed in
Table 1. Here, the column Operationalizationshows how the vulnerability
indicators miðsjÞwere determined and which statistical data were used. Note that
the operationalization may be case specic and depends on available data.
In the following, the rationale for some of the selected indicators is presented.
The vulnerability dimension infrastructure dependency is split up into the depen-
dency dimensions power dependency, water dependency, and transportation depen-
dency. Power dependency is further disaggregated into specic power consumption
and a power dependency factor from literature (ATC 2002) (fragility indicators) and
the degree of power self-supply (resilience indicator). The use of sector-specic
power consumption is based on the assumption that the higher the consumption is,
the more difcult it becomes to cover it alternatively (e.g. by emergency power
supply). While equipment and labor dependency are represented by just one depen-
dency dimension each, supply chain dependency is modeled by supply dependency
and demand dependency of a sector. Since the supply chain design is highly com-
pany dependent and data is condential (Peck 2005), the indicatorsquantication
on the sector level is challenging. In order to operationalize the upstream and down-
stream integration of an industrial sector within the supply chain, information from
national inputoutput tables was used. For example, the supply dependency is
depicted by the degree of in-house processing and the degree of supply-side
integration where the former is measured by the ratio of the in-house supply in a
sector and its supply from other sectors and the latter by a backward multiplier
based on the Leontief Inverse of the national inputoutput table (for the calculation
procedure see Schaffer 2002 or West and Brown 2003).
Journal of Risk Research 1083
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
Table 1. Hierarchical system of industrial vulnerability indicators and their operationalization.
Vulnerability
dimension Dependency
dimension Indicator Operationalization(determination
of indicator values miðsjÞ) Data source
Resilience
factor (RF)/
fragility
factor (FF)
Equipment
dependency Equipment
dependency I
1
: Equipment
intensity Value of xed assets of a sector/gross value added
of a sector (e)Destatis (2008) FF
Labor
dependency Labor
dependency I
2
: Labor intensity Number of employees of a sector/gross value added
of a sector (e)Destatis (2008) FF
I
3
: Degree of labor
specialization Specic labor costs (e) per employee of a sector FF
Infrastructure
dependency Power
dependency I
4
: Specic power
consumption Consumption of electricity of a sector (TJ)/gross value
added of a sector (e)Destatis (2008) FF
I
5
: Power importance ATC dependency factor power ATC (2002) FF
I
6
: Degree of power
self-supply Electricity generation (TJ)/electricity consumption (TJ) Destatis (2008) RF
Water
dependency I
7
: Specic water
consumption Volume of used freshwater of a sector (m
3
)/gross value
added of a sector (e)Destatis (2008) FF
I
8
: Water importance ATC dependency factor water ATC (2002) FF
I
9
: Degree of water
self-supply Degree of water self-supply (%) Destatis (2008) RF
Transport
dependency I
10
: Specic freight
transport road Transport volume of a sector on roads (Tkm)/gross value
added of a sector (e)Destatis (2008) FF
I
11
: Specic freight
transport railway Transport volume of a sector on railway (Tkm)/gross value
added of a sector (e)Destatis (2008),
KBA (2008) FF
Supply chain
dependency Supply
dependency I
12
: Material intensity Material costs of a sector (e)/gross value added of
a sector (e)Destatis (2008) FF
I
13
: Degree of
supply-side
integration
Backward multiplier (based on the Leontief Inverse) Destatis (2008) FF
I
14
: In-house
processing Value of in-house supply of a sector/supply from
other sectors to a sector Destatis (2008) RF
(Continued)
1084 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
Table 1. (Continued ).
Vulnerability
dimension Dependency
dimension Indicator Operationalization(determination
of indicator values miðsjÞ) Data source
Resilience
factor (RF)/
fragility
factor (FF)
Demand
dependency I
15
: Degree of
demand-side
integration
Forward multiplier (based on the Gosh Inverse) Destatis (2008) FF
I
16
: Customer
proximity Direct (consumption of goods produced by a sector by
households and the state/total amount of produced
goods of a sector
Destatis (2008) RF
Journal of Risk Research 1085
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
4.3. Normalization, weighting and aggregation
Individual indicator values miðsjÞof each sector s
j
and for each indicator iare
determined using the operationalization method dened in the previous step (see
Table 1, column operationalizationfor more details), mainly making use of ofcial
statistical data, in particular, Destatis data provided by the German Federal
Statistical Ofce. To make the different units and ranges of the individual indicators
comparable, the indicator values miðsjÞare normalized before combining them to
one composite indicator. Within the normalization step, similar to MAVT, so-called
vulnerability value functions v
i
are used to derive the indicator values miðsjÞof each
sector s
j
for an indicator ion a scale between 0 and 1:
vi
h1;0
miðsjÞ#viðmiðsjÞÞ
ð4Þ
Here, the value 1 represents the highest degree of vulnerability and 0 the lowest.
In the following case study, linear vulnerability value functions have been applied,
which means that a linear relation between an indicator value and its impact on
vulnerability is assumed. Linear vulnerability value functions for indicators with an
aggravating effect on vulnerability (fragility indicators) are dened by:
viðmiðsjÞÞ ¼ miðsjÞmmin
iðsjÞ
mmax
iðsjÞmmin
iðsjÞð5Þ
Resilience indicators are normalized by negative linear vulnerability value
functions:
viðmiðsjÞÞ ¼ mmax
iðsjÞmiðsjÞ
mmax
iðsjÞmmin
iðsjÞð6Þ
Here, mmin
iðsjÞis the lowest and mmax
iðsjÞthe highest value of an indicator i
measured across all jsectors.
The normalized indicators are aggregated into a composite indicator, represent-
ing the overall vulnerability of an industrial sector. Due to its comprehensibility, the
additive aggregation rule is used:
Vs¼vðsjÞ¼X
n
i¼1
wim
iviðmiðsjÞÞ ð7Þ
with: v(s
j
) = vulnerability of a sector s
j
;v
i
(m
i
(s
j
)) = normalized value of indicator
i.wim
i=importance weight of indicator i
The elicitation of weights for the individual indicators is especially important for
the quality of the results (Khazai et al. 2013). The weights wim
iexpress the relative
importance of the individual indicators. The weighting vector wim ¼ðwim
1; :::; wim
nÞ
contains the weights of all individual indicators i(i¼1; :::; n) of the composite
indicator model. For w
im
,it must be ensured that the constraints
X
n
i¼1
wim
i¼1;wim
i0 for all ið8Þ
1086 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
are satised. The assignment of weights requires a deep understanding of the
processes shaping industrial vulnerability and its underlying fragility and resilience
factors. Therefore, the retrieval of weighting factors is ideally done in workshops or
moderated group discussions with experts from industry and disaster science (Hiete
and Merz 2009). For the identication of importance weights wim
i, various methods
can be applied. Often, Analytical Hierarchy Process, SWING method, SMARTER
method, or DIRECT weighting approaches are used (Bertsch 2008). For the
following case study, the importance weights have been assigned such that the four
vulnerability dimensions are weighted equally regardless of the number of sub-indi-
cators.
