160315 Thesis Pietro Crupi

POLITECNICO DI TORINO - DISEG G. P. CIMELLARO
Pietro Crupi
M.S. student in Civil Engineering
Politecnico di Torino
M.S. Thesis
The City College of New York
(September 2015 – February 2016)
Advisor: Prof. Gian Paolo Cimellaro
Host Advisor: Prof. Anil Kumar Agrawal
Modeling Interdependencies of
Critical Infrastructures After
Hurricane Sandy
PIETRO CRUPI
March 15th, 2016
Politecnico di Torino, Sala Consiglio di Facoltà
Pietro Crupi

Outline
 Hurricane Sandy:
 “Superstorm” overview
 Damage to critical infrastructures systems (c.i.s.) in the
metropolitan area of New York
 Initiatives to increase resilience:
 New York City Government report (2013) after Sandy
 Further categorization criterias
 Application of the Inoperability Input-Output Model:
 Supporting data
 Model formulation
 Discussion of results
 Dynamic extension
 Conclusions
PIETRO CRUPIPietro Crupi

Hurricane Sandy
 Last hurricane of the 2012 Atlantic season
 “Superstorm” characteristics:
 unusual westbound track
 interaction with other storms
 1800 km max diameter
 superposition of events
 1000 km impacted U.S. coastline
 New York Metropolitan Area:
 New York City and New Jersey
 Vulnerable region
 Concentration of c.i.s.

Damage to c.i.s.
 Direct: physical damage due to Sandy
 Indirect: disruption in a specific sector due to functional
problems occurred in other sectors

Initiatives
 Increase resilience of c.i.s. against future similar events
 Focus on those proposed for:
 utilities
 liquid fuel
 transportation
 More directly damaged
 Importance in the network:
 High dependency of other
sectors
 Cause of the majority of
indirect damage
 High number of facilities in
the area under analysis

Inoperability Input-Output Model
 Haimes and Jang adaptation of the original Leontief’s
input-output model for economy
 Inoperability: “inability of a system to perform its intended
function, assumed to be a continuous variable between 0 and 1”
 Demand-reduction Regional Model
 Model usage:
 Numerically definition of the interconnectivity among c.i.s.
 Quantify the effect of external perturbation on the network
 Identification of the priority initiatives
 Supporting economic data:
 BEA database of national input-output accounts
 RIMS II multipliers for regional decomposition

 “make” matrix [vij: “industry i produces commodity j”]
 “use” matrix [uij: “commodity i is consumed by industry j”]
 Interdependency matrix:
11 1 1 11 1 1 1
1 1 1
1 1 1
/ / /
ˆ / / /
/ / /
j n j j n n
i ij in i ij j in n
m mj mn m mj j mn n
u u u u x u x u x
u u u u x u x u xU U
u u u u x u x u x
   
   
   
     
   
   
      
11 1 1 11 1 1 1
1 1 1
1 1 1
/ / /
ˆ / / /
/ / /
j m j j m m
i i j im i ij j im m
n nj nm n nj j nm m
v v v v y v y v y
v v v v y v y v yV V
v v v v y v y v y
   
   
   
     
   
   
      
BEA database
 ˆ ˆ ˆ îj ik kj
k
A VU a v u  
j ij
i
y v
j ij
i
x u
/îj ij j
v v y
/îj ij j
u u x

 Multipliers tables for regional decomposition purchased for:
 New York City counties
 New Jersey counties in the metropolitan area
 Multipliers relative to the sector subjected to perturbation
RIMS II accounts
Code Industries liutilities litransp limining
11 Agriculture, forestry, fishing, and hunting 0 0 0
21 Mining 0.0006 0.00015 1.002567
22 Utilities 1.0058 0.007488 0.007567
23 Construction 0.0135 0.008325 0.010867
31G Manufacturing 0.0164 0.032513 0.020433
42 Wholesale trade 0.014 0.031438 0.017433
44RT Retail trade 0.0034 0.005925 0.001933
48TW Transportation and warehousing 0.0294 1.0778 0.009467
51 Information 0.0121 0.0184 0.0102
FIRE Finance, insurance, real estate, rental, and leasing 0.0709 0.1215 0.0564
PROF Professional and business services (includes waste management) 0.0514 0.044425 0.036367
6 Educational services, health care, and social assistance 0.0009 0.00085 0.0007
7 Arts, entertainment, recreation, accommodation, and food services 0.0093 0.006425 0.003967
81 Other services, except government 0.0102 0.010125 0.002833
G Government 0.008903 0.020372 0.000262
Pietro Crupi

