POLITECNICO DI TORINO

Repository ISTITUZIONALE

PEOPLES: A Framework for Evaluating Resilience

Original
PEOPLES: A Framework for Evaluating Resilience / Cimellaro, GIAN PAOLO; Renschler, Chris; Reinhorn, Andrei M.;
Arendt, Lucy. - In: JOURNAL OF STRUCTURAL ENGINEERING. - ISSN 0733-9445. - ELETTRONICO. - 142:10(2016),
p. 04016063.

Availability:
This version is available at: 11583/2652922 since: 2019-10-18T12:50:35Z

Publisher:
American Society of Civil Engineers (ASCE)

Published
DOI:10.1061/(ASCE)ST.1943-541X.0001514

Terms of use:
openAccess
This article is made available under terms and conditions as specified in the corresponding bibliographic description in
the repository

Publisher copyright

(Article begins on next page)

11 June 2020
 1 PEOPLES: A FRAMEWORK FOR EVALUATING RESILIENCE
2
3 Gian Paolo Cimellaro1 , Chris Renschler2 , Andrei M. Reinhorn3 , Lucy Arendt4 ,

4 ABSTRACT

5 In recent years, the concept of resilience has been introduced to the engineering field in particular

6 related to disaster mitigation and management. However, the built environment is only part of the

7 elements that support community functions. Maintaining community functionality during and after a

8 disaster, defined as resilience, is influenced by multiple components. The paper is proposing a

9 framework for measuring community resilience at different spatial and temporal scales. Seven

10 dimensions are identified for measuring the community resilience: Population and Demographics,

11 Environmental/Ecosystem, Organized Governmental Services, Physical Infrastructures, Lifestyle and

12 Community Competence, Economic Development, and Social-Cultural Capital. They are

13 summarized with the acronym PEOPLES. Each dimension is characterized by a corresponding

14 performance metric that is combined with the other dimensions using a multi-layered approach.

15 Therefore, once a hybrid model of the community is defined, the proposed framework can be applied

16 to measure its performance against any type of extreme event during emergency and in long term

17 post-disaster phases. A resilience index can be determined to reflect all, or part, of the dimensions

18 influencing the events. Several applications of part of such framework can already be found in

19 literature for different types of infrastructures, physical and organizational (e.g. gas network, water

20 distribution networks, health care facilities etc.). The proposed framework can be used as decision

21 support by stakeholders and managers and it can help planners in selecting the optimal restoration

22 strategies that enhance the community resilience index.

1 Vis iting Pr ofes s or , D epar tm ent of Str uctur al & Env ir onm ental Engineer ing, U niv er s ity of Califor nia, Ber k eley, Ber k eley, U SA, E-
mail: [email protected]
2 As s ociate Pr ofes s or , Depar tm ent of G eogr aphy, U niv ers ity at Buffalo, SU NY, 116 Wilk eson Q uad, Buffalo, NY 14261, U .S.A. E-
mail: rens [email protected].
3 Pr ofes s or Em er itus , D epar tm ent of Civ il, Str uctur al & Env ir onm ental Engineering, U niv er s ity at Buffalo (SU NY), 135 Ketter H all,
North Cam pus, 14260-4300, Buffalo, NY, U .S.A. [email protected]
4 As s ociate Pr ofes s or of Managem ent, As s ociate D ean, College of Pr ofes s ional Studies , D irector , Aus tin E. Cofr in School of Bus ines s ,
Univ ersity of Wiscons in-Green Bay, 2420 Nicolet Dr, G reen Bay, WI 54311, USA, [email protected]

1
23 Keywords: Resilience; disaster resilience; infrastructures, performance metric; community resilience;
24 environment/ecosystem; organized governmental services; physical infrastructure.
25

26 INTRODUCTION AND DEFINITION OF RESILIENCE

27 Recent disasters around the world have shown clearly that not all the threats can be averted. Modern

28 societies are trying to enhance their resilience against extreme events after realizing that they cannot

29 prevent every risk from being realized, but rather they must manage risks and adapt minimizing the

30 impact on population and their support systems.

31 The concept of resilience has several definitions, because of its broad utilization in ecology, social

32 science, economy and engineering fields, with different meanings and implications. As Klein et al.

33 stated (2003), the term derives from the Latin word ‘resilio’ that means ‘to jump back’. The term has

34 been used first in psychology and psychiatry in 1940s, and it is mainly accredited to Norman

35 Garmezy, Emmy Werner and Ruth Smith (Garmezy, 1974; Werner and Smith, 1989). Later the

36 concept of resilience established in the field of ecology by Holling (1973) who stated that the

37 resilience of an ecological system is “a measure of the persistence of systems and of their ability to

38 absorb change and disturbance and still maintain the same relationships between populations or state

39 variables. Stability represents the ability of a system to return to an equilibrium state after a

40 temporary disturbance; the more rapidly it returns to equilibrium and the less it fluctuates, the more

41 stable it would be”. An extended literature review about resilience has been assembled in the past

42 (see Table -1) with each contribution adding new nuances. Primarily resilience has been defined in

43 context to the speed of systems to go towards equilibrium (Adger, 2000) or capability to cope and

44 bounce back, ability to adapt to new situations (Comfort, 1999), be inherently strong, flexible and

45 adaptive (Tierney & Bruneau, 2007), or ability to withstand external impacts and recover with least

46 outside interferences (Mileti, 1999). After the original definition of resilience in ecological systems,

47 the word expanded its meaning to engineering, social and economical fields.

48 Engineering resilience is defined as the capability of a system to maintain its functionality and to

49 degrade gracefully in the face of internal and external changes (Allenby and Fink, 2005). The main

2
50 difference in defining resilience arises between the engineering approach where resilience occurs by

51 recovering towards a previous or an improved stable state (Bruneau et al., 2003), and the ecological

52 approach where resilience is achieved moving towards a different system state (Handmer & Dovers,

53 1996).

54 Social resilience is defined as the ability of groups or societies to cope with external stresses and

55 disturbances because of social, political, and environmental change (Adger, 2000).

56 Economic resilience is defined as the inherent ability and adaptive response that enables individual

57 business firms and entire regions to avoid maximum potential losses (Rose and Liao, 2005). It has

58 mainly been studied in context to seismic response and recovery (Tierney, 1997), community

59 behavior (Chang & Shinozuka, 2004) and disaster hazard analysis (Rose, 2004).

60 Research advancements have proven that resilience should be addressed at the large-scale level and

61 not just locally. Bruneau et al. (2003) identified four types of resilience that should be adequately

62 measured: technical; organizational; social; and economical, (TOSE). Technical and economical

63 resilience, are mainly related to the physical systems, while organizational and social resilience, are

64 related to the society and the non physical systems.

65  Technical resilience describes the capability of a system to function and perform adequately.
66  Organizational resilience describes the ability of the organization(s) to manage the system. For
67 example, measures of organizational resilience could include how well emergency units function,
68 how quickly spare parts are replaced, how quickly repair crews are able to reach the affected
69 components of a system, etc.
70  Social resilience concerns how well society copes with the loss of services because of a disaster.
71 For example, social resilience can become the most critical dimension of the global resilience,
72 because of severe blackouts during a disaster.
73  Economic resilience describes the capability to reduce both indirect and direct economic losses
74 (Rose and Liao, 2005).

75 Following the initial resilience framework by Bruneau et al. (2003), other frameworks have been

76 developed expanding and identifying different metrics to quantify resilience. For example, Chang

77 and Shinozuka (2004) refined the method proposed by Bruneau (2003) by proposing a metric of

3
 78 system functionality Q that is evaluated comparing the extreme events scenario with the pre-event

79 conditions and they applied the method to the case study of Memphis water system. Miles and

80 Chang (2006) presented a comprehensive functionality restoration model that establishes the

81 relationships between community’s household, business and lifeline networks. The same year

82 Cagnan et al. (2006) developed a discrete event simulation model for modeling the post-earthquake

83 restoration process of an electric power system. The resilience concept as input to decision support

84 methodologies has been applied to hospitals (Cimellaro et al., 2010b; Cimellaro and Pique`, 2014a),

85 lifeline structures (Ouyang and Duenas-Osorio, 2011, Cimellaro et al., 2014b-c) and cities (Chang et

86 al, 2014) using different optimization methods based on economic (Chang and Shinozuka, 2004),

87 downtime (Cagnan et al., 2006) or multi-criteria analysis (Javanbarg et al., 2012).