Due to the additive aggregation, dependencies among the individual indicators
of the composite framework may lead to distorted results (Khazai et al. 2013). Par-
ticularly, the causal dependencies among the individual indicators can lead to an
overestimation or underestimation of single vulnerability dimensions or indicators
(Wu and Olson 2008). However, it is not possible to select completely independent
indicators, since they are all describing the same phenomenon industrial disaster
vulnerability (Saisana and Tarantola 2002). Within the existing composite indicator
models, the interindicator dependencies are mostly not considered.
In general, statistical as well as expert-based methodologies can be used to
consider the relationship among the individual indicators (for statistical methods see
e.g. Nardo et al. 2005b; Nicolett, Scarpeta, and Boylaud 2000). Within this paper,
the importance weights have been corrected for interdependencies using the
Decision Making Trial and Evaluation Laboratory (DEMATEL) method developed
by Fontela and Gabus (1976) and applied in many different situations (cf. Liou,
Yen, and Tzeng 2008; Wu, Chen, and Shieh 2010). The DEMATEL method enables
the quantication of the degree of direct and indirect dependencies between vari-
ables based on digraphs representing the direct dependencies assessed by experts.
Compared to statistical methods, the major benet of the DEMATEL method is the
fact that the method captures causal relationships and not only dependencies inher-
ent in the data (Hiete et al. 2011). To correct the importance weights wim
ifor struc-
tural relations between the individual indicators, the DEMATEL results were used
to calculate dependency weights wdp
iwith 0 wdp
i1. For indicators that consider-
ably inuence other indicators, wdp
iis considerably smaller than 1. To calculate the
overall weight of an indicator, the importance weights wim
iwere combined with the
dependency weights wdp
i:
wi¼wim
iwdp
ið9Þ
For detailed information on the calculation of wdp
i, see Hiete et al. (2011) or
Khazai et al. (2013).
4.4. Sensitivity analysis
Sensitivity analyses contribute to the understanding of the indicator-based
vulnerability assessment and, therefore, enhance the transparency of industrial risk
management (French 2003; Nardo et al. 2005a). The results of a sensitivity
analysis deliver valuable information concerning the design process and give
Journal of Risk Research 1087
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
implications for the adaptation of the model (e.g. with respect to indicator selec-
tion) and the selection of modeling parameters (e.g. weighting factors) (Saltelli
et al. 2009). Since the results of an indicator-based vulnerability assessment can
hardly be validated, sensitivity analyses can reveal important information about
the reliability of the results (Schmidtlein et al. 2008). Within composite indicators,
especially, model uncertainties are important. These uncertainties result among
other things from the selection of indicators, the operationalization of the individ-
ual indicators, the assignment of weights, or the selection of the aggregation rule
(e.g. additive vs. multiplicative) (Munda and Nardo 2009). Furthermore, data
uncertainties associated with the sectorsvalues with respect to the different indi-
cators can lead to indicator results that are not robust (Bertsch 2008). Though
sensitivity analyses are helpful to handle the uncertainties in a transparent manner
(Freudenberg 2003; Saisana, Saltelli, and Tarantola 2005), they were performed in
the design of a few composite indicator models only, e.g. Environmental Sustain-
ability Index (Saisana, Nardo, and Saltelli 2005), Environmental Protection Index
(Saisana and Saltelli 2008), Technology Achievement Index (Freudenberg 2003),
or SoVI (Schmidtlein et al. 2008).
4.5. Regionalization of sector specic vulnerability
In order to determine the industrial vulnerability V
r
of a region r, the structural
vulnerability SV
r
of that region representing fragility and resilience characteristics of
the regional industry was multiplied with the regional industrial exposure E
r
.
Vr¼SVrEr
with :
Vr= regional industrial vulnerability of region r
SV r= structural vulnerability of region r
Er= industrial exposure of region r
ð10Þ
For the calculation of the structural vulnerability SV
r
, the sector-specic
vulnerability index estimated with the above developed composite indicator model
was used:
SVr¼Pn
s¼1VsGVAsr
GVAr
with :
SV r= structural vulnerability of region r
Vs= sector specific vulnerability index sector s
GVAsr = gross value added of sector s in region r
GVAr= gross value added of all industrial sectors in region r.
ð11Þ
Values for GVA
sr
were taken from ofcial statistical data (Landesamt für
Statistik Baden-Wuerttemberg, pers. comm.). When the sector specic vulnerability
index V
s
of a sector s is multiplied with the relative share a, sector s has of the total
GVA in a region r, given by GVAsr
GVAr. Finally, the structural vulnerability SV
r
is
multiplied with the industrial exposure (represented by the industry density) E
r
of a
region r, which is calculated as:
1088 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
Er¼GVAr
Ar
mit
Er¼industrial exposure of region r
GVAr¼industrial gross value added of region r
Ar¼area region r
ð12Þ
5. Application of the model to an exemplar case study
5.1. Characteristics and aim of the case study
In order to demonstrate the suitability of the indicator-based approach for vulnera-
bility assessment, the developed approach was exemplarily applied to the federal
state of Baden-Wuerttemberg in southwestern Germany. The composite indicator
model was used to assess the indirect disaster vulnerability of 16 industrial sectors
in the state. These values were regionalized according to the procedure described
above. As a result, the industrial vulnerabilities of 44 administrative districts were
determined.
The case study area covers the entire federal state of Baden-Wuerttemberg with
a population of 11 million people, an area of 36 thousand square kilometers, and a
GDP 2009 of 344 billion e(Destatis 2010; Statistical Ofce of Baden-Wuerttem-
berg 2009). The most important industry sectors are automotive manufacturing and
mechanical and electrical engineering (e.g. Daimler AG, Robert Bosch GmbH)
(Statistical Ofce of Baden-Wuerttemberg 2009). Furthermore, companies in envi-
ronmental engineering, microsystem technology, healthcare and biotechnology, and
a dense network of small business enterprises have developed (Khazai et al. 2013).
The 16 industrial activities examined on the sector level are classied according
to the European NACE codes (Nomenclature Statistique des Activités Économiques
dans la Communauté Européenne) (Destatis 2003). In order to ensure a high repro-
ducibility, only ofcial statistical data provided by the Federal Ofce of Statistics in
Germany for 2007 were used (Destatis 2008) with the exception of the indicators
describing infrastructure dependencies, which were from ATC (2002). The exemplar
indicator values of the 16 evaluated industrial sectors of the case study are given in
Table 2.
5.2. Structural indirect disaster vulnerability of industrial sectors
Using the developed methodological approach and applying the weighting factors
wi,wim
i, and wdp
igiven in Table 2, the vulnerability indices V
s
are calculated for 16
industrial sectors s. The vulnerability indices V
s
provide information on the relative
vulnerability of the industrial sectors and differ considerably between them (see
Figure 2).
While the sectors rened petroleum products manufacturing,energy, gas, and
water supply and transportation equipment manufacturing show a relatively high
vulnerability against indirect disaster losses, other sectors such as the construction
and machinery and equipment manufacturing are not. When analyzing the contribu-
tion of the individual vulnerability dimensions to the overall vulnerability, equip-
ment dependency plays an important role for the vulnerability of the sectors energy,
gas, and water supply and rened petroleum products manufacturing. Therefore, in
order to minimize the economic disaster losses in these sectors, risk management
Journal of Risk Research 1089
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
Table 2. Normalized Indicator values of 16 industrial sectors calculated based on ofcial statistical data provided by the Federal Ofce of Statistics
in Germany for 2007.
