Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 84 (2014) 116 121

2014ISSST, 2014 International Symposium on Safety Science and Technology

Vulnerability and resilience under effects of tsunamis: case of

industrial plants
Ahmed MEBARKIa, , Adrien WILLOTb, Sandra JEREZc, Mathieu REIMERINGERb,
Gatan PRODHOMMEb
b

a
University Paris-Est, Lab. Modlisation et Simulation Multi Echelle, UMR 8208 CNRS, 5 Bd Descartes, 77454, Marne-La-Valle, France
INERIS, Institut National de lEnvironnement Industriel et des Risques, Parc Technologique ALATA,BP 2 - 60550 Verneuil-en-Halatte, France
c
EscuelaColombiana de Ingeniera, Av. 13 No.205-59, Bogot, Colombia

Abstract
The damages caused by natural or technological hazards to physical systems may generate disastrous situations. However, the
damaged systems can adapt, recover and be restored or strengthened in given acceptable recovery period for resilient systems.
The metrics to evaluate objectively the resilience are still to be developed for several systems such as industrial plants, for
instance. In fact, the resilience is intimately depending on both expected damages level and adaptation measures. The mechanical
vulnerability is therefore an influent governing parameter. The present paper focuses on the case of industrial tanks and their
mechanical vulnerability under tsunamis effect. The resilience of coastal industrial plants is implicitly investigated by running
sensitivity analysis:
Since tsunamis may cause structural failure of the tanks. Several modes of damages are considered, i.e. uplift by buoyancy,
rigid overturning, rupture of anchorages or rigid sliding effect, and excessive stress by bending or buckling.
For a wide range of tanks dimensions and filling ratios, the vulnerability and fragility curves are developed and discussed.

by Elsevier
Ltd.
is an Ltd.
open access article under the CC BY-NC-ND license
2014
2014 Published
The Authors.
Published
byThis
Elsevier
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of scientific committee of Beijing Institute of Technology.
Peer-review under responsibility of scientific committee of Beijing Institute of Technology
Keywords: Tsunamis; Resilience; Vulnerability; Fragility; Hazard; Risk; Industrial plants; Metal tanks

Corresponding author (A. Mebarki, Professor). Tel.: +33.1.60 95 77 87; fax: +33.1.60 95 77 99
E-mail address:Ahmed.Mebarki@u-pem.fr

1877-7058 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of scientific committee of Beijing Institute of Technology
doi:10.1016/j.proeng.2014.10.520

Ahmed Mebarki et al. / Procedia Engineering 84 (2014) 116 121

1. Introduction
The concept of resilience is well adapted to address the case of industrial plants and their vulnerability under the
effects of natural hazards such as tsunamis, or industrial hazards such as explosions or fires. It investigates the
capacity of the physical system and its related socio-economic aspects to stand the disturbances and damages, to
adapt, recover and return into accepted conditions within a given recovery period, [1-16]. The resilience analysis of
industrial plants still requires the development of rigorous and objective metrics [6], i.e:
- Identification and evaluation of given utility functions,
- Analysis and assessment of these functions losses, such as mechanical damages, under the effect of
expected hazards(Fig.1),
- Definition of these functions threshold values, recovery periods for post-disaster stage, below which the
system is considered as unable to become resilient.

(a)- General layout

(b)- Applications and case studies

Fig. 1. Layout for integrated framework: Hazard Vulnerability Risk Resilience[6].

117

118

Ahmed Mebarki et al. / Procedia Engineering 84 (2014) 116 121

Obviously, the resilience depends intimately on the damage level caused by the potential hazards. For excessive
values of the damage, the system will be unable to recover. It is therefore of great importance to be able to evaluate
accurately the values of the expected or observed damages and to be able to develop objective and measurable
indicators for the systems resilience, [6]:
FR  t 

Tref

F (t ). 1- D  .> (1  ) r (t ) F r (t )) @.dv

(1)