88 Several methods for the quantification of infrastructures’ resilience have been proposed that can be

89 grouped in probabilistic methods (Miller-Hooks et al, 2012, Queiroz et al., 2013), graph theory

90 methods (Berche et al, 2009; Dorbritz, 2011), fuzzy logic methods (Heaslip et al., 2010) and

91 analytical methods (Cimellaro et al., 2010a; Tamvakis and Xenidis, 2013). For example, Tamvakis

92 and Xenidis (2013) proposed a framework base on entropy theory concepts. Entropy describes the

93 system’s disorder at a given point in time and it is measurable in a single metric, analogous to

94 resilience, which describes the system’s potential of recovering to a desired system’s condition.

95 It should be noted that the literature review presented above it is not exhaustive; however, most of

96 the works cited herein summarize previous works to quantify resilience, therefore this review is

97 adequate for the classification of the different trends in the quantification of resilience for

98 infrastructures and communities in general. However, due to its complexity, a comprehensive model

99 that quantifies resilience of local, metropolitan or disperses communities and considers all

100 infrastructures and their interaction is still missing.

101 This paper suggests a novel framework to evaluate resilience of a community and to assess the

102 performance of critical infrastructures and their interdependencies while taking into account the

103 influence of the human behavior, societal, organizational, and economic issues. The framework is

4
104 based on seven major groups of characteristics, defined here as dimensions, which can measure

105 resilience at different scales. These are: Population and Demographics, Environmental/Ecosystem,

106 Organized Governmental Services, Physical Infrastructure, Lifestyle and Community Competence,

107 Economic Development, and Social-Cultural Capital and are identified with the acronym

108 PEOPLES: The framework can be used for resilience-based design (RBD) at different spatial

109 (local, regional etc.) and temporal (emergency response, recovery and reconstruction phase, etc.)

110 scales. It can also be used by decision makers for disaster and post-disaster management,

111 minimizing all the possible consequences following an extreme event, both natural and man-made

112 allowing the perturbed system to return to the initial conditions as quickly as possible.

113 Mathematical definition of Resilience

114 The definition of Resilience used in this paper is the one described also by Cimellaro et al.,

115 (2010a), where a resilience index R of a system is defined as the normalized area underneath the

116 functionality-performance function Q(t) shown in Figure -1Error! Reference source not found.,

117 while analytically is defined as

tOE TLC
118 R r    QTOT  r , t  TLC dt (1)
tOE

119 where QTOT(t) is the global functionality-performance function of the area considered (local,

120 regional, etc) which is described in the next paragraph; TLC is a control time for the period of interest;

121 t 0E is the time instant when the event happens; r is a spatial vector defining the position P in the

122 region where the resilience index is evaluated (Cimellaro et al. 2010b). In general, the resilience

123 index can be applied to different fields (e.g. engineering, economic, social science etc.) and it can be

124 used at various temporal and spatial scales. The first step to quantify the resilience index (R) is to

125 define the spatial scale (e.g. individual building, city, region, state, etc.) of the problem of interest,

126 because large disasters tend to expand over interacting large spaces. The second step is to define the

127 temporal scale (emergency response vs. long term reconstruction phase) of the problem of interest;

5
128 the selection of the control period TLC affects the resilience index R, therefore it should be maintained

129 fixed when comparing different scenarios.

130 THE SEVEN DIMENSIONS

131 The proposed framework expands the initial research at the Multidisciplinary Center of Earthquake

132 Engineering Research (MCEER) and links with the previously identified resilience characteristics

133 (technical, organizational, societal, and economic) and with the resilience attributes (r4 : robustness,

134 redundancy, resourcefulness, and rapidity) (Bruneau et al. 2003; Bruneau and Reinhorn, 2007;

135 Cimellaro et al. 2010b).

136 The new framework, identified by the acronym PEOPLES, incorporates the initial MCEER’s

137 definitions of service functionality of community components (assets, services, demographics) and

138 parameters influencing resilience, all assembled into a layered approach. The seven dimension

139 groups of the PEOPLES’ framework (Renschler et al. 2010, 2011), listed below, are further

140 explained in this section.

141 (1) Population and demographics;

142 (2) Environment/ecosystem;
143 (3) Organized government services;
144 (4) Physical infrastructure;
145 (5) Lifestyle and community competence;
146 (6) Economic development;
147 (7) Social-cultural capital;

148 The specific dimensions represent groups of interwoven societal, technical, economic and

149 organizational issues. Although other definitions of multiple dimensions were described by

150 Rockefeller Foundation (Huq et al, 2007), United Nations (2013), the Institute for Social and

151 Environmental Transition (ISET) (Tyler and Moench, 2012), Arup (da Silva and Morera, 2014), the

152 aforementioned dimension groups were selected based on similar characteristics. A resilience index

153 can be established for each of the above dimensions; however, the whole community resilience

154 would be influenced by all, or only some dominant dimensions, as it is described in the following.

155 Table -2 shows the extended list of components and sub-components of the “PEOPLES Framework”,

6
156 while the detailed description of each dimension is given in the next paragraphs. The description of

157 the dimensions follows the order of the PEOPLES acronym, so it is not based on a specific

158 hierarchy. Additional details can also be found in Renschler et al. (2010, 2011).

159 Population and demographics

160 The Population and demographics dimension describes and differentiate the communities using

161 specific parameters (e.g. the median income, the age distribution etc.) which might be critical for

162 understanding its economics, health, etc. These parameters help describing the social vulnerability

163 which is defined as the incapacity of societies, organizations and citizens to resist at the exposure of

164 multiple undesirable events. These events are generated by the interaction in the society, the

165 institutions and the systems of different cultural values. Social vulnerability is a pre-existing state of

166 the community that affects the society’s capacity to get ready for and recover from an undesirable

167 event.

168 This dimension can be measured using a social index that describes the socioeconomic status, the

169 composition of the population (e.g. elderly and children), the population density, the rural

170 agriculture, the race, the gender, the ethnicity, the infrastructure employment, and the county

171 debt/revenue.

172 Following the general definition of Resilience given in Equation (1), a possible functionality-

173 performance metric (Qp ) for the Population and Demographic dimension could be the social

174 vulnerability index (SVI) (Barry et al., 2011). The domains that form the basis of the Social

175 Vulnerability Index (SVI) are 1) socioeconomic status, 2) household composition and disability, 3)

176 minority status and language, and 4) housing and transportation. The data can be collected from the

177 European Census of Population and Housing at the census tract level. Each of the domains can be

178 described as per following variables:

179  Socioeconomic status comprising income, poverty, employment, and education variables;
180  Household Composition and Disability, comprising age, single parenting, and disability
181 variables;

7
182  Minority Status and Language, comprising race, ethnicity, and language proficiency
183 variables;
184  Housing and Transportation, comprising housing structure, crowding, and vehicle access
185 variables.
186

187 Each of the above census variables, except per capita income, could be ranked from highest to

188 lowest across all census tracts, to construct the SVI. Per capita income should be ranked from lowest

189 to highest because, unlike the other variables, a higher value indicates less vulnerability. A percentile

190 rank is then calculated for each census tract over each of these variables. A percentile rank is defined

191 as the proportion of scores in a distribution that a specific score is greater than or equal to. Percentile

192 ranks are calculated by using the formula:

193 Percentile Rank   Rank  1 /  N  1 (2)

194 where N is the total number of data points, and all sequences of ties are assigned the smallest of the

195 corresponding ranks. In addition, a tract-level percentile rank is calculated for each of the four

196 domains based on an across-the-board sum of the percentile ranks of the variables comprising that

197 domain. Finally, an overall percentile rank for each tract could be calculated as the sum of the

198 domain percentile rankings. This process of percentile ranking—for all variables, for each domain,

199 and for an overall SVI— is then repeated for the individual communities.

200 Others social vulnerability index (SoVI) (Cutter, 1996; Cutter et al., 2000) can be used as well, such

201 as the one proposed by Cutter that integrates exposure to hazards with the social conditions that

202 make people vulnerable to them.

203

204 Environmental/Ecosystem

205 In the PEOPLES Framework, the environmental and ecosystem dimension measure the capability of

206 the ecological system to go back to its pre-event condition defined as its basic functionality. This

8
207 dimension measures the capability of an ecosystem to deal with disturbance, but also the amount of

208 disturbance an ecosystem can absorb without considerably varying its processes and structures.