Industrial sector I
1
()I
2
()I
3
()I
4
()I
5
()I
6
(%) I
7
()I
8
()I
9
(%) I
10
()I
11
()I
12
()I
13
(%) I
14
()I
15
()I
16
(%)
Food and tobacco 0.26 0.48 0.10 0.34 0.90 12.76 0.37 0.70 68.90 0.66 0.05 0.34 4.31 1.23 0.75 50.64
Textiles 0.31 0.47 0.17 0.27 1.00 3.37 0.24 0.70 75.40 0.25 0.01 0.23 3.17 0.97 0.72 30.37
Leather products 0.33 0.54 0.01 0.03 1.00 0.00 0.19 0.50 89.70 0.61 0.01 0.27 0.05 0.96 0.70 37.74
Wood products 0.19 0.26 0.21 0.46 1.00 30.62 0.13 0.50 69.90 0.44 0.21 0.22 6.52 1.23 1.16 13.93
Pulp, paper, and paper
products 0.25 0.31 0.36 0.57 0.55 25.14 0.45 0.45 96.10 0.46 0.22 0.18 6.59 1.06 1.05 21.23
Rened petroleum
products 0.65 0.25 0.95 1.00 1.00 76.21 0.69 0.50 76.90 0.97 1.00 1.00 7.86 0.76 1.09 30.54
Chemical products 0.38 0.34 0.70 0.76 0.90 26.25 0.70 0.80 78.60 0.43 0.25 0.37 33.58 1.24 0.99 7.71
Rubber and plastic 0.15 0.43 0.29 0.44 1.00 0.00 0.22 0.50 74.80 0.30 0.18 0.25 2.06 1.01 1.02 2.87
Nonmetallic mineral
products 0.28 0.32 0.36 0.55 1.00 1.55 0.29 0.50 87.80 0.70 0.21 0.15 0.78 1.34 1.11 6.25
Metal products 0.16 0.33 0.40 0.59 0.95 12.19 0.40 0.85 89.80 0.32 0.27 0.28 27.00 1.12 1.16 5.45
Machinery 0.09 0.39 0.50 0.06 1.00 0.00 0.03 0.60 80.40 0.08 0.02 0.25 0.02 1.05 0.74 20.99
Electrical and optical
equipment 0.17 0.35 0.52 0.08 1.00 0.00 0.13 0.90 74.10 0.09 0.01 0.31 0.83 0.98 0.83 16.61
Transport equipment 0.35 0.77 0.73 0.30 1.00 0.00 0.16 0.60 66.10 0.29 0.15 0.53 9.09 1.20 0.80 18.74
Manufacturing n.e.c. 0.20 0.38 0.26 0.11 0.95 0.00 0.01 1.00 72.70 0.22 0.03 0.24 0.11 1.02 0.74 44.85
Electricity and water 1.00 0.23 1.00 0.48 0.80 47.00 1.00 0.40 98.00 0.01 0.05 0.61 4.76 1.04 1.12 65.36
Construction 0.02 0.15 0.20 0.27 0.40 0.00 0.12 0.50 10.00 0.64 0.36 0.01 0.00 1.11 0.73 79.36
1090 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
should focus on the restoration of the production equipment. Furthermore, the sector
rened petroleum products manufacturing as well as the sectors manufacturing of
basic metals and metal products and manufacturing of chemical products are partic-
ularly dependent on the functioning of different infrastructure systems. The sector
showing the highest dependency on work force is transportation equipment manu-
facturing. The relative contribution of the vulnerability dimension supply chain
dependency is almost similar for all analyzed sectors. While the sector manufactur-
ing of basic metal and metal products is marked by a surpassing demand depen-
dency, the sector manufacturing of food and tobacco shows a relatively high supply
dependency.
The presented approach allows considering the contribution of the individual
vulnerability dimensions or single indicators prior to aggregation. Figure 3 shows
the infrastructure dependency of the 16 sectors and the relative contribution of the
dependency on electricity, water, and transportation. While the dependency on elec-
tricity is relatively high in all industrial sectors, the dependency on water supply
and on transportation systems varies considerably. For example, the sector energy,
gas, and water supply shows a high dependency on the availability of water, and
the sector manufacturing of rened petroleum products strongly depends on the
proper functioning of transportation systems. The knowledge about the dependency
of the industrial sectors on infrastructure systems, for example, constitutes valuable
0,0 0,1 0,2 0,3 0,4 0,5 0,6
construction
manufacturing n.e.c.
machinery and equipment
electrical and optical equipment
wood and wood products
leather and leather products
textiles and wearing apparels
food and tobacco
pulp, paper and paper products
rubber and plastic products
basic metals and metal products
non-metallic mineral products
chemical products
transport equipment
electricity, gas and water supply
refined petroleum products
capital dependency labor Dependency electricity dependency water dependency
transportation dependency supply dependency demand dependency
0.529
0.528
0.424
0.408
0.357
0.331
0.310
0.310
0.298
0.295
0.190
0.264
0.291
0.289
0.290
0.255
Figure 3. Structural sector specic vulnerability index V
s
of 16 industrial sectors in the
German federal state of Baden-Wuerttemberg.
Journal of Risk Research 1091
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
information for disaster risk management since this shows which types of infrastruc-
ture system should be protected most in which industrial sectors.
In order to assess the robustness of the results to the variation of modeling
parameters and model congurations, a sensitivity analysis has been performed.
Within ve indicator scenarios, the shape of the vulnerability value functions, the
indicator weights, and the aggregation method were varied. Based on theoretical
considerations, one scenario using exponential vulnerability value functions, one
scenario applying a multiplicative aggregation method, and three scenarios with
varying weighting factors were assessed.
To measure the changes caused by the different scenarios, Spearman correlation
coefcient rsand average shift in rank positions
Rrelative to the base scenario were
calculated (Table 3). These characteristics point out that the results are relatively
robust to model and parameter variations and that the deviation of the assessed
scenarios from the reference scenario with an average shift in rank positions
Rof
0.9 1.4 is moderate (Table 4). This is also shown by the relatively high Spear-
man correlation coefcients rsbetween 0.89 and 0.92. However, especially, model
uncertainties (e.g. the aggregation method chosen) have a certain inuence on the
results, which should be made explicit.
5.3. Regional structural industrial vulnerability in the German federal state of
Baden-Wuerttemberg
The structural vulnerability SV
r
(see Equation (11)) and the industrial exposure E
r
(see Equation (12)) of the r assessed regions (administrative districts) have been
also calculated based on ofcial statistical data published by the Statistical Ofce of
Baden-Wuerttemberg (2006a, 2006b). For the spatial visualization of structural vul-
nerability SV
r
(see Equation (11)) and the industrial exposure E
r
(see Equation (12))
of the administrative districts r, SV
r
and E
r
were normalized using a MinMax
Normalization and the geographical information systems (GIS) software ArcGIS®
(Figure 4).
Table 3. Indicator weights wi,wim
iand wdp
iused in the case study.