 Regional interdependency matrix:
 Demand-reduction regional interdependency matrix:
 GDP proportion:
 Equation for the model:
 qR: inoperability of the regional network (output)
 c*R: external perturbation to the regional network (input)
* 1 *
ˆ
[( ( )) ( ( ))]ˆ ˆ
ˆ
R
jR R R R R R
ij ij R
i
x
A diag x A diag x a a
x

  
    
  
[min( , )] min( ,1)R R
ij i ij
A diag l A a l a  
Regional IIM
* 1 *
( )R R R
q I A c
 
0.1 (1/10)NYC NJ
US
GDP
GDP



 Inoperability ranking due to 10% functionality reduction of
utilities sector
Results
Code Industries q
R
[%]
21 Mining 0.50
48TW Transportation and warehousing 0.14
23 Construction 0.05
PROF Professional and business services 0.05
31G Manufacturing 0.04
42 Wholesale trade 0.03
FIRE Finance, insurance, real estate, rental, and leasing 0.02
7 Arts, entertainment, recreation, accommodation, and food services 0.02
51 Information 0.01
81 Other services, except government 0.01
G Government 0.01
44RT Retail trade 0.00
11 Agriculture, forestry, fishing, and hunting 0.00
6 Educational services, health care, and social assistance 0.00
* 1 *
( )R R R
q I A c
   *
0 0 10 0 ... 0
T
R
c 

Assumption
 Correspondence between economic industries and
critical infrastructure sectors
* includes waste management service
 Sector : New York City Government report
 Sector : Department of Homeland Security (DHS) definition
Code Industries Critical infrastructure sectors
11 Agriculture, forestry, fishing, and hunting Food and Agricolture
21 Mining Liquid Fuels
22 Utilities Utilities
23 Construction Buildings
31G Manufacturing Critical Manufacturing
42 Wholesale trade Commercial Facilities
44RT Retail trade Commercial Facilities
48TW Transportation and warehousing Transportation
51 Information Communications
FIRE Finance, insurance, real estate, rental, and leasing Financial Services
PROF Professional and business services* Solid Waste, Water and Wastewater
6 Educational services, health care, and social assistance Healthcare and Public Health
7 Arts, entertainment, recreation, accommodation, and food services Commercial Facilities
81 Other services, except government Emergencies Services
G Government Government Facilities

 Not reasonable percentages of inoperability
 Used as “magnitudes” to scale the inoperability of the sectors
proportionally to the inoperability of the sector subjected to
functionality reduction
Utilities
LiquidFuels
Transportation
Buildings
SolidWaste,Water
andWastewater
Critical
Manufacturing
Commercial
Facilities
FinancialServices
Communications
Emergencies
Services
Government
Facilities
Foodand
Agricolture
Healthcareand
PublicHealth
10.00 0.50 0.14 0.05 0.05 0.04 0.03 0.02 0.01 0.01 0.01 0.00 0.00
20.00 1.00 0.28 0.10 0.09 0.08 0.05 0.03 0.03 0.02 0.01 0.00 0.00
30.00 1.50 0.42 0.15 0.14 0.12 0.08 0.05 0.04 0.03 0.02 0.00 0.00
40.00 1.99 0.56 0.20 0.18 0.16 0.11 0.07 0.05 0.04 0.03 0.00 0.00
50.00 2.49 0.70 0.25 0.23 0.20 0.14 0.08 0.06 0.05 0.04 0.00 0.00
60.00 2.99 0.84 0.30 0.28 0.24 0.16 0.10 0.08 0.06 0.04 0.00 0.00
70.00 3.49 0.98 0.35 0.32 0.28 0.19 0.11 0.09 0.07 0.05 0.00 0.00
80.00 3.99 1.12 0.40 0.37 0.32 0.22 0.13 0.10 0.08 0.06 0.00 0.00
90.00 4.49 1.26 0.45 0.42 0.36 0.25 0.15 0.12 0.09 0.07 0.00 0.00
100.00 4.98 1.40 0.51 0.46 0.40 0.27 0.16 0.13 0.10 0.07 0.00 0.00
%FUNCTIONALITY
REDUCTIONOFUTILITIES
SECTOR
% INOPERABILITY FOR SECTORS
Results