209 In order to measure the environmental/ecosystem dimension of functionality and resilience, key

210 indicators should be integrated together such as air, water and soil quality, biodiversity, and other

211 natural resources.

212 One possible functionality-performance metric for this dimension is the Normalized Difference

213 Vegetation Index (NDVI), which is evaluated from satellite-derived remote sensing images that

214 analyze the density of green vegetation across an area (Rouse et al., 1973). The NDVI index (≤1) is

215 given by

216

217 NDVI   NIR – Red   NIR  Red  (3)

218

219 where Red are the visible (red) infrared absorption bands and NIR are the near infrared absorption

220 bands. Indeed, the NDVI index is highly correlated with the Aboveground Net Primary Productivity

221 index (ANPP) (Pettorelli, 2005; Olofsson et al., 2007), that is based on filed measurements of the

222 biomass accumulation and therefore can be considered as an indicator of the ecosystem resilience.

223 Several applications can be found in literature where the NDVI values obtained from Landsat images

224 have been used to observe the restoration of the vegetation after a fire (Diaz-Delgado et al., 2002)

225 and using time series analysis (Simoniello et al., 2008).

226 The NDVI index in Equation (3) can be used to quantify the Environmental/Ecosystem dimension by

227 comparing the NDVI values before and after the event, to determine the variations of ecosystem

228 productivity through the space and the time caused by natural disasters such as fire, flood,

229 hurricanes, tsunami, etc. Instead, in other types of disasters such as blizzards, terrorist attacks etc.,

230 the variation of this index could be negligible, because the vegetation density might not be altered,

231 while other indicators could be more relevant.

232
9
233 Organized governmental services

234 The dimension of organized governmental services includes legal and security services (e.g. Police,

235 emergency departments, fire departments, the military etc.), and also the public health, the hygiene

236 departments, the cultural heritage departments etc.. Each of the above mentioned organized

237 government services play a key role in sustaining societies before and after an extreme event.

238 Key indicators for this dimension include the number of available response units and their capacity,

239 if they are opportunely normalized with respect to the number of residents involved. This dimension

240 can provide a measure on how much the various organized government services participate in

241 emergency preparedness planning (e.g. survey, etc.) developing a memorandum of understanding

242 (MOUs) and other mutual aid agreements (Tierney, 2009).

243 Other examples of performance metric for this dimension can be the patient waiting time (WT), that

244 is the time the patient waits before receiving assistance (Cimellaro et al., 2011), in the case when the

245 organized service is the Emergency Department (ED) of an hospital. This specific indicator

246 measures the ability of the ED to provide service to all patients after a disaster. Analytically the

247 functionality-performance metric (Qo ) is given by

WT0
248 Qo  (3)
WT

249 where WT0 is the waiting time in normal operating conditions, while WT is the waiting time during

250 the emergency.

251 The deficiencies associated with this resilience dimension have been observed during the 2010 Haiti

252 Earthquake, where the lack of organized government services and orderly control together with a

253 perception that the government could not deal with the disaster reduced the response and recovery

254 processes. In contrast, this resilience’s dimension dominated the post 2010 Darfield earthquake in

255 New Zealand, because the local, territorial and national government services were well organized to

256 provide a quick restoration process. The organizational response during an emergency is most likely

257 to be effective and improve resilience when it blends discipline and agility (Harrald, 2006).

10
258 Discipline and proper reaction are guaranteed by emergency plans, training activities, exercises and

259 mutual aid agreements that encourage action toward common goals (Weick, 1995; Weick et al.,

260 2005). Agility, flexibility, adaptability, and improvisations are entities which enhance resilience of

261 a society, through volunteers, spontaneous helping behavior, and emergency groups which infuse

262 resources and creativity into disaster response activities (Stallings and Quarantelli, 1985; Drabek and

263 McEntire, 2002). The emergency management system following a disaster involves different groups

264 such as the emergency response teams, the volunteers, the mass media, the economic network etc.

265 These groups, instead of transferring information in a hierarchic way in the top-down direction, use

266 an upward flow of information, which is the most preferred direction of communication used during

267 disasters. In fact, the experience in the field has shown that decentralized networks with flatter

268 organizations and less hierarchical structure are quicker in responding to disasters because they

269 promote a free flow of information (Simoniello and Quarantelli, 1985).

270

271 Physical infrastructure

272 The physical infrastructure dimension includes facilities (e.g. housing, commercial and industrial

273 facilities, and cultural facilities) and lifelines (food supply, utilities, transportation, communication

274 networks etc.) within a built environment (Cimellaro et al., 2014b). While facilities are traditional

275 essential life support for its population, lifelines are essential utilities which serve communities

276 across all jurisdictions such as: (a) energy utilities (e.g. power and natural gas networks (Bruneau et

277 al, 2003, Cimellaro et al., 2014a)); (b) transportation systems (e.g. highways, railroads, airports,

278 seaports etc.); (c) water, storm-water and sewerage pipelines; (d) communication systems; and (e)

279 health care facilities (e.g. hospitals, etc) (Cimellaro et al., 2011), etc. Functionality of physical

280 infrastructures has an important impact on the restoration process following a disaster; therefore, the

281 organized government services work actively to restore their functionality. Such interactions are

282 essential in resilient communities.

11
283 For example, following Hurricane Katrina in 2005, after the evacuation of New Orleans, attention

284 has shifted towards the restoration of the physical infrastructures. The pictures of damages have

285 been used to communicate to the media in the world the consequences of the hurricane and of the

286 subsequent flood (e.g. collapse of critical facilities such as churches, schools, and hospitals). The

287 critical facilities were not able to provide their services without water and electricity. The damaged

288 schools affected the community’s self confidence to overcome the disaster and restore the initial

289 functionality. The roads full of debris created an obstacle to the supply chain, therefore the economy

290 in the region could not restart, because even if shops and companies re-opened they could not be

291 accessible and even if they relocated for a short term, the previous customers were having some

292 difficulties in finding the new location.

293 After a disaster, the restoration of physical infrastructures remains a technical problem that is also

294 related to the socio-political events and the economic situation. The resilience dimension of physical

295 infrastructure should also take into account the interdependencies between the different types of

296 infrastructures and sectors during the analysis (Cimellaro and Solari, 2014c). Different functionality-

297 performance metrics for this dimension are available in literature (Cimellaro et al., 2014a-b-c) and

298 vary for every type of infrastructure (e.g. gas, water, transportation, etc). However, a general

299 definition of functionality-performance metric (Qph ) for this dimension which applies to every type

300 of infrastructure is given by

 n(t )
QPh (t ) 
t0 E
301 (3)
nTOT

302 where n(t) is the number of households without service at a given instant t and ntot is the total

303 number of households with service before the emergency.

304 There are also other examples for housing units where a possible functionality-performance metric

305 might be the proportion of housing stock not rated as substandard or hazardous and vacancy rates for

306 rental housing (Tierney, 2009). Examples of functionality-performance metrics for the

12
307 communication networks might be the (i) acceptable linkages between official and unofficial

308 information sources, (ii) the number of ties between the mass media and the emergency management

309 entities, (iii) the sufficiency of measures for communicating the public’s need and information after

310 the disaster (Tierney, 2009).

311

312 Lifestyle and Community Competence

313 Lifestyle Community competence dimension deals with flexibility, creativity and problem solving

314 skills of a community through also political partnerships (Norris et al., 2008). Principal elements of

315 this dimension include collective actions and decision making, collective efficacy and empowerment

316 and quality of life. This dimension captures both the raw abilities of a community (e.g., skills to find

317 multifaceted solutions to complex problems through the engagement in political networks) and the

318 perceptions of a community (e.g. perception to have the ability to do a positive change through a

319 common effort that relies on peoples’ aptitude to resourcefully envision a new future and then move

320 in that direction) (Brown and Kulig, 1996). In fact, the societies that believe that they can restore,

321 renew and rebuild themselves are expected to be more determined when facing a disaster or in

322 general, any type of changes.

323 Quality of life surveys can be used as indicators of this perception, because they reveal whether

324 people inside the community are devoted to their community and willing to engage in the activities

325 necessary to maintain the community alive, before or after the disaster strikes. Examples of

326 performance metrics for the community competence in normal condition before the disaster might be

327 the number of immigrants, the number of citizens involved in politics, etc.

328 Specific performance metrics for this dimension directly related to the disaster might be the

329 extensiveness of community warning procedures and plans, measured using for example the number

330 of citizens involved, the number of organizational disaster training programs, etc. (Tierney, 2009).