Indicator wiwim
iwdp
i
Equipment dependency Equipment intensity 0.218 0.25 0.87
Labor dependency Labor intensity 0.118 0.125 0.94
Degree of labor specialization 0.105 0.125 0.84
Infrastructure dependency Specic power consumption 0.019 0.028 0.69
Power importance 0.021 0.028 0.76
Degree of power self-supply 0.021 0.028 0.75
Specic water consumption 0.018 0.028 0.63
Water importance 0.021 0.028 0.76
Degree of water self-supply 0.021 0.028 0.77
Specic freight transport road 0.033 0.042 0.80
Specic freight transport railway 0.033 0.042 0.80
Supply chain dependency Material intensity 0.031 0.042 0.75
Degree of supply-side integration 0.030 0.042 0.76
In-house processing 0.032 0.042 0.72
Degree of demand-side integration 0.051 0.063 0.82
Customer proximity 0.047 0.063 0.76
1092 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
The structural vulnerability SV
r
is inuenced by both the types of industrial
sectors located in an administrative district and the sector specic vulnerability of
these sectors. Administrative districts, characterized by a particularly vulnerable
industrial structure, are, for example, the state capital Stuttgart and the administra-
tive districts along the Rhine River forming the western border of Baden-Wuerttem-
berg (Figure 5, left side). The industrial exposure E
r
, depicted by the industrial
density (GVA [e]/area [km
2
]), is also relatively high in the greater metropolitan
area Stuttgart (Figure 5, right side).
The vulnerability value V
r
shows how vulnerable the industry located in the
administrative districts is against business interruptions caused by natural disasters
(Figure 6). While in the urban areas around Stuttgart and Mannheim, both the struc-
tural vulnerability and the industrial exposition contribute to high vulnerability
values, in the administrative districts Böblingen and Heilbronn, the high industrial
Table 4. Statistical characteristics summarizing the results of the sensitivity analyses.
Sensitivity analysis Spearman correlation
coefcient rs
Average shift in
rank positions
R
Number of
scenarios
Variation of weights 0.92 0.9 3
Variation of vulnerability functions 0.96 1.2 1
Variation of aggregation method 0.89 1.4 1
0,150,100,050,00
machinery and equipment
electrical and optical equipment
manufacturing n.e.c.
textiles and wearing apparels
transport equipment
leather and leather products
wood and wood products
rubber and plastic products
pulp, paper and paper products
electricity, gas and water supply
food and tobacco
construction
non-metallic mineral products
chemical products
basic metals and metal products
refined petroleum products
power dependency water dependency transport dependency
0.146
0.108
0.105
0.096
0.090
0.089
0.087
0.081
0.080
0.079
0.049
0.055
0.059
0.072
0.070
0.072
Figure 4. Infrastructure dependency of 16 industrial sectors in the German federal state of
Baden-Wuerttemberg.
Journal of Risk Research 1093
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
density leads to an enhanced vulnerability. Despite their relatively low industrial
exposure (density), high vulnerability values are assigned to the administrative
districts Baden-Baden and Karlsruhe. This is caused by the extremely susceptible
industrial sectors (e.g. manufacturing of rened petroleum products,manufacturing
of chemical products and energy, gas, and water supply) located in these adminis-
trative districts.
(SVr)norm
0.0 – 0.1
0.1 – 0.2
0.2 – 0.4
0.8 – 1.0
0.4 – 0.8
0.0 – 0.1
0.1 – 0.2
0.2 – 0.4
0.6 – 1.0
0.4 – 0.8
(Er)norm
Stuttgart
Stuttgart
Mannheim
Boeblingen
Karlsruhe
Figure 5. Structural industrial vulnerability SV
r
and industrial exposure E
r
of the
administrative districts r in the German federal state of Baden-Wuerttemberg.
0.0 – 0.1
0.1 – 0.2
0.2 – 0.4
0.8 – 1.0
0.4 – 0.8
(Vr)norm
Stuttgart
Mannheim
Boeblingen
Karlsruhe
Baden-Baden
Heilbronn
Figure 6. Regional industrial vulnerability V
r
of the administrative districts r in the German
federal state of Baden-Wuerttemberg.
1094 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
6. Conclusion
In the event of natural disasters, industrial production sites are affected by direct
(physical) damage to production equipment, buildings, and materials and by indirect
losses due to production interruptions and domino effects in interlaced supply
chains. In order to reduce the often severe consequences of industrial disaster
damage, risk management measures must be implemented on the company as well
as on a regional level (e.g. administrative districts).
To provide a sound basis for decisions on risk prevention and mitigation
measures, the industrial risk of different regions must be assessed. Since the inten-
sity of a hazard can hardly be altered by risk management measures, the evaluation
of the vulnerability of different regions serves as an important input for risk
management decisions. Due to the complexity of industrial production systems and
the concept of vulnerability itself, the structural vulnerability of industrial produc-
tion systems cannot be measured directly. Composite indicators, which are widely
applied in social vulnerability assessment, are considered as a useful tool for indus-
trial vulnerability assessment by analyzing the impact factors driving vulnerability
and resilience and decomposing the vulnerability accordingly.
Within this paper, an indicator-based approach to assess the structural indirect
industrial vulnerability of administrative districts has been presented. In order to
consider industry specic vulnerability factors on a spatial assessment level, a two-
stage approach using a sector-specic indicator model in combination with a newly
developed regionalization method has been developed.
For the development of the sector-specic indicator model, methods from the
eld of multicriteria decision analysis (MAVT) have been used. In order to correct
the results for interdependencies among the single indicators, the DEMATEL
method has been applied.
The applicability of the developed approach has been demonstrated by means of
a case study that assesses the industrial vulnerability of 44 administrative districts
in the federal state of Baden-Wuerttemberg in southwestern Germany. Sensitivity
analyses performed within the case study showed that the results of the methodo-
logical approach are overall robust. However, it became evident that, especially,
model uncertainties should not be neglected: it is important to show the inuence
of parameter and model variations when presenting the results in a transparent
manner.
The results on the regional industrial vulnerability help to identify hotspots of
vulnerability and to point out where prevention or mitigation measures could be
implemented most effectively. Furthermore, the assessment based on sector-specic
vulnerability values helps to identify central vulnerability factors and facilitates the
selection of adequate prevention and mitigation procedures. Therefore, the results of
the developed approach supply decision-makers with a multifaceted picture of the
industrial disaster loss potential and deliver valuable information to the overall
reduction of risk.
The presented approach has a number of limitations and leaves room for further
research. As every model, the sector specic indicator model is a simplication of
the real world, and only the vulnerability factors considered most important were
operationalized as far as (mainly ofcial) statistical data allowed this. However, due
to the open structure of a composite indicator model, further indicators (e.g. depict-
ing the dependency on infrastructures like gas supply or air freight) can be easily
Journal of Risk Research 1095
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
integrated if deemed necessary and feasible. The structural vulnerability assessment
focussing on the systemsstate can be understood as a stepping stone towards risk
mitigation: by identifying the most vulnerable sectors, risk management measures
enhancing the resilience can be dened. Future work consists in extending the
indicator framework to take the risk management into account. In this manner, the
impact of different risk management strategies can be compared. Furthermore,
within the developed indicator model, only generic indicators have been used such
that the model does not reect the hazard dependency of vulnerability. To generate
hazard-specic vulnerability results, the selection of indicators and their weightings
may be adapted according to the hazard prole.