Utilities
LiquidFuels
Transportation
Buildings
SolidWaste,Water
andWastewater
Critical
Manufacturing
Commercial
Facilities
FinancialServices
Communications
Emergencies
Services
Government
Facilities
Foodand
Agricolture
Healthcareand
PublicHealth
10.00 5.87 1.65 0.60 0.54 0.47 0.32 0.19 0.15 0.12 0.09 0.01 0.00
20.00 11.74 3.29 1.19 1.09 0.93 0.64 0.39 0.30 0.23 0.17 0.01 0.01
30.00 17.61 4.94 1.79 1.63 1.40 0.97 0.58 0.45 0.35 0.26 0.02 0.01
40.00 23.48 6.59 2.38 2.18 1.86 1.29 0.77 0.60 0.46 0.34 0.02 0.02
50.00 29.35 8.23 2.98 2.72 2.33 1.61 0.96 0.76 0.58 0.43 0.03 0.02
60.00 35.22 9.88 3.57 3.27 2.80 1.93 1.16 0.91 0.69 0.51 0.03 0.03
70.00 41.09 11.53 4.17 3.81 3.26 2.26 1.35 1.06 0.81 0.60 0.04 0.03
80.00 46.96 13.17 4.76 4.36 3.73 2.58 1.54 1.21 0.92 0.69 0.04 0.04
90.00 52.83 14.82 5.36 4.90 4.20 2.90 1.73 1.36 1.04 0.77 0.05 0.04
100.00 58.70 16.46 5.95 5.45 4.66 3.22 1.93 1.51 1.15 0.86 0.06 0.05
%FUNCTIONALITY
REDUCTIONOFUTILITIES
SECTOR
% INOPERABILITY FOR SECTORS
Results
 New inoperability:
R
jR R
j scaled pR
j
q
q q
q

qR
j: inoperability of the jth sector not subjected to
qR
p: inoperability of the sector affected by

 Comparison for 10% functionality reduction of utilities sector
Results
PIETRO CRUPI
Code Critical Infrastructure sectors
Ut Utilities
LF Liquid Fuel
Tr Transportation
Bu Buildings
SW Solid Waste, Water and Wastewater
CM Critical Manufacturing
CF Commercial Facilities
Code Critical Infrastructure sectors
FS Financial Services
Co Communications
ES Emergency Services
GF Goverment Facilities
FA Food and Agricolture
HP Healthcare and Public Health
Pietro Crupi

 Function of the dependency from the perturbed sector
 Ratios for functionality reductions of utilities, transportation,
and liquid fuel sectors:
Inoperability ratio
R
j scaled
pj R
p
q
Q
q

qR
j scaled: new induced inoperability
qR
p: inoperability of the sector affected by
α% 0.16α% 0.59α%
0.09β% β% 0.05β%
0.13γ% 0.22γ% γ%
Transportation
LiquidFuel
UTILITIES
TRANSPORTATION
LIQUID FUEL
Utilities
α%:
inoperability ratio for utilities sector
when it is subjected to functionality
reduction
0.59α%:
inoperability of the liquid fuel sector
always equal to the 59% of the
inoperability of the utilities sector

 Percentages and lack of reciprocity justified by several
examples regarding the influence among these three sectors
during Hurricane Sandy:
 Closure of gas stations because of no power to pump fuel
or no possibility to fast connect to backup generators
(utilities  liquid fuel)
 Power outage and damage to electric equipment caused
the suspension of train and subway services
(utilities  transportation)
 Selection and ranking of the priority initiatives that bring to a
reduction of the inoperability ratios between different sectors:
 Indirect damage amount not negligible
 Induced inoperability is a considerable component of the
overall inoperability of the sector
Considerations

Priority initiatives
PRIMARY INITIATIVES FOR FUNCTIONALITY REDUCTION OF UTILITIES
UTILITIES
α%
LIQUID FUEL
0.59α%
Causes Effects Initiatives
Power outage
Not functioning backup
generators
Shutdown of refineries and
pipelines or reduction of their
operations
1: Develop a fuel infrastructure hardening strategy
Power outage
Damage to terminals electric
equipment
Shutdown of terminals or
reduction of their operations,
impossibility to discharge fuel
tankers
6: Creation of a transportation fuel reserve
Power outage
No possibility to fast connect to
backup generators
Closure of gas stations 5: Ensure that a subset of gas stations and terminals have
access to backup generators in case of widespread power
outages
Lack of planning of backup
generators prepositioning
Closure of gas stations 4: Provision of incentives for the hardening of gas stations
Damage to electric systems and
equipment
Bottlenecks along pipelines and
delays in fuel supply
3: Build pipeline booster stations in New York City
Damage to fuel facilities
electric equipment
Reduction of capacity to
dispense fuel to delivery trucks
8: Development of a package of City, State, and Federal
regulatory actions to address liquid fuel shortages during
emergencies