331

13
332 Economic development

333 The economic development dimension is composed of both a static and a dynamic assessment. The

334 static assessment is the market activity of the current economy of a community, while the dynamic

335 assessment corresponds to the economic development which is the community’s ability to

336 continuously sustaining the economic growth. Resilient communities are characterized by the

337 community’s capacity to replace goods, services, shift employment patterns when is needed. In other

338 words, they are associated to the employment, the variety in production and services. The economic

339 dimension consists of three sub-categories: (i) the production within the industry, (ii) the distribution

340 of employments within the industry, and (iii) the financial services.

341 The key indicators of the economic development dimension can be: (i) the percentage of the

342 inhabitants that are working in the diverse industries, and (ii) the variability of the distribution of

343 employments in the different industries which are in the community (iii) the literacy rate, (iv) the life

344 expectancy, (v) the poverty rates. Other examples of indicators for this dimension are related to the

345 community performance following a disaster and are: (i) the adequacy of plans for inspecting

346 damaged buildings following disasters, (ii) the extent of evacuation plans and drills for high-

347 occupancy structures, (ii) the adequacy of plans for post-disaster commercial restoration, etc.

348 (Tierney , 2009). Because of these indicators, this dimension is interdependent with the Population

349 and Demographics dimension.

350 Analytically one possible functionality-performance metric (QE) for this dimension is given by

351

per _ inco  med _ inco  emply  hsg _ value  buss  insurance

352 QE  (3)
6

353

354 where per_inco= per capita income; med_inco= median household income; employ = employed

355 civilian population; hsg_value= median value of owner occupied housing units; buss = business

356 establishments; insurance = population with health insurance.

14
357

358 Social/cultural capital

359 The Social/cultural capital dimension includes numerous sub-categories such as: (i) education

360 services, (ii) child and elderly services, (iii) cultural and heritage services, (iv) community

361 participation etc. The key indicators in normal condition for this dimension are: (i) the number of

362 members belonging to the diverse civil and community organizations, (ii) the surveys of leaders and

363 their perception. The key indicators in emergency conditions are: (i) the existence of community

364 plans targeting transportation-disadvantaged residents, (ii) the adequacy of post-disaster sheltering

365 plans, (iii) the adequacy of plans for incorporating volunteers into official response activities, (iv) the

366 adequacy of donations management plans, (v) the community’s plans to manage various networks

367 (Tierney, 2009).

368 In relation to disaster phases’ activities, socio-cultural capital dimension can be measured using the

369 following six components suggested in the literature (Mayunga, 2009):

370 1. Participation in voluntary organizations (volunteerism): this component was measured using

371 registered non-profit organizations;

372 2. Involvement in social groups (association densities): the involvement in social groups was

373 measured using recreational centers (bowling centers, and fitness centers), golf clubs, and

374 sport organizations;

375 3. Civic and political participation: this social capital component was measured using three

376 indicators including registered voters, civic and political organizations, and Census response

377 rates for the decennial population and housing survey;

378 4. Religious participation: it was measured using religious organizations;

379 5. Community attachment: the community attachment component was measured using owner-

380 occupied housing units;

15
381 6. Connection to working places: this element was measured using two indicators including

382 professional organizations and business organizations;

383 Then a three-step procedure is employed in calculation of the socio-cultural capital dimension: (i)

384 scale adjustment of indicators, (ii) standardization or normalization, and (iii) creation of the socio

385 cultural community resilience index.

386 In addition, the social support underlies several services connected with the social/cultural capital,

387 such as “helping behaviors within family and friendship networks” and the “relationships between

388 individuals and their larger neighborhoods and communities” (Norris et al., 2008). In fact, the

389 habitants of a community tend to manifest their sense of community and to bond with other members

390 of the same group by providing social and cultural services. However, this emotional connection to

391 the community is not necessary related only to the residents in those places (Manzo and Perkins,

392 2006). For example, several displaced residents of New Orleans after Hurricane Katrina expressed

393 the desire to return home with a strong “place attachment”, regardless the job they had and the

394 people they knew. These residents are an important resource for the community, because if they will

395 be provided with housing and employment after the disaster, they will act in order to restore the

396 community to the initial condition before the disaster. The citizen participation in community

397 organizations (e.g. religious congregations, school and resident associations, neighborhood watches,

398 self-help groups etc.) is a way of demonstrating one’s care for their community, one’s care for

399 meeting and understanding one’s fellow citizens and it increases individuals’ circle of influence and

400 perception of control (Norris et al., 2008).

401

402 MATHEMATICAL FORMULATION OF THE PEOPLES FRAMEWORK

403 General description of the methodology and the community hybrid model

404 The main part of the methodology consists in developing a community hybrid model, coupling the

405 Network models which will be used to model the physical infrastructures networks such as the power

16
406 and the water, with the Agent based models which will be used to model the socio-technical

407 networks such as the Emergency Medical Technicians and the fire brigade (Figure -2). Inside the

408 ABM models, the emotions in the agents will be modeled using the extended version of Belief-

409 Desire-Intention modeling framework proposed by Zoumpanaki (2010) that has been expanded and

410 adapted to the proposed methodology (Figure 3).

411 Both types of models will be integrated in a hybrid framework and a matrix approach will be used to

412 describe the interdependencies between the different layers. Each layer represents an infrastructure

413 (Figure 3) and is described by an adjacency matrix A, while a D matrix will describe the

414 interdependencies between the nodes of the different layers (e.g. D Water Power) and it will be obtained

415 using an extended version of the Haimes’ input-output inoperability matrix (IIM) (Haimes et al.,

416 2005). For example, in Figure -2, the hospital is a node of the EMT layer and it is interdependent

417 with the power and the water network. Therefore, a D matrix describing the interdependencies

418 between the EMT layer and the water and power layer will be determined using Haimes model.

419 The matrix D is composed of constant scalars terms if the system does not change though the time.

420 In reality, the interdependent networks might change through the time their interconnectedness as

421 shown in some applications (Fantini et al., 2014), however the proposed approach can also be

422 applied in this case. The proposed approach will require substantial computational power if the

423 spatial and temporal dimensions of the problem increase, therefore the use of parallel computing is

424 recommended in these cases. Below is shown in simple terms how the agent base models and the

425 network model interact in the proposed methodology. Once the hazard is determined (e.g.

426 earthquake event), the corresponding damage in the infrastructure networks is determined using

427 fragility analysis combined with graph theory in order to identify the nodes of the network that will

428 not be functional following the extreme event. Because of physical infrastructure disservice, also the

429 socio-technical networks operating in the community will be affected. For example the road

430 transportation disservice, might limit the capacity of the emergency rescue teams to move and

431 operate in the community. The water network disservice might limit the capacity of the fire brigade

17
432 to extinguish fires etc. In order to study these interdependencies the network models and the agent-

433 based model need to run simultaneously. So the output of the damage analysis in the network models

434 should be used as input to modify the characteristics of the agent-based models, such as the

435 extension of the environmental space (e.g. roads), the capacity to perform certain actions (e.g.

436 extinguish fire) of the agents, etc.

437 Da qui

438

439 Resilience index and performance metrics

440 Once the hybrid model in Figure -2 is built, it is necessary to identify the performance metrics to

441 estimate the resilience of each infrastructure. Several approaches exist in literature for hospitals

442 (Cimellaro et al., 2011), lifeline structures (Ouyang et al., 2012, Cimellaro et al., 2014c) and cities

443 (Chang et al, 2014). Once the proper performance metric is selected, the degree of interdependency

444 between an infrastructure x and y is described using a matrix D x  y which is able to identify the exact

445 location of the interdependency in the network (e.g. node or link). However, sometimes it is also

446 useful to identify a global index I that measures the degree of interdependency between the different

447 infrastructures, in order to have a global evaluation of the community performance and to assign an

448 unbiased evaluation of the weight (or important factor) to each infrastructure. This index can be

449 determined using time series analysis (Cimellaro et al., 2014c) or from linear algebra manipulation of

450 the D x  y matrix etc. Then the indices I can be grouped into an infrastructure Interdependency

451 Matrix (IM) (Figure 4). The infrastructures considered in the analysis of the community are listed in

452 the rows and the columns, while in each cell is shown the degree of interdependency (from 0 to1)

453 between them. The sum over the columns gives the dependent factor of the specific lifeline, while

454 the sum over the rows gives the importance factor of a specific lifeline. Ideally, the target is to

455 realize a community where all lifelines are independent, so IM will be an identity matrix. As

456 mentioned above, the IM can also be used to have an unbiased estimation of the weight coefficients

18
457 to assign to each infrastructure considered in the layered approach shown in Figure -2. Once the

458 spatial and temporal boundaries of the problem at hand are defined, the performance metrics of all

459 the resilience dimensions are aggregated following the procedure described in the paper of Cimellaro

460 et al. (2014c). The global resilience indicator at the community level is evaluated using the

461 following equation

462 R    Ri  wi  (6)
i

463 where Ri is calculated using Equation (1). For example if it is considered the physical infrastructure

464 dimension, Ri is the resilience indicator of a specific infrastructure, while wi is the weight factor

465 describing the interdependencies between the different indicators. The coefficient wi are determined

466 using a time series analysis approach borrowed by the economic field which is based on the analysis

467 of the cross correlation function (CCF). The procedure can be applied to all the components and

468 subcomponents of the PEOPLES framework in order to take into account the interdependencies

469 between the different variables. Further details about the methodology can be found in Cimellaro et

470 al. (2014c).