Acknowledgments
We would like to thank the Center for Disaster Management and Risk Reduction
Technology (http://www.cedim.de), an interdisciplinary research center in the eld of disaster
management founded by GFZ German Research Centre for Geosciences and Karlsruhe
Institute of Technology (KIT), for nancially supporting Mirjam Merz.
References
Adger, W.N. 2006. Vulnerability. Global Environmental Change 16, no. 3: 5763.
Alexander, D. 2000. Confronting catastrophe. Hertfordshire: Terra.
ATC. 2002. Seismic vulnerability and impact of disruption of lifelines in the conterminous
United States. Washington, DC: Earthquake Hazard Reduction Series, 58, Applied
Technology Council.
Baas, S., S. Ramasamy, J.D. DePryck, and F. Battista. 2008. Disaster risk management
systems analysis, environment, climate change, and bioenergy divisions. Rome: Food
and Agriculture Organisation of the United Nations (FAO).
Bertsch, V. 2008. Uncertainty handling in multi-attribute decision support for industrial risk
management. Karlsruhe: Universitätsverlag Karlsruhe. http://digbib.ubka.uni-karlsruhe.de/
volltexte/1000007378.
Belton, V., and T. Stewart. 2002. Multiple criteria decision analysis an integrated
approach. Boston: Kluwer Academic Press.
Birkmann, J. 2006a. Indicators and criteria for measuring vulnerability: Theoretical bases
and requirements. In Measuring vulnerability to hazards of natural origin towards
disaster resilient society, ed. J. Birkmann, 5577. Tokyo: UNU Press.
Birkmann, J. 2006b. Measuring vulnerability to promote disaster-resilience societies: Con-
ceptual frameworks and denitions. In Measuring vulnerability to hazards of natural ori-
gin towards disaster resilient society, ed. J. Birkmann, 954. Tokyo: UNU Press.
Birkmann, J. 2007. Risk and vulnerability indicators at different scales: Applicability,
usefulness and policy implications. Environmental Hazards 7: 2031.
Birkmann, J., and B. Wisner. 2006. Measuring the un-measurable: The challenges of vulner-
ability. SOURCE, Publication Series of UNU-EHS, 5, Bonn.
Cardona, O.D. 2004. The need for rethinking the concepts of vulnerability and risk from a
holistic perspective: A necessary review and criticism for effective risk management. In
Mapping vulnerability, disasters, development, and people, ed. G. Bankoff, G. Frerks,
and D. Hilhorst, 3751. London: Earthscan.
Cruz, A.M., and N. Okada. 2008. Consideration of natural hazards in the design and risk
management of industrial facilities. Natural Hazards 44, no. 2: 21327.
Cutter, S.L., B.J. Boruff, and W.L. Shirley. 2003. Social vulnerability to environmental
hazards. Social Science Quarterly 84, no. 2: 24261.
Cutter, S.L., Ch.G. Burton, and C. Emrich. 2010. Disaster resilience indicators for bench-
marking baseline conditions. Journal of Homeland Security and Emergency Management
7, no. 1: 122.
Dastous, P.-A., J. Nikiema, and S. Megaloconomos. 2008. Risk management: All stakehold-
ers must do their part. Journal of Loss Prevention in the Process Industries 21: 36773.
1096 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
Destatis, Federal Statistical Ofce Germany. 2003. Klassikation der Wirtschaftszweige mit
Erläuterungen [Classication of sectors of the economy with explanations]. Wiesbaden:
Federal Statistical Ofce.
Destatis, Federal Statistical Ofce Germany. 2008. Kostenstruktur im Produzierenden Gew-
erbe 2006 [Cost structure in industry 2006], In Statistisches Jahrbuch für die Bundesre-
publik Deutschland 2008, 366399. Wiesbaden: Statistisches Bundesamt.
Dilley, M., R.S. Chen, U. Deichmann, A.L. Lerner-Lam, and M. Arnold. 2005. Natural
disaster hotspots: A global risk analysis. Washington, DC: Weltbank.
Ellinger, T., and R. Haupt. 1996. Produktions- und Kostentheorie [Production and cost theory].
Stuttgart: Schäffer-Poeschel Verlag.
European Commission (EC). 2010. Risk assessment and mapping guidelines for disaster
management. Brussels: Commission Staff Working Paper.
Fekete, A. 2009. Validation of a social vulnerability index in context to river-oods in Ger-
many. Natural Hazards and Earth System Sciences 9: 393403.
Fontela, E., and A. Gabus. 1976. The DEMATEL observer, DEMATEL report 1976. Genf:
Battelle Geneva Research Center.
French, S. 2003. Modelling, making inferences and making decisions: The roles of sensitiv-
ity analysis. TOP 11, no. 2: 22951.
Freudenberg, M. 2003. Composite indicators of country performance a critical assessment,
OECD Science, Technology and Industry Working Papers 2003/16, DSTI/DOC(2003)
16-JT00153477. Paris: OECD.
Gall, M. 2007. Indices of social vulnerability to natural hazards: a comparative evaluation.
PhD thesis, University of South Carolina, Columbia City.
Gheorghe, A.V., R. Mock, and W. Kröger. 2000. Risk assessment of regional systems.
Reliability Engineering & System Safety 70: 14156.
Green, C.H., and A. van der Veen. 2007. Indirect economic damage: Concepts and guide-
line. In Evaluating ood damages: Guidance and recommendations on principles and
methods, eds. F. Messner, E. Penning-Rowsell, C.H. Green, V. Meyer, S. Tunstall and A.
van der Veen, 95105. FLOODsite Report, T09-06-01, Wallington.
Hahn, H. 2003. Indicators and other instruments for local risk management for communities and
local governments. Eschborn: Gesellschaft für Technische Zusammenarbeit GmbH (GTZ).
Hiete, M., and M. Merz. 2009. An indicator framework to assess the vulnerability of indus-
trial sectors against indirect disaster losses. Proceedings of the International Conference
on Information Systems for Crisis Response and Management (ISCRAM), 2009, Gothen-
burg, Paper Nr. 131.
Hiete, M., M. Merz, T. Comes, and F. Schultmann. 2011. Trapezoidal fuzzy DEMATEL
method to analyze and correct for relations between variables in a composite indicator
for disaster resilience, OR Spectrum. http://www.springerlink.com/content/qg561n2l2632
7573/fulltext.pdf.
Hossini, V. 2008. The role of vulnerability in risk management Summary of the third PhD
block course. Working Paper 8/2008, UNU Institute for Environment and Human
Security (UNU-EHS), Bonn.
KBA. 2008. Verkehrsaufkommen im Jahr 2008 nach Güterabteilungen [Trafc volumes by type
of goods in 2008]. In Verkehr deutscher Lastkraftfahrzeuge Verkehrsaufkommen, Eigens-
chaften der Ladung, Kraftfahrt Bundesamt (KBA) [Trafc of German freight vehicles - vol-
ume of trafc, freight properties], Flensburg: German Federal Motor Transport Authority.
Khazai, B., M. Merz, C. Schulz, and D. Borst. 2013. An integrated indicator framework for
spatial assessment of industrial and social vulnerability to indirect disaster losses. Natu-
ral Hazards,123. Published online 20 January 2013, 10.1007/s11069-013-0551-z.