SECONDARY INITIATIVES FOR FUNCTIONALITY REDUCTION OF UTILITIES
UTILITIES
α%
TRANSPORTATION
0.16α%
Power outage No functioning traffic signals 3: Elevation of traffic signals and provision of backup electrical
power
Damage to overhead power
lines tore down by tree
branches and/or wind
Closure of streets 6: Hardening of vulnerable overhead lines against winds
Power outage
Damage to tunnel electrical
equipment and control systems
Closure of road and rail tunnels 4: Protection of NYCDOT tunnels from flooding
Power outage
Damage to bridges electrical
equipment and control systems
Inoperability of moveable
bridges
5: Installation of watertight barriers for mechanical
equipment of bridges
Reparation or replacement of
old and damaged subway
electric equipment
Delayed restoration of subway
service
1: Develop a cost-effective upgrade plan of utilities systems
Power outage
Damage to key electric
equipment
Suspension of train and subway
services, overwhelming of other
transportation systems that do
not rely on power lines, and
more private vehicles traffic
9: Planning for temporary transit services in the event of
subway system suspensions
12: Planning and installation of new pedestrian and bicycle
facilities
11: Implementation of High-Occupancy Vehicle (HOV)
requirements

PRIMARY INITIATIVES FOR FUNCTIONALITY REDUCTION OF TRANSPORTATION
TRANSPORTATION
β%
UTILITIES
0.09β%
Street damage and closure Delayed utilities restoration
efforts and collection of
damages information
13: Implementation of smart grid technologies
Street damage Limited access for restoration
crews to critical customers
affected by utilities damages
14: Speed up service restoration for critical customers via
system configuration
23: Improvement of backup generation for critical customers
SECONDARY INITIATIVES FOR FUNCTIONALITY REDUCTION OF TRANSPORTATION
TRANSPORTATION
β%
LIQUID FUEL
0.05β%
Street damage Limited access to fuel facilities 8: Development of a package of City, State, and Federal
regulatory actions to address liquid fuel shortages during
emergencies
Street damage Delays in fuel supply and fuel
delivery trucks detours
9: Hardening of municipal fueling stations and enhancing of
mobile fueling capability

Dynamic recovery
 Resilience coefficient:
 Equation for the model:
 Data for utilities recovery:
 qi(0) = 47% (customers affected by power outages)
 qi(Ti) = 1% (99% recovery achieved in Ti)
 Ti = 30 days
*
*
1
1
ln[ (0) / ( )] 1
1
0.1289 /
i
ii
i i i
i ii
k
a
q q T
T a
day


 
   
 
   

λ: recovery constant
τ: recovery time Ti
aii
*: diagonal element of matrix A*R
qi(0): inoperability of i sector at
perturbation (t=0)
qi(Ti): inoperability of i sector at Ti
*
(1 )
( ) (0)i iik a t
i i
q t e q 


Results
*
(1 )
( ) (0)i iik a t
i i
q t e q 

*
(1 )
1 ( ) 1 (0)i iik a t
i i
q t e q 
  
 Recovery of the utilities sector in terms of
reduction of inoperability and increase of functionality

 Interconnectivity among c.i.s. not negligible when
planning to increase their resilience:
 Crucial for c.i.s. functioning both in normal conditions
and emergency situations
 Determine perturbations propagation
 IIM realistically defines the interconnectivity and the
effects on the network of a perturbation to one system
 Decision-maker need to be guided in the policy selection
 Selection and ranking of priority initiatives through IIM:
 Reduction of inoperability ratio
 Primary and secondary initiatives
Conclusions

POLITECNICO DI TORINO - DISEG G. P. CIMELLAROPIETRO CRUPI
Thank you for your attention!
Pietro Crupi

160315 Thesis Pietro Crupi

Recommended

Recommended

More Related Content

Similar to 160315 Thesis Pietro Crupi

Similar to 160315 Thesis Pietro Crupi (20)

160315 Thesis Pietro Crupi