471 The selection of the proper performance metric for the critical infrastructures plays a key role in the

472 analysis. Even if a realistic and predictive model is developed, the results might be affected by the

473 selection of the final performance function adopted to evaluate the community resilience index using

474 the methodology shown in Figure -2. Different innovative approaches to measure functionality are

475 available in literature and they include agent-based modeling, input-output models, mathematical

476 models and game theory (Pederson et al., 2006). Therefore, once the approach and the geographic

477 scale is selected, the global performance indicator QTOT can be plotted over the region of interest

478 using a contour plot at a given instant of time t, so the time-dependent functionality maps can be

479 obtained. When also the control time TLC is defined, the resilience contour map of the region of

480 interest can also be plotted. The Resilience contour maps are obtained by integrating the

481 functionality maps over time using Equation (1), therefore the resilience maps will be time

19
482 independent, but they will vary in space from point to point in the selected region. Finally, the

483 community resilience index Rcom is given by the double integral over time and space as follows

tOE TLC
484 Rcom   R  r  AC dr    QTOT  r , t   ACTLC  dtdr (3)
AC AC tOE

485 where Ac is the area of the selected region. The contour plot of each dimension can be combined

486 with the other plots using a layered approach. Then a radar graph is built (Figure -2) and the internal

487 area will determine the final score of the resilience index that will be used to recognize the priority

488 resilience actions to be taken in the community.

489

490 Resilience performance levels

491 The objective of Performance Based Seismic Engineering (PBSE) is to design, construct and

492 maintain facilities with better damage control, coupling the expected or desired performance levels

493 with the levels of seismic hazard. Generally the levels focus on the performances a structure can

494 hold during the shaking and are related to engineering demand parameters such as deformations.

495 More recently SPUR (Bonowitz, 2009), the San Francisco planning and Urban Research

496 Association, introduced other definitions of performance levels for infrastructures based on recovery

497 target states combining safety and recovery time. Five performance measures for buildings have

498 been identified: (i) Safe and Operational; (ii) Safe and usable during repair; (iii) Safe and usable

499 after repair; (iv) Safe but not repairable; (v) Unsafe.

500 The proposed Resilience Performance Levels (RPL) focus on building performance after the

501 earthquake, recognizing the importance of the temporal dimension (Recovery time TRE) in the

502 assessment of the RPLs of structures and communities in general.

503 In this paper a 2-dimensional performance domain consisting of Performance Levels PL(i, j), defined

504 by the combination of functionality (index j) and recovery time (index i) is proposed. By accounting

20
505 for the effect of the temporal dimension, a 3-dimensional performance matrix (Figure -5) can be

506 visualized as a set of predefined joined performance domains (“masks”) for different seismic

507 intensity level, IM and different RPLs.

508 The resilience performance levels can be defined using expert opinions as well as public interviews

509 which will allow identifying the acceptable and desired performance levels by citizens for different

510 type of infrastructures, for example.

511 Restoration models and recovery time

512 The restoration phase and the recovery time are key element for the quantification of the resilience

513 index, but they are also the most uncertain and difficult to be computed. The first step for its

514 evaluation consists in the definition of a performance index. In general, the community performance

515 indices are function of time t and of other parameters that depend on the type of a community

516 considered. Numerous models have been listed in Cimellaro et al. (2010a) to describe the restoration

517 function. They can be either empirical or analytical depending on the type of analysis and data.

518 Empirical recovery models are based on test or field data interpretation and engineering judgment.

519 They can be built using Monte Carlo simulations based on data from past events or maximum

520 likelihood method. Since the complexity of the problem changes case by case, no specific model is

521 presented in this part.

522 Analytical recovery models are developed using response data from numerical simulations

523 (nonlinear time history analysis, response spectral analysis, etc) of system models. Few example of

524 analytical recovery models (e.g. discrete event simulation models, metamodels, etc.) for critical

525 facilities like hospitals can be found in Cimellaro et al., (2011, 2014d).

526 Step by step procedure for resilience evaluation

527 A schematic step-by-step procedure of the methodology described in is the following:

528 (8) Define the extreme event scenarios (e.g. PSHA and ground motion selection);

21
529 (9) Definition, calibration and validation of the hybrid model of the community;

530 (10) Run the analysis and evaluate the response of the model;

531 (11) Evaluate the performance metrics (e.g. losses, restoration time, performance index, resilience

532 index) for different scenarios and compare with different performance levels;

533 (12) Recognize remedial mitigation actions (e.g. advanced technologies such as base isolation,

534 passive dampers, etc.) and/or resilience actions (e.g. resourcefulness, redundancy, etc.);

535 The proposed design approach has analogies with the feedback loop taken from control theory and it

536 can be applied both to communities and single structures (e.g. hospital, city hall, etc).

537 Several applications of this approach can be found in literature to hospitals (Cimellaro et al., 2011),

538 natural gas distribution networks (Cimellaro et al., 2014a), water distribution network (Cimellaro et

539 al., 2015), but they are not reported in this paper due to the lack of space.

540

541 CONCLUDING REMARKS

542 After the recent disasters, the general public became aware that Resilience is the solution to face

543 natural and manmade threats. The paper presents a holistic framework to evaluate the resiliency of a

544 community at various geographical and temporal scales and identifies the gaps in the definitions and

545 quantification of resilience at the community level. The suggested framework summarized with the

546 acronyms P.E.O.P.L.E.S. is combining different dimensions of resilience together using a layered

547 approach. The main contribution in the field is the development of a community hybrid model

548 combining network models to model the physical infrastructures (electric power, water, gas etc.) with

549 agent-based models to model the socio technical networks (e.g. Emergency medical technicians, fire

550 brigade, police, etc). Furthermore, special attention is given to the human behavior and its emotions

551 which plays a key role during the emergency and they have been modeled using the extended version

552 of Belief-Desire-Intention (BDI) modeling framework proposed by Zoumpanaki in 2010.

22
553 Each dimension of the framework is made of components and sub-components with their respective

554 performance indicators. These indicators can be grouped according to their difficulty in evaluating

555 them, their complexity as well as spatial and temporal scales. Some of them might be valid on a

556 multi hazard approach, while others might be valid only for certain type of hazard. In the paper are

557 shown some examples of indicators, while is made reference to the several applications already

558 available in literature of the PEOPLES methodology because the framework has been the result of a

559 NIST project developed in 2009.

560 In the long term, the proposed framework can be used as decision support software by decision

561 makers and by planners/engineers to help implementing Resilience-Based Design (RBD) techniques.

562 The goal is to make individual structures and communities safe and resilient with both advanced

563 technologies (e.g. base isolation, passive dampers etc.) and resilience actions that allow each system

564 to recover its functionality in a short time by selecting the optimal restoration strategy and enhancing

565 the community resilience index by comparing it with the resilience levels targets.