Kleindorfer, P.R., and H.G. Saad. 2005. Managing disruption risks in supply chains.
Production and Operations Management 14, no. 1: 5368.
Leitch, M. 2010. ISO 31000: 2009 - the new international standard on risk management.
Risk Analysis 30: 88792.
Liou, J.J.H., L. Yen, and G.H. Tzeng. 2008. Building an effective safety management system
for airlines. Journal of Air Transport Management 14: 206.
Merz, M., M. Hiete, and F. Schultmann. 2010. An indicator framework for the assessment of the
indirect disaster vulnerability of industrial production systems. Proceedings of the Interna-
tional Conference on Disaster Risk Reduction (IDRC), May 2010, Davos, Paper Nr. 131.
Journal of Risk Research 1097
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
Messner, F., and C. Green. 2007. Fundamental issues in the economic evaluation of ood
damage. In Evaluating ood damages: Guidance and recommendations on principles
and methods, ed. F. Messner, E. Penning-Rowsell, C.H. Green, V. Meyer, S. Tunstall
and A. van der Veen, 95105. FLOODsite Report, T09-06-01, Wallington, 95105.
Messner, F., and V. Meyer. 2005. Flood damage, vulnerability and risk perception chal-
lenges for ood damage Research, UFZ Discussion Papers 13/2005, Umweltforschungs-
zentrum Leipzig, Germany.
Munda, G., and M. Nardo. 2009. Non-compensatory/nonlinear composite indicators for
ranking countries: a defensible setting. Applied Economics 41, no. 12: 151223.
Nardo, M., M. Saisana, A. Saltelli, and S. Tarantola. 2005a. Tools for composite indicator
building. Report, EUR 221682 EN. Ispra: Joint Research Center.
Nardo, M., M. Saisana, A. Saltelli, M. Tarantola, and E. Giovanni. 2005b. Handbook on
constructing composite indicators: Methodology and user guide, OECD Statistics Work-
ing Paper, TD/DOC(2005)3-JT00188147. Paris: OECD.
Nicolett, G., S. Scarpeta, and O. Boylaud. 2000. Summary indicators of market regulation
with an extension to employment protection legislation, OECD Economic Department
Working Papers, 226 ECO/WKP (99)18, Paris.
Okuyama, Y. 2007. Economic modeling for disaster impact analysis: Past. Present and
Future, Economic systems research 19, no. 2: 11524.
Peck, H. 2005. Drivers of supply chain vulnerability: An integrated framework. International
Journal of Physical Distribution & Logistics Management 35, no. 4: 21032.
Purdy, G. 2010. ISO 31000:2009 setting a new standard for risk management. Risk Analy-
sis 30, no. 6: 8816.
Saisana, M., M. Nardo, and A. Saltelli. 2005. 2005 ESI sensitivity analysis, 2005 Environ-
mental Sustainability Index Report, Yale Center for Environmental Law and Policy, New
Haven, CT. http://www.yale.edu/esi.
Saisana, M., and A. Saltelli. 2008. Sensitivity analysis of the 2008 Environmental Perfor-
mance Index, JRC Scientic and Technical Reports, EUR 23485 EN 2008, Joint
Research Center (European Commission), Brüssel.
Saisana, M., A. Saltelli, and S. Tarantola. 2005. Uncertainty and sensitivity analysis tech-
niques as tools for the quality assessment of composite indicators. Journal of the Royal
Statistical Society 168, no. 2: 30723.
Saisana, M., and S. Tarantola. 2002. State-of-the-art report on current methodologies and
practices for composite indicator development, JRC Report, EUR 20408 EN. Ispra: Joint
Research Centre.
Saltelli, A., M. Ratto, T. Andres, F. Campolongo, J. Caribani, D. Gatelli, M. Saisana, and S.
Tarantola. 2009. Global sensitivity analysis the primer. Chichester: Wiley & Sons.
Sauerwein, E., and M. Thurner. 1998. Der Risikomanagementprozess im Überblick [An
overview of the risk management process]. In Betriebliches Risikomanagement [Opera-
tional risk management], ed. H. Hinterhuber, E. Sauerwein, and Ch. Fohler-Norek, 19
39. Wien: Verlag Österreich.
Schaffer, A. 2002. Ecological input-output analysis: ECOLIO’–a model for conventional
and ecological key sector analyses in Germany. Karlsruher Beiträge zur wirtschaftspoli-
tischen Forschung, Vol.13. Baden-Baden: Nomos-Verlags-Gesellschaft.
Schmidtlein, M.C., R.C. Deutsch, W.W. Piegorsch, and S.L. Cutter. 2008. Sensitivity analy-
sis of the social vulnerability index. Risk Analysis 4: 1099114.
Statistical Ofce of Baden-Wuerttemberg. 2006a. Bruttowertschöpfung in jeweiligen Preisen
für Deutschland und Baden-Württemberg [Gross value added in corresponding prices for
Germany and Baden-Wuerttemberg] (available upon request in electronic form). Lande-
samt für Statistik Baden-Württemberg.
Statistical Ofce of Baden-Wuerttemberg. 2006b. Sozialversicherungspichtig Beschäftigte
am Arbeitsort nach Kreisen/kreisfreien Städten [Employees subject to social insurance
contribution at place of work differentiated according to districts and urban municipali-
ties] (available upon request in electronic form). Landesamt für Statistik Baden-Württem-
berg.
Statistical Ofce of Baden-Wuerttemberg. 2009. Baden-Württemberg ein Standort im
Vergleich [Baden-Wuerttemberg a location in comparison]. http://www.statistik.baden-
wuerttemberg.de/Veroeffentl/803609002.pdf (accessed December 27, 2010).
1098 M. Merz et al.
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
Steinberg, L., H. Sengul, and A.M. Cruz. 2008. NATECH risk and management: An
assessment of the state of the art. Natural Hazards 46, no. 2: 14352.
Thywissen, K. 2006. Components of risk a comparative glossary. SOURCE, 02/2006,
UNU Institute for Environment and Human Security (UNU-EHS), Bonn.
Turner, B.L., R.E. Kasperson, P.A. Matsone, J.J. McCarthy, R.W. Corellg, L. Christensene,
N. Eckley, et al. 2003. A framework for vulnerability analysis in sustainability science.
Proceedings of the National Academy of Science (PNAS) 100, no. 14: 80749.
Twigg, B.L. 2001. Sustainable livelihoods and vulnerability to disasters. Disaster Manage-
ment Working Paper, 2/2001, Beneld Greig Hazard Research Centre, London.
UNDP. 2004. A global report, reducing disaster risk: A challenge for development. Genf:
United Nations Development Program.
UN/ISDR. 2005. Hyogo framework for action 20052015: Building the resilience of nations
and communities to disasters. Genf: United Nations International Strategy for Disaster
Reduction.
UN/ISDR. 2004. Living with risk. A global review of disaster reduction initiatives. Genf:
United Nations International Strategy for Disaster Reduction.
Villagrán de Léon, J.C. 2006a. Vulnerability assessment. The sectoral approach. In Measur-
ing vulnerability to hazards of natural origin towards disaster resilient society, ed. J.