where: FR(.)= Resilience function of the system during the recovery period or reference period Tref; V= volume of
the system; Tref= recovery or reference period; t= instant ranging between the hazard occurrence and Tref; D(.)=
damage value or vulnerability ranging within [0..1]; r(.)= recovery and evolution function, and Fr(.t)= resilience
capability and availability function which depends on availability of resources (internal or external) and the capacity
to react adequately, such as past experience, knowledge and preparedness.
Relevant resilience analysis requires that the vulnerability and fragility functions are accurately modeled. The
present paper focuses on several parts of the integrated resilience framework for different mechanical systems:
- Mechanical vulnerability of industrial plants under the effect of tsunamis,
- Sensitivity analysis of the resilience indicator, i.e. utility functions depending on the intensity of damage
and vulnerability, their threshold values and recovery time.
2. Vulnerability modeling
Under the hydrodynamic effects of the tsunamis, the structural vulnerability is investigated against several
mechanical phenomena, (Fig.2):
- Tanks uplift due to buoyancy,
- Debris impacts, perforation or collapse of tanks or sections of tank with the ensuing escape of stored
products (oil, other liquids and gases),
- Excessive bending or shear, as well as rigid sliding and overturning,
- Circumferential as well as longitudinal buckling,
- Rupture of the pipes connected to the tanks and the metal roofs.
The vulnerability of the tanks is investigated for given levels of the tsunamis effects, i.e. its height and velocity,
against the height of the liquid stored in the tanks, in order to consider various exploitation conditions. However,
due to unknown conditions about the contact of the tank with its concrete support on the ground, this contact is
described by a friction coefficient assumed to have a constant value all over the concrete support. This tank-fullness
ratio is considered as a random variable following a theoretical Gamma distribution, [6]. The tank may fail when it
cannot resist any of uplift effect, overturning moment, sliding effect as well as circumferential buckling effects. The
failure event, Ef, is therefore described as being a serial combination of probabilistic elementary failures, [6]:

Ef

Ne

*E

f ,i

and

Pf

> @

P Ef

(2)

i 1

where: Ef= failure event of the system; Pf= failure probability or vulnerability; Ef,i= i-th failure event among the total
number Ne of failure events: for instance i=1 for Buoyancy, i=2 for perforation by debris impacts.
3. Fragility curves and failure risks
The failure risk of the metal tanks is defined as, [6-9]:

Pf

( R  S d 0)

(3)

119

Ahmed Mebarki et al. / Procedia Engineering 84 (2014) 116 121

S0

D1: SLIDING
D2: OVERTURNING

D3: BUOYANCY
D4: SHELL BUCKLING

D5: DEBRIS IMPACT

(a)

D.H

2.R

(b)- Tanks: 5 sizes {T1: H=8m, R=5.57m}; { T2: H=10m, R=14m}; { T3: H=19m, R=10m}; { T4: H=30m, R=20m}; { T4: H=30m, R=40m}
Fig. 2. Cylindrical metal tanks for storing oil in industrial plants: (a)- Failure modes of tanks; (b)- Various dimensions[6].

phydraulic

Hw 

Vw2
2.g

(4)

where : R= structural capacity of the tank against any of the elementary mechanical events, i.e. uplift effect,
overturning moment, sliding effect as well as circumferential buckling effects); S= mechanical effect of the tsunami,
i.e. the effect of the hydraulic pressure phydraulic; phydraulic= hydraulic pressure resulting from the height (Hw) and the
velocity (Vw) of flow, [6-7].
Monte Carlo simulations are used to calculate the risk of failure, [6-9]. The fragility curves express the
probability of failure vs. vs. tsunami height (Hw). The risk analysis of the entire industrial plant, erected in a zone
prone to tsunamis, relies on the use of these fragility curves specific to each type of tank, (Fig. 3).
4. Conclusions
The concept of resilience is considered and discussed in order to address the case of industrial plants and their
physical vulnerability under the effects of tsunamis hazards. Coastal industrial plants containing metal tanks and
their fragility curves are investigated.
Simplified model valid at the shoreline (at normal sea level) and inland inasmuch as the terrain is flat have
already been issued and the expected height and velocity of the tsunami flow are considered as known, [7].
The tsunami may cause mechanical failure of tanks by rigid sliding, buoyancy uplift, buckling or rigid
overturning. New fragility curves vs. tsunami height are developed for typical tanks ranging from 8 m to 30 m high
and from 10 to 80 m in diameter, used for oil storage. Two kinds of tanks are reported in the present paper.

120

Ahmed Mebarki et al. / Procedia Engineering 84 (2014) 116 121

(a) Multi failure modes= large size tanks

(b)Multi failure modes= small size tanks[6]

Fig 3. Cylindrical metal tanks for storing oil in industrial plants.