566

567 ACKNOWLEDGEMENTS

568 The research leading to these results has received funding from the European Research Council

569 under the Grant Agreement n° ERC_IDEAL RESCUE_637842 of the project IDEAL RESCUE-

570 Integrated Design and Control of Sustainable Communities during Emergencies.

571 The research leading to these results has received also funding from the European Community’s

572 Seventh Framework Programme - Marie Curie International Outgoing Fellowship (IOF) Actions-

573 FP7/2007-2013 under the Grant Agreement n°PIOF-GA-2012-329871 of the project IRUSAT—

574 Improving Resilience of Urban Societies through Advanced Technologies.

23
575 References

576 Adger, W. N. (2000) Social and ecological resilience: are they related? Progress in Human
577 Geography 24(3): 347-64

578 Allenby, B., and Fink, J. (2005). "Toward inherently secure and resilient societies." Science,

579 309(5737), 1034-1036.

580 Arcidiacono, V., G. P. Cimellaro, et al. (2012) Road network resilience assignment methodology. In:
581 6th International Conference on Bridge Maintenance, Safety and Management
582 (IABMAS2012). F. B. D. Frangopol. CRC Press Taylor & Francis Group, Stresa, Lake
583 Maggiore, July 8-12, 2012.

584 Berche, B., von Ferber, C., Holovatch, T., and Holovatch, Y. (2009). "Resilience of public transport

585 networks against attacks." European Physical Journal B, 71(1), 125-137.

586 Bonowitz, D. (2009). "The Dilemma of existing buildings: Private property, public risks." San
587 Francisco Planning Urban Research Association (SPUR), San Francisco, California.

588 Brown, D. and J.Kulig (1996) The conceptof resiliency: Theoretical lessons from community
589 research. Health and Canadian Society 4: 29-52.

590 Bruneau, M. and A. M. Reinhorn (2007) Exploring the Concept of Seismic Resilience for Acute
591 Care Facilities. Earthquake Spectra 23(1): 41-62.

592 Bruneau, M., Chang, S., Eguchi, R., Lee, G., O’Rourke, T., Reinhorn, A. M., Shinozuka, M.,

593 Tierney, K., Wallace, W., and Winterfelt, D. v. (2003). "A framework to Quantitatively

594 Assess and Enhance the Seismic Resilience of Communities." Earthquake Spectra, 19(4),

595 733-752.

596 Cagnan, Z., Davidson, R. A., and Guikema, S. D. (2006). "Post-Earthquake Restoration Planning for

597 Los Angeles Electric Power." Earthquake Spectra, 22(3), 589-608.

598 Chang, S. and M. Shinozuka (2004) Measuring Improvements in the Disaster Resilience of
599 Communities. Earthquake Spectra 20(3): 739-55.

24
600 Chang, S. E., McDaniels, T., Fox, J., Dhariwal, R., and Longstaff, H. (2014). "Toward Disaster-
601 Resilient Cities: Characterizing Resilience of Infrastructure Systems with Expert Judgments."
602 Risk Analysis, 34(3), 416-434.

603 Cimellaro, G. P., Reinhorn, A. M., and Bruneau, M. (2010a). "Framework for analytical
604 quantification of disaster resilience." Engineering Structures, 32(11), 3639–3649.

605 Cimellaro, G. P., Reinhorn, A. M., and Bruneau, M. (2010b). "Seismic resilience of a hospital
606 system." Structure and Infrastructure Engineering, 6(1-2), 127-144.

607 Cimellaro, G. P., A. M. Reinhorn, et al. (2011) Performance-based metamodel for health care
608 facilities. Earthquake Engineering & Structural Dynamics 40 1197–217.

609 Cimellaro, G. P., Villa, O., and Bruneau, M. (2014a). "Resilience-Based Design of Natural gas
610 distribution networks." Journal of Infrastructure Systems, ASCE, 10.1061/(ASCE)IS.1943-
611 555X.0000204.

612 Cimellaro, G. P., Scura, G., Renschler, C., Reinhorn, A. M., and Kim, H. (2014b). "Rapid building
613 damage assessment system using mobile phone technology" Earthquake Engineering and
614 Engineering Vibration, 13(3), 519-533.

615 Cimellaro, G. P., Solari, D., and Bruneau, M. (2014c). "Physical infrastructure Interdependency and
616 regional resilience index after the 2011 Tohoku earthquake in Japan." Earthquake
617 Engineering & Structural Dynamics, 43(12), 1763-1784.

618 Cimellaro, G. P., Nagarajaiah, S., and Kunnath, S. (2014d). "Computational Methods, Seismic
619 Protection, Hybrid Testing and Resilience in Earthquake Engineering - A tribute to the
620 research contribution of Prof. Andrei Reinhorn." Geotechnical, Geological and Earthquake
621 Engineering Series, C. Springer International Publishing AG, ed., Springer, Netherland, 250.

622 Cimellaro, G. P., Tinebra, A., Renschler, C., and Fragiadakis, M. (2015). "Resilience-based design of
623 an urban water distribution system." Journal of Structural Engineering, ASCE, in press.

624 Comfort, L. K. (1999) Shared Risk: Complex Systems in Seismic Response. Pergamon, New York.

625 Crooks, A. T., and Heppenstall, A. J. (2012). "Introduction to agent-based modeling." Agent-based
626 models of geographical systems, C. Heppenstall, See, & Batty, ed., Dordrecht: Springer, 85-
627 105.

25
628 Cutter, S. L. and J. T. Mitchell (2000) Revealing the vulnerability of people and places: A case study
629 of Georgetown County, South Carolina. Annals of American Geographers 90(4): 713-37.

630 Cutter, S. (1996) Vulnerability to Environmental Hazards. Progress in Human Geography 20(4):
631 529-39.

632 Department of Homeland Security (DHS) (2008). DHS Risk Lexicon, Risk Steering Committee,
633 U.S. Department of Homeland Security, Washington, D.C.

634 Díaz-Delgado, R., F. Lloret, et al. (2002) Satellite Evidence of Decreasing Resilience in
635 Mediterranean Plant Communities After Recurrent Wildfires. Ecology 83(8): 2293-303.

636 Dorbritz, R. (2011). "Assessing the resilience of transportation systems in case of large-scale
637 disastrous events." The 8th International Conference on Environmental Engineering, Vilnius,
638 Lithuania, 19-20 May 2011, 1070-1076.

639 Drabek, T. E. and D. A. McEntire (2002) Emergent Phenomena and Multiorganizational

640 Coordination in Disasters: Lessons from the Research Literature. International Journal of
641 Mass Emergencies and Disasters 20: 197-224.

642 Ettema, D., de Jong, K., Timmermans, H., and Bakema, A. (2007). "PUMA: Multi-agent modelling
643 of urban systems." 45th Congress of the European Regional Science Association, Vrije
644 Universiteit, Amsterdam.

645 Fantini, P., Cimellaro, G. P., and Mahin, S. A. (2014). "Measuring lifeline emergency response using
646 temporal network models " Proceedings of the 15th U.S.-Japan Workshop on the
647 Improvement of Structural Engineering and Resiliency., ATC- Applied Technology Council,
648 December 3-5, 2014, Island of Hawaii.

649 Fiksel, J. (2003) Designing resilient, sustainable systems. Environmental Science and Technology
650 37(23): 5330-39.

651 Gunderson, L. H., C. S. Holling, et al. (2002) Resilience of large-scale resource systems,
652 Washington, D.C.

653 Haimes, Y. Y., Horowitz, B. M., Lambert, J. H., Santos, J. R., Lian, C., and Crowther, K. G. (2005).

654 "Inoperability input-output model for interdependent infrastructure sectors. I: Theory and

655 methodology." Journal of Infrastructure Systems, 11(2), 67-79.

26
656 Haimes, Y. Y. (2009) On the definition of resilience in systems. Risk Analysis 29(4): 498-501.

657 Haimes, Y. Y., N.C.Matalas, et al. (1998) Reducing vulnerability of water supply systems to attack.
658 Reducing vulnerability of water supply systems to attack 4(4): 164-77.

659 Handmer, J. and S. Dovers (1996) A typology of resilience: Rethinking institutions for sustainable
660 development. Organization & Environment 9: 482-511.

661 Harrald, J. (2006) Agility and Discipline: Critical Success Factors for Disaster Response. Annals of
662 the American Academy of Political and Social Science 604: 256–72.

663 Heaslip, K., Louisell, W. C., Collura, J., and Serulle, N. U. (2010). "A sketch level method for
664 assessing transportation network resiliency to natural disasters and man-made events." The
665 89th Annual Meeting of the Transportation Research Board, Washington, D.C., U.S.A., 10-
666 14 January 2010.
667 Holling, C. S. (1973) Resilience and Stability of Ecological Systems. Annual Review of Ecology and
668 Systematics 4: 1-23.