Birkmann, 30015. Tokyo: UNU Press.
Villagrán de Léon, J.C. 2006b. Vulnerability a conceptional and methodological review,
SOURCE, 04/2006, UNU Institute for Environment and Human Security (UNU-EHS),
Bonn.
West, G.R., and R.P.C. Brown. 2003. Structural Change, Intersectoral Linkages And Hollow-
ing-Out in the Taiwanese Economy, 19761994. Discussion Paper No 327, School of
Economics, University of Queensland.
Wu, D., and D.L. Olson. 2008. A comparison of stochastic dominance and stochastic DEA
for vendor evaluation. International Journal of Production Research 46, no. 8: 231327.
Wu, H.H., H.K. Chen, and J.I. Shieh. 2010. Evaluating performance criteria of employment
service outreach program personnel by DEMATEL method. Expert Systems with Appli-
cations 37, no. 7: 521923.
Zhou, P., L.-W. Fan, and D.-Q. Zhou. 2010. Data aggregation in constructing composite
indicators: A perspective of information loss. Expert Systems with Applications 37, no.
1: 3605.
Journal of Risk Research 1099
Downloaded by [Universitetsbiblioteket i Agder] at 01:48 18 April 2014
... Indicators are typically designed to measure and compare the level of performance with respect to benchmark cases. They can be designed to measure and compare the resilience of different systems, defined by geographical areas and communities [1], industrial sectors [54] or critical infrastructures [55,56]. Social resilience is inherently embedded in a system, but there are many ways to understand J o u r n a l P r e -p r o o f that embeddedness. ...
... Despite or maybe because of the many definitions of resilience, the concept itself is quite abstract and complex. Hence, there are many calls from practice to operationalize resilience, to turn it into a concrete and measurable concept that lends itself to decision-and policy-making [54,84]. Any attempt to measure and operationalize resilience, however, also has a definitory character -determining for the user what resilience means, and how it can be applied. ...
Article
Full-text available
More than any other facet of resilience, social resilience raises the inherent tension within the concept between identity or persistence, and transformation. Is a community the people who make it up, or the geography or physical infrastructure they share? What about the resilience of communities that transform, as a result of a sudden disaster or over time? In this paper, we explore the impact of this tension on how social resilience indicators can be developed and used. Beginning with a close look at the ways in which our concepts of resilience and our use of indicators interact, several points are raised. First, that how we identify a community and frame its resilience conveys particular conceptualisations of resilience, which in turn have normative implications for the communities themselves. In part, this is because of the difficulty in capturing important adaptations and transformative actions within and by those communities. Further, measuring and comparing the resilience of communities, and aspects of quantification that go along with selecting, aggregating and comparing indicator values, ensure that the decisions made about how indicators ought to be used carry normative weight. Through this exploration, we identify several normative implications of choices in indicator design and application. We conclude with recommendations for moving forward with greater transparency and responsibility toward those communities whose social resilience we hope to measure in order to improve.
... The average absolute shift in rank position ( Rs ) is an adapted tool for testing the robustness, stability, and reliability of the findings (Hudrliková 2013;Nazeer and Bork 2021). Rs values of 1.10, 1.13, and 1.19 for the scenarios RA.LA, ZC.LA, and MM.LA respectively are moderate (Merz et al. 2013) implying a moderate deviation from the reference scenario (median ranking). The results are therefore relatively robust to the variation in the initial indicator normalization and aggregation techniques. ...
Article
Full-text available
Climate change has severe impacts on the livelihoods of West-African communities with the floods of the late 2000s and early 2010s serving as factual evidence. Focusing on the assessment of observed and future vulnerability to extreme rainfall in the tropical Ouémé River Basin, this study aims to provide scientific evidence to inform national adaptation plans. Observed climate variables, historical and future outputs from regional climate models, topographic, land cover, and socioeconomic data were used in the vulnerability assessment. This assessment was based on four indicator normalization methods (min–max, z-scores, distance to target, and ranking), two aggregation techniques (linear and geometric), four classification methods (quantile, standard deviation, equal intervals, natural breaks), and three robustness evaluation approaches (spearman correlation, Akaike Information Criterion (AIC), and average shift in ranks). Based on the AIC, it was found that “equal intervals” is the overall best classification method and the min–max normalization with linear aggregation (MM.LA) outperformed other methods. The median scenario indicates that the population of the Ouémé Basin is vulnerable to the adverse impacts of climate change for the historical (1970–2015) and future periods (2020–2050) as a result of low adaptive capacity. By 2050, the southern part of the Ouémé Basin will be highly vulnerable to pluvial flooding under RCP 4.5. Vulnerable municipalities will continue to suffer from flooding if adequate adaptation measures including water control infrastructure (development and expansion of rainwater and wastewater drainage systems) and appropriate early warning systems to strengthen community members’ resilience are not taken.
... Задача рассматривается как построение многофакторной модели некоторого сложного явления. Существует множество методов оценки и картирования опасности/риска природных и техногенных бедствий (Kohler, 2004;EU Commission Staff Paper, 2010;Fekete, 2012;Merz et al., 2013;Orencio and Fujii, 2014;Papathoma-Köhle et al., 2016, Council Directive 67/548/EEC, 2016. Причинами такого разнообразия являются различные подходы к решению этих задач, а также объективное различие сущности и физических особенностей опасных объектов, процессов и явлений. ...
... Natural disasters can hit Industry potentially, as reviewed previously, there are different types of natural disasters or other man hazard risk that can jeopardize the production system of the industry. Merza (2013), divides the impacts affecting the industry in direct and indirect ways. The first type of impact, the direct, Merza states in Cruz (2015) and will have to do with damages on buildings, materials, and production equipment that will obstruct labors to continue working on the production. ...
Chapter
Recently, the concept of humanitarian relief was introduced, and there remains work to do to help people in any situation of disaster. The present research is focused on determining an inventory level for relief kits that can benefit 24 municipalities that belong to the State of Puebla, Mexico. Historically, these municipalities have been vulnerable to hydrometeorological phenomena. From 2001 to 2017, Puebla has had 1,632 emergency declarations, of which 59.7% were classified as a hydrometeorological issue. According to this historical, the disasters increased during August, September, and October, so it is proposed with this research to have an adequate inventory level of kits before the disasters happen in accordance to the months above. In the official records, there are not registered the number of affected people of these municipalities, so to determine demand, the frequency was found using the historical data regarding affected people by hydrometeorological phenomena at a national level. The lot size was calculated using the Newsboy Inventory Model, and the demand was separated in different age range and gender to make kits following the necessities.
... In the context of disaster risk reduction, most predictive approaches focus on a combination of hazard (event), exposure (elements at risk) and vulnerability or resilience (Djalante et al., 2011;Merz et al., 2013). The hazard dimension is typically modelled through dedicated meteorological, geological, seismic etc. models, which provide an assessment about the magnitude of specific events that have a given likelihood of occurrence (Karimi and Höllermeier, 2007). ...