Ahmed Mebarki et al. / Procedia Engineering 84 (2014) 116 121

The results show that sliding has more severe effects than buckling, buoyancy or overturning in the case of small
tanks, even if they are not empty. Lateral protective barrier should be considered even for large tanks since their
resistance to sliding is also very weak. Otherwise, these tanks could slide and the pipes connected to them could
break even if the tsunami is less than 3 meters high.
If adequate protections are considered against sliding, they can withstand tsunamis of almost 10 m before
buckling and 15 m before they are damaged by buoyancy or overturning effects.
This sensitivity analysis is helpful for risk managers and their preparedness either to face potential hazards such
as tsunamis or to design protective systems such as dikes, early warning and alert devices.
Acknowledgements
Several research projects have been helpful for collecting results and methodological developments: research
projects with partial financial support by Agence Nationale de la Recherche (VULCAIN: ANR-PGCU 2007;
INTERNATECH: ANR-Flash Japan 2011), the Chinese-French program PHC-Xu Guangqi (27939XK, 2012),
PHC-CaiYuanpei (28020PB, 2012-2014) and PHC-CaiYuanpei (Havu-Risk: Chemical industrial plants & domino
effect: hazards, vulnerability, risks & sustainability 32114TE, 2014-2016).
References
[1] Barker K., Ramirez-Marquez J.E., Rocco C.M. (2012) Resilience-based network component important measures. Reliability Engineering and
System Safety 117:89-97
[2] Barroca B., Serre D. (2013) Behind the barriers : a resilience conceptual model. SAPIENS.http://sapiens.revues.org/1529
[3] Cimellaro G.P., Reinhorn A.M., Bruneau M. (2010) Framework for analytical quantification of disaster resilience. Engineering Structures
32:36393649
[4] Francis R., Bekera B. (2010) A metric and framework for resilience analysis of engineered and infrastructure systems. Reliability Engineering
and System Safety 121:90-103
[5] Dinh L.T.T., Pasman H., Gao X., Mannan M.S. (2012) Resilience engineering of industrial processes: Principles and contributing factors.
Journal of Loss Prevention in the Process Industries 25:233-241
[6] Mebarki A., Barroca B. (2014) Resilience and vulnerability analysis for restoration after tsunamis and floods: the case of dwellings and
industrial plants. In Vicente Santiago-Fandio,Yev A. Kontar and YoshiyukiKaneda. (eds) Post-Tsunami Hazard Reconstruction and
Restoration, Advances in Natural and Technological Hazards Research, Springer (in Press)
[7] Mebarki A., Jerez S., Matasic I., Prodhomme G., Reimeringer M., Pense V., Vu Q.A., Willot A. (2014) Domino effects and industrial risks :
integrated probabilistic framework Case of tsunamis effects. In Y.A. Kontar et al. (eds) Tsunami Events and Lessons
Learned :Environnemental and Societal Significance, Advances in Natural and Technological Hazards Research 35, doi:10.1007/978-94-0077269-4_15, Springer
[8] Mebarki A., Valencia N., Salagnac J.L., Barroca B. (2012) Flood hazards and masonry constructions: a probabilistic framework for damage,
risk and resilience at urban scale. Nat. Hazards Earth Syst. Sci. 12:17991809, doi:10.5194/nhess-12-1799-2012
[9] Mebarki A., Genatios C., Lafuente M. (2008) RisquesNaturels et Technologiques: Alas, Vulnrabilit et Fiabilit des Constructions
versune formulation probabilisteintgre, Presses Ponts et Chausses Ed., ISBN 978-2-85978-436-2, Paris
[10] Miller-Hooks E., Zhang X., Faturechi R. (2012) Measuring and maximizing resilience of freight transportation networks. Computers &
Operations Research 39 :1633-1643
[11] Ouedraogo K.A., Enjalbert S., Vanderhaegen F. (2013) How to learn from the resilience of Human-Machine Systems ?. Engineering
Applications of Artificial Intelligence 26:24-34
[12] Ouyang M., Dueas-Osorio L., Min X. (2012) A three-stage resilience analysis framework for urban infrastructure systems. Structural
Safety 36-37:23-31
[13] Pant R., Barker K., Ramirez-Marquez J.E., Rocco C.M. (2014) Stochastic measures of resilience and their application to container terminals.
Computers & Industrial Engineering 70:183-194
[14] Shirali G.H.A., Motamedzade M., Mohammadfam I., Ebrahimpour V., Moghimbeigi A. (2012) Challenges in building resilience engineering
(RE) and adaptive capacity: A field study in chemical plant. Process Safety and Environmental Protection 90:83-90
[15] Steen R., Aven T. (2011) A risk perspective suitable for resilience engineering. Safety Science 49: 292297
[16] Tisserand S. (2007) La Resilience. Que Sais-je ? PUF, Ed. Point Delta (in French)

121