669 Hollnagel, E. (2006) Resilience - the challenge of the unstable, Hampshire.

670 Holmgren, A. (2007) A framework for vulnerability assessment of electric power systems., New
671 York, NY.

672 Horne, J. F. and J. E. Orr (1998) Assessing behaviors that create resilient organizations. Employment
673 Relations Today 24(4): 29-39.

674 Javanbarg, M. B., Scawthorn, C., Kiyono, J., and Shahbodaghkhan, B. (2012). "Fuzzy AHP-based

675 multicriteria decision making systems using particle swarm optimization." Expert Systems

676 with Applications, 39(1), 960-966.

677 Klein, R. J. T., R. J. Nicholls, et al. (2003) Resilience to natural hazards: How useful is this concept?
678 . Global Environmental Change Part B: Environmental Hazards 5(1-2): 35-45.

679 Manyena, S. B. (2006) The concept of resilience revisited. Disasters 30: 434.

680 Manzo, L. and D. Perkins (2006) Finding common ground: The importance of place attachment to
681 community participation and planning. Journal of Planning Literature 20: 335-50.

27
682 Miles, S., and Chang, S. (2006). "Modeling Community Recovery from Earthquakes." Earthquake

683 Spectra, 22(2), 439-458.

684 Miles, S. B. and S. E. Chang (2011) ResilUS: A community based disaster resilience model. Journal
685 of Cartography and GIS (CAGIS) 38(1): 36-51.

686 Mileti, D. (1999). Disasters by Design: A Reassessment of Natural Hazards in the United States,

687 Joseph Henry Press (May 18, 1999) Washington D.C.

688 Miller-Hooks, E., Zhang, X., and Faturechi, R. (2012). "Measuring and maximizing resilience of

689 freight transportation networks." Computers & Operations Research, 39(7), 1633-1643.

690 Norris, F. H., S. P. Stevens, et al. (2008) Community Resilience as a Metaphor, Theory, Set of
691 Capacities, and Strategy for Disaster Readiness. Am J Community Psychol 41: 127-50.

692 Olofsson, P., L. Eklundh, et al. (2007) Estimating net primary production for Scandinavian forests
693 using data from Terra/MODIS. Space Res 39: 125-30.

694 Ouyang, M., and Duenas-Osorio, L. (2011). "An approach to design interface topologies across
695 interdependent urban infrastructure systems." Reliability Engineering & System Safety,
696 96(11), 1462-1473.

697 Pederson, P., D. Dudenhoeffer, S. Hartley and M. Permann. (2006). Critical Infrastructure Modeling:
698 A Survey of US and International research, Research Report prepared at the Idaho National
699 Laboratory, Idaho Falls, Idaho 83415 INL/EXT-06-11464.

700 Pettorelli, N., J. Vik, et al. (2005) Using the satellite-derived NDVI to assess ecological responses to
701 environmental change. TRENDS in Ecology and Evolution 20(9): 503-10.

702 Queiroz, C., Garg, S. K., and Tari, Z. (2013). "A probabilistic model for quantifying the resilience of

703 networked systems." Ibm Journal of Research and Development, 57(5).

704 Renschler, C., Frazier, A., Arendt, L., Cimellaro, G. P., Reinhorn, A. M., and Bruneau, M. (2010).
705 "Framework for Defining and Measuring Resilience at the Community Scale: The PEOPLES
706 Resilience Framework." MCEER Technical Report –MCEER-10-006, pp. 91, University at
707 Buffalo (SUNY), The State University of New York, Buffalo, New York.

708 Renschler, C., Reinhorn, A. M., Arendt, L., and Cimellaro, G. P. "The PEOPLES Resilience
709 Framework: A conceptual approach to quantify community resilience." Proceedings of the
28
710 3rd International Conference on Computational Methods in Structural Dynamics and
711 Earthquake Engineering (COMPDYN 2011), Corfu`, Greece, May 26-28, 2011.
712 RICSA (2010) RISCA Poverty Project. URL:
713 http://web.uct.ac.za/depts/ricsa/projects/publicli/poverty.

714 Rose, A. (2004). "Economic principles, issues, and research priorities in hazard loss estimation."

715 Modeling Spatial and Economic Impacts of Disasters, S. C. Yasuhide Okuyama, ed., Springer

716 Berlin Heidelberg, 13-36.

717 Rose, A. and S. Y. Liao (2005) Modeling regional economic resilience to disasters: A computable
718 general equilibrium analysis of water service disruptions. Journal of Regional Science 45(1):
719 75-112.

720 SEAOC. (1995). "Vision 2000 - A Framework for Performance Based Earthquake Engineering ",
721 Structural Engineers Association of California, Vision 2000 Committee, Sacramento,
722 California.
723 Simoniello, T., Lanfredi, M., Liberti, M., Coppola, R., and Macchiato, M. (2008). Estimation of
724 vegetation cover resilience from satellite time series. Hydrol. Earth Syst. Sci., 12, 1053-1064.

725 Simoniello, R. A. and E. L. Quarantelli (1985) Emergent Citizen Groups and Emergency
726 Management. Public Administration Review 45: 93-100.

727 Tamvakis, P., and Xenidis, Y. (2013). "Comparative Evaluation of Resilience Quantification
728 Methods for Infrastructure Systems." Selected Papers from the 26th Ipma (International
729 Project Management Association), World Congress, 74, 339-348.

730 Tierney, K. (1997) Business impacts of the northridge earthquake. Journal of Contingencies and
731 Crisis Management 5: 87-97.

732 Tierney, K. and M. Bruneau (2007) Conceptualizing and Measuring Resilience. TR News 250: 14-
733 17.

734 Tierney, K. (2009) Disaster Responce: Research Findings and their implications for Resilience
735 Measures. In. CARRI Research Report 6, Community & Regional Resilience Institute,
736 Colorado Boulder.

737 Vugrin, E. D., D. E. Warren, et al. (2010) A Framework for assessing the resilience of infrastructure
738 and economic systems, Springer, Berlin, Germany.

29
739 Weick, K. E. (1995). Sensemaking in Organizations (Foundations for Organizational Science),
740 SAGE Publications, Inc., Thousand Oaks, California.

741 Weick, K. E., K. Sutcliffe, et al. (2005) Organizing and the Process of Sensemaking. Organization
742 Science 16: 409-21.