Chapter
Full-text available
Literature about humanitarian logistics (HL) has developed a lot of innovative decision support systems during the last decades to support decisions such as location, routing, supply, or inventory management. Most of those contributions are based on quantitative models but, generally, are not used by practitioners who are not confident with. This can be explained by the fact that scenarios and datasets used to design and validate those HL models are often too simple compared to the real situations. In this chapter, a scenario-based approach based on a five-step methodology has been developed to bridge this gap by designing a set of valid scenarios able to assess disaster needs in regions subject to recurrent disasters. The contribution, usable by both scholars and practitioners, demonstrates that defining such valid scenario sets is possible for recurrent disasters. Finally, the proposal is validated on a concrete application case based on Peruvian recurrent flood and earthquake disasters.
Article
Chemical clusters are attributed with large inventories of hazardous materials whose release could result in catastrophic events, as observed in the Tianjin Port accident. Such events are typically high-impact low-probability (HILP) accidents since multiple robust safety barriers can significantly ensure the integrity of installations. However, the consequences are extremely serious if the safety barriers broke down due to disaster factors. The current risk assessment methods cannot capture the complex multi-hazard scenarios and the interaction of escalation factors causing domino effects. To overcome such gaps, the present study proposes a quantitative vulnerability assessment method for multi-hazard scenarios triggered by natural events. The vulnerability assessment method considers the exposure of hazards, the sensitivity of causes, and the resilience of asset in modelling the primary event and the possible domino accidents. The proposed method assists in analyzing the risk of domino effects triggered by natural disasters and optimizing the deployment of safety barriers in chemical clusters. The application of the method is demonstrated through a detailed case study.
Article
One of the most heavily employed tools to define countries' policies are the attractiveness indexes. The main purpose of our paper is to formally verify the capacity of an attractiveness index to describe the phenomenon of Foreign Direct Investment (FDI). Severe weaknesses were detected through the use of a framework (principal components analysis and Cronbach's alpha reliability) to analyse indexes, regarding the theoretical background and the adequate selection of indicators, especially the ones from the bottom. Additionally, the study demonstrated that the analyzed FDI attractiveness index built to formally aggregate the factors that impacts foreign direct investment can hardly be applied to emerging economies.
Chapter
Literature about humanitarian logistics (HL) has developed a lot of innovative decision support systems during the last decades to support decisions such as location, routing, supply, or inventory management. Most of those contributions are based on quantitative models but, generally, are not used by practitioners who are not confident with. This can be explained by the fact that scenarios and datasets used to design and validate those HL models are often too simple compared to the real situations. In this chapter, a scenario-based approach based on a five-step methodology has been developed to bridge this gap by designing a set of valid scenarios able to assess disaster needs in regions subject to recurrent disasters. The contribution, usable by both scholars and practitioners, demonstrates that defining such valid scenario sets is possible for recurrent disasters. Finally, the proposal is validated on a concrete application case based on Peruvian recurrent flood and earthquake disasters.
Chapter
This paper focuses its study in the generation of a Supply Chain Resilience strategy for a metal-transformation company located in the City of Puebla, Mexico. The study tends to strengthen the capacity of the company, in terms of resilience in case of any logistic or operational disruption caused by the negative impacts of a disaster. It was suggested to start with a Risk Management Analysis (RMA) following by a Business Continuity Plan implementation. Using the (Define, Measure, Analyze, Improve, and Control) DMAIC methodology, disturbing agents from a national federal agency were analyzed to detect potential risks on the complete Mapping Production Process of the company, to sort those risks per weighted damage impact later. The strategy set up would help to the Tool Manufacturer to control risks better and improve the resilience culture of the company. The risk cost impact was estimated to be reduced from 1.2 M USD to USD 500 k USD. In the second scenario, an AHP was used, but considering other aspects like infrastructure, roads, and so forth, the safety sites were found in the northwest, center-west, center-east, southwest, southeastern, and central areas of the state of Puebla.
Technical Report
Full-text available
An assessment of the robustness of the 2008EPI results requires the evaluation of uncertainties underlying the index and the sensitivity of the country scores and rankings to the methodological choices made during the development of the Index. To test this robustness, the EPI team has continued its partnership with the Joint Research Centre (JRC) of the European Commission in Ispra, Italy. This JRC report shows that the 2008 EPI has an architecture that highlights the complexity of translating environmental stewardship into straightforward, clear-cut policy recipes. The trade-offs within the index dimensions are a reminder of the danger of compensability between dimensions while identifying the areas where more work is needed to achieve a coherent framework in particular in terms of the relative importance of the indicators that compose the EPI framework. The 2008 Environmental Performance Index (EPI) is developed for 149 countries and is based on 25 indicators in six policy categories: Environmental Health, Air Pollution, Water, Biodiversity and Habitat, Productive Natural Resources, Climate Change. The EPI aims to bring a data-driven, fact-based and empirical approach to environmental protection and global sustainability. The validity of the EPI scoring and respective ranking is assessed by evaluating how sensitive the country ranks are to the assumptions made on the index structure and the aggregation of the 25 underlying indicators. The assumptions tested are: • measurement error of the raw data, • choice of capping at selected targets for the 25 indicators, • choice to correct for skewed distributions in the indicators values, • weights of the indicators and/or of the subcomponents of the index, and finally • aggregation function at the policy level (six policy categories).
Research
Full-text available
Composite indicators (composite indicators) have received substantial attention in recent years and various methodologies have been developed to handle different aspects of the issue. This report examines a number of methodologies with a view to clarifying how they relate to the development of composite indicator. Several methods are investigated such as aggregation systems, multiple linear regression models, Principal components analysis and factor analysis, Cronbach alpha, Neutralization of correlation effect, Efficiency frontier, Distance to targets, Experts opinion (budget allocation), Public opinion, and Analytic Hierarchy Process. The report further examines twenty-four published studies on this topic in a number of fields such as environment, economy, research, technology and health, including practices from the Directorates General of the European Commission. In this report, we offer for each composite indicator reviewed general information on the number and type of sub-indicators, on the preliminary treatment (normalisation, detrending etc.) and on the weighting system considered. Finally, each composite indicator is briefly commented.
Article
This paper reviews research traditions of vulnerability to environmental change and the challenges for present vulnerability research in integrating with the domains of resilience and adaptation. Vulnerability is the state of susceptibility to harm from exposure to stresses associated with environmental and social change and from the absence of capacity to adapt. Antecedent traditions include theories of vulnerability as entitlement failure and theories of hazard. Each of these areas has contributed to present formulations of vulnerability to environmental change as a characteristic of social-ecological systems linked to resilience. Research on vulnerability to the impacts of climate change spans all the antecedent and successor traditions. The challenges for vulnerability research are to develop robust and credible measures, to incorporate diverse methods that include perceptions of risk and vulnerability, and to incorporate governance research on the mechanisms that mediate vulnerability and promote adaptive action and resilience. These challenges are common to the domains of vulnerability, adaptation and resilience and form common ground for consilience and integration.
Article
Composite indicators are increasingly used for bench-marking countries’ performances. Yet doubts are often raised about the robustness of the resulting countries’ rankings and about the significance of the associated policy message.We propose the use of uncertainty analysis and sensitivity analysis to gain useful insights during the process of building composite indicators, including a contribution to the indicators’ definition of quality and an assessment of the reliability of countries’ rankings.We discuss to what extent the use of uncertainty and sensitivity analysis may increase transparency or make policy inference more defensible by applying the methodology to a known composite indicator: the United Nations technology achievement index.