743 Wildavsky, A. B. (1991). Searching for Safety, Oxford Transaction Publisher, New Brunswick.

744 Woods, D. D. (2006) Essential characteristics of resilience, Aldershot, UK.

745 Zoumpoulaki, A., Avradinis, N., and Vosinakis, S. (2010). "Multi-Agent Simulation Framework for

746 Emergency Evacuations Incorporating Personality and Emotions." Artificial Intelligence:

747 Theories, Models and Applications, Springer Berlin Heidelberg, 423-428.

748

30
749 Table -1 Literature review about resilience definitions
Author Definition
Holling Ecological systems resilience is a measure of the persistence of systems and of their ability to absorb
(1973) change and disturbance and still maintain the same relationships between populations or state variables.
Wildavsky Resilience is the capacity to cope with unanticipated dangers after they have become manifest, learning
(1991) to bounce back.
Horne and Orr Resilience is the ability of a system to withstand stresses of ‘environmental loading’... [it is] a
(1998) fundamental quality found in individuals, groups, organizations, and systems as a whole.
Haimes et al. Resilience is the ability of system to return to its optimal condition in a short period of time. Considering
(1998) resilience one of four strategies for hardening a system, together with security, redundancy and
robustness.
Mileti Local resiliency with regard to disasters means that a locale is able to withstand an extreme natural event
(1999) without suffering devastating losses, damage, diminished productivity, or quality of life and without a
large amount of assistance from outside the community.
Comfort Resilience is the capacity to adapt existing resources and skills to new situations and operating
(1999) conditions.
Adger (2000) Social resilience is the ability of groups or communities to cope with external stresses and disturbances
as a result of social, political, and environmental change.
Gunderson et Engineering resilience […] is the speed of return to the steady state following a perturbation […]
al. (2002) ecological resilience […] is measured by the magnitude of disturbance that can be absorbed before the
system is restructured….
Fiksel (2003) Resilience is the essence of sustainability […] the ability to resist disorder.
Bruneau et al. Resilience is defined in terms of three stages: the ability of a system to reduce the probability of an
(2003) adverse event, to absorb the shock if the adverse event occurs, and to quickly re -establish normal
operating conditions. So resilience thus encompasses the four characteristics of robustness, redundancy,
resourcefulness, and rapidity. Are considered four types of resilience: technical; organizational;
economic; and social.
Allenby and Resiliency is defined as the capability of a system to maintain its functions and structure in the face of
Fink (2005) internal and external change and to degrade gracefully when it must.
Rose and Liao Regional economic resilience is the inherent ability and adaptive response that enables firms and regions
(2005) to avoid maximu m potential losses.
Hollnagel Resilience is defined as the intrinsic ability of an organization (system) to maintain or regain a
(2006) dynamically stable state, which allows it to continue operations after a major mishap and/or in the
presence of a continuous stress.
Manyena Evaluating all the possible definitions provided from the 90’s to nowadays, resilience could be viewed
(2006) as the intrinsic capacity of a system, community or society predisposed to a shock or stress to adapt and
survive by changing its non essential attributes and rebuilding itself.
Woods Evaluating all the possible definitions provided from the 90’s to nowadays, resilience could be viewed
(2006) as the intrinsic capacity of a system, community or society predisposed to a shock or stress to adapt and
survive by changing its non essential attributes and rebuilding itself.
Holmgren Resilience is the ability of the system to return to a stable condition after a disruption. Distinguishing
(2007) robustness and resilience, using robustness to imply that the system will remain (nearly) unchanged even
in the face of disruption.
Tierney and Resilience is both the inherent strength and ability to be flexible and adaptable after environmental
Bruneau shocks and disruptive events.
(2007)
DHS Resilience is the ability of systems, infrastructures, government, business, and citizenry to resist, absorb,
(2008) recover from, or adapt to an adverse occurrence that may cause harm, destruction, or loss of national
significance.
Haimes Resilience is defined as the ability of the system to withstand a major disruption within acceptable
(2009) degradation parameters and to recover within an acceptable time and composite costs and risk.
Vugrin et al. Given the occurrence of a particular disruptive event (or set of events), the resilience of a system to that
(2010) event (or events) is the ability to efficiently reduce both the magnitude and duration of the deviation
from targeted system performance levels.
750

751

31
752 Table -2 Complete list of components and subcomponents of PEOPLES framework
1) POPULATION AND DEMOGRAPHICS
a) Distribution/Density b) Composition c) Socio-Economic Status
i) Urban i) Age i) Educational Attainment iv) Home Ownership
ii) Suburban ii) Gender ii) Income v) Housing Vacancies
iii) Rural iii) Immigrant Status iii) Poverty vi) Occupation
iv) Wildland iv) Race/Ethnicity
2) ENVIRONMENTAL/ECOSYSTEM
a) Water Quality/Quantity b) Air Quality c) Soil Quality d) Biodiversity
e) Biomass (Vegetation) f) Other Natural Resources
3) ORGANIZED GOVERNMENTAL SERVICES
a) Executive/Administrative b) Judicial c) Legal/Security
i) Emergency Response and ii) Health and Hygiene
Rescue 4) PHYSICAL INFRASTRUCTURE
a) Facilities b) Lifelines
i) Residential i) Communications
(1) Housing Units (1) Internet (2) Phones (3) TV (4) Radio (5) Postal
(2) Shelters ii) Health Care
ii) Commercial (1) Acute Care (2) Long-Term Acute Care (4) Psychiatric
(1) Distribution Facilities (3) Manufacturing Facilities (3) Primary Care (5) Specialty
(2) Hotels - Accommodations (4) Office Buildings iii) Food Supply
iii) Cultural iv) Utilities
(1) Entertainment Venues (4) Schools (1) Electrical (2) Fuel/Gas/Energy (3) Waste
(2) Museums (5) Sports/Recreation Venues v)(4) Water
Transportation
(3) Religious Institutions (1) Aviation (2) Bridges (3) Highways
(4) Railways (5) Transit (6) Vehicles (7) Waterways
5) LIFESTYLE AND COMMUNITY COMPETENCE
a) Collective Action and Decision Making b) Collective Efficacy and c) Quality of Life
i) Conflict Resolution ii) Self-Organization Empowerment
6) ECONOMIC DEVELOPMENT
a) Financial Services b) Industry – Employment - c) Industry – Production
Services
i) Asset Base of Financial i) Agriculture x) Number of Corporate i) Food Supply
Institutions Headquarters
ii) Checking Account Balances ii) Construction xi) Other Business Services ii) Manufacturing
(Personal and Commercial)
iii) Consumer Price Index iii) Education and Health xii) Professional and Business
Services Services
iv) Insurance iv) Finance, Insurance and (1) Employment Services
Real Estate
v) Number and Average Amount of v) Fortune 1000 (a) Flexibilities
Loans
vi) Number of Bank and Credit Union vi) Fortune 500 (b) Opportunities
Members
vii) Number of Banks and Credit vii) Information, Professional (c) Placement
Unions Business, Other
viii) Savings Account Balances viii) Leisure and Hospitality (2) Transport and Utilities
(Personal and Commercial)
ix) Stock Market ix) Manufacturing (3) Wholesale and Retail
7) SOCIAL/CULTURAL CAPITAL
a) Child and Elderly Services b) Commercial Centers c) Community Participation d) Cultural and Heritage
Services
e) Education Services f) Non-Profit Organizations g) Place Attachment
32
Q(t) %
100

L0

50
Res
Q0
0
t0E1 t0E1+TRE1 t0E2 t0E2+TRE2
Time
TLC

Figure -1 Resilience (adapted from Cimellaro et al., 2010a)

33
 REAL & SIMULATED
DATA
CALIBRATION &

HYBRID MODEL
vALIDATION

Water Pump Water

Tower House Gas
Physical Plant
B.1 B.2
lifelines
(Network Pump Power
Models) House Gas
B.1 Plant
B.2
Substation

Gas EMT
B.1
Plant
Non- B.2
Physical
lifelines
(ABM) B.2 Gas
Fire
Plant
B.1 brigade

Scenario
analysis

Performance
Robustness measures
Resilience actions

Rapidity Population/ Environmental/

Demographic Ecosystems

Resourcefulness Social-Cultural Organized

Capital governmental
services

Redundancy
Economic Physical
Development Infrastructures Performance levels
Resource & Opp. Lifestyle and Community
Evaluation Competence
Intensity measures (IM)

Gaps & Priority

No Community
Identification Resilience index
Performance
Objective

Recovery Time (T RE)

Yes

Figure -2 Methodology for Resilience-based design (RBD) based on control (feedback loop)

approach and hybrid layered model

34
 REAL & SIMULATED
DATA
CALIBRATION &
vALIDATION
HYBRID MODEL
INTERDEPENDENCY
Water Pump Water
Gas D(water-power)
Tower House
Physical Plant
B.1 B.2
lifelines
(Network Pump Power D(power-EMT)
Models) House Gas
B.1 B.2 Plant
Substation
D(EMT-Fire brigade)
Gas EMT
B.1
Non- Plant
B.2
Physical
lifelines
B.2 Gas
(ABM) Fire
B.1 Plant
brigade

Extended Belief, Desire, Intention BDI architecture

(Zoumpoulaki et al., 2010)

Belief Personality
PERCEP 1.Knowledgeof 1.Leader/Follower
TION the world 2. Altruism APPRAISAL
2. Emotioanl 3. OCEAN (Orthony,
State 2003)
Simulation 3. Group Status
Environ. 4. Physical Emotion (W)
(agents) Status
1.Emotional State
5. Desire Status

DECISION MAKING
ACTION Intentions P(t+h)=SP(t)+CM(t+h)W(t+h)

Desires
1.Evacuation
2. Group Related
3. Information
exchange

Figure 3 Methodology to model the interdependency and the human behavior within the community
hybrid model

35
Figure 4 Sketch of a typical IM matrix.

36
 Recovery Time
Short term Long term

Intensity measures (IM)

Midterm
(Emergency) (Reconstruction)
Fully
Basic
Operational Unacceptable Unacceptable
Objective
Functionality Performance Levels

(Q1)

Operational Essential Basic

Unacceptable
(Q2) Objective Objective

Partially
Critical Essential
Operational Basic Objective
Objective Objective
(Q3)
Near not Performance
Critical Objective
Operational Not feasible Basic Objective
Objective
(Q4)
Recovery Time (TRE)

Figure -5 Tridimensional Resilience Performance levels matrix for structures, communities, systems
etc.

37