Vulnerability Assessment of Spatial Networks:

Models and Solutions

Eduardo lvarez-Miranda1,2 , Alfredo Candia-Vjar2 , Emilio Carrizosa3 , and

Francisco Prez-Galarce2
1
DEI, Universit di Bologna, Italy [email protected]
2
DMGI, Universidad de Talca, Chile {ealvarez,acandia,franperez}@utalca.cl
3
Faculdad de Matemticas, Universidad de Sevilla, Spain [email protected]

Abstract. In this paper we present a collection of combinatorial op-

timization problems that allows to assess the vulnerability of spatial
networks in the presence of disruptions. The proposed measures of vul-
nerability along with the model of failure are suitable in many applica-
tions where the consideration of failures in the transportation system is
crucial. By means of computational results, we show how the proposed
methodology allows us to find useful information regarding the capacity
of a network to resist disruptions and under which circumstances the
network collapses.

1. Introduction
Shortest path problems correspond to an old and very known class of problems
in combinatorial optimization. A variant of one of these basic problem consists
on analyzing the effects of removing arcs from a network. In [14] the problem of
removing k arcs that cause the greatest decrease in the maximum flow from a
source to a sink in a planar network is studied. This problem is a special case of
a broad class of network optimization problems known as interdiction problems.
Applied to the shortest s, t-path problem, the interdiction problem can be de-
fined in the following way. Given a graph G = (V, E) with a non-negative length
function on its arcs l : E R and two terminals s, t V , the goal is to destroy
all (or the best) paths from s to t in G by optimally eliminating as many arcs
of A as possible (usually respecting a so-called interdiction budget). Interdiction
problems are often used to measure the robustness of solutions of network op-
timization problems. In [9] several versions of these problems are studied; they
consider the case of total limited interdiction when a fixed number of k arcs can
be removed, and node-wise limited interdiction (for each node v V a fixed
number k(v) of out-going arcs can be removed). For a complete survey on early
interdiction problems with different underlying network properties the reader is
referred to [2]. For a more general discussion regarding network vulnerability
approaches we suggest to see [10].
Based on a well-known network interdiction model we formulate a framework
of combinatorial optimization problems whose solutions can be used for assess-
ing the vulnerability of spatial networks in the case of disruptions. We design
2

a flexible model of network disruption based on the geometric characteristics of

spatial networks. This model incorporates the nature of the disruptions present
in different situations such as military planning [5, 8], terrorist attacks [12] or
emergency control of infectious disease spreading [1]. The proposed problems,
along with the model of disruption, span several realizations of network inter-
diction providing a useful tool to characterize network vulnerability. Our aim is
to propose a methodology that uses network optimization problems to charac-
terize the robustness of a network in the presence of multiple failures.
In 2 we present the optimization framework for vulnerability assessment;
in 3 we report computational results on realistic instances; these results show the
versatility of the proposed models to characterize the robustness of the network
infrastructure. Finally, in 4 we draw final conclusions and propose paths for
future work.

2. Vulnerability Measures as Optimization Problems

Notation Let G = (V, E) be a spatial network such that |V | = n and |E| = m.

Let s, t V be a source and a target node respectively; le , e : {i, j} E, be
the cost of edge e (distance between i and j); and ` be the cost of the shortest
s, t-path on G with edge costs given by le , e E.
Let X R2 be an arbitrary sub-region of R2 . An element x X is a point in
X ; for a given point x and a given edge e, let d(x, e) be the minimum distance
between x and the line segment defined by e (recall that e : {i, j} links node i
with node j, whose positions are given). For a given R R>0 and a given x X ,
let Ex = {e E | d(e, x) > R} and Ex = {e E | d(e, x) R}. In other words,
Ex is the set of edges that are not reached by the disk of radius R centered at x
(the disruption disk (x, R)), and Ex is the set of disrupted or interdiced edges.
We will refer to Gx = (V, Ex ) as the operating network with respect to (x, R).
Note that Gx might be disconnected.
The model of failure represented by (x, R) embodies a characteristic of dis-
ruption produced by many different sources: instead of having isolated failures,
we have a set of failures all of them circumscribed within a delimited area. This
naturally occurs in the application contexts that we have already mentioned.

2.1 The Max-Cost Single-Failure Shortest Path Problem

Let us assume that X is a finite set of points x in R2 and that R can take values
in R which is a finite subset of R>0 . Given a radius R R and a discrete set
X , we are interested in knowing what is the maximum length of a shortest
s, t-path across all possible locations x X of the disruption disk (x, R).
Knowing is threefold: (i) It tells us how severe a disruption can be by
comparing the value of with respect to `; in other words, the increase of the
transportation time between s and t induced by a failure located in the worst
location x = argxX {}. (ii) From the tactical point of view, preventive actions
can be taken in order to reduce the chances that a failure can be produced at x
 3

(a) No failure disk. (b) A disk (x1 , R1 ). (c) Two disks (x1 , R1 ) and
(x2 , R2 ).

Fig. 1. Example of a network G = (V, E), nodes s and t, a region X and different
interdiction cases.

or the edges Ex can be reinforced to increase their reliability. And (iii) we can
know whether the network is so vulnerable that s and t might be disconnected,
which can be verify if = .

The problem of calculating will be called the Max-Cost Single-Failure

Shortest Path Problem (MCSFSPP). Therefore, the MCSFSPP is an optimiza-
tion problem whose objective function value is a vulnerability measure of the
network on which it is solved. Intuitively, the MCSFSPP can be solved as fol-
lows. For a given x X , let `x be the cost of the shortest s, t-path on Gx
with edge costs lex defined as lex = le if e Ex and lex = M if e Ex , with
M = O(m maxeE le ); therefore, = maxxX `x . If > M then there is at
least one x for which s and t cannot be connected.

In Figure 1(a) it is shown a network G = (V, E) where s and t correspond to

the nodes represented with triangles and X is represented by a grid of 87 points
in the background of part of G; an optimal s, t-path is shown with bold edges.
In Figure 1(b) we show the case where a disruption disk (x1 , R1 ) interdicts
the network such that an alternative (an more expensive) s, t-path has to be
established ( < M ). And in Figure 1(c) a more complex situation is shown; here
two disruption disks, (x1 , R1 ) and (x2 , R2 ), are simultaneously interdicting the
network. In the latter case all possible s, t-paths (one of them is shown in bold
dashed lines) have at least one interdicted edge, i.e., > M .

The MCSFSPP is closely related with the network interdiction problems

studied in [4, 5, 11, 3, 8, 7]. In the following, we will use this basic definition to
construct generalizations addressing different, but complementary, measures of
vulnerability under different models of failure.

Mixed Integer Programming Formulation for the MCSFSPP

4

Let f [0, 1]m be a vector of [0, 1]-flow variables. An s, t-path p in G is

induced by a given allocation of flows f if the following constraints are satisfied:

X X 1,
if j = s
fj,k fi,j = 0, if j V \ {s, t} (SP.1)

kV |e:{j,k}E iV |e:{i,j}E
1, if j = t.

For a given x X , the problem of finding `x can be defined as

( )
X
x m
`x = min le fe | (SP.1) and f [0, 1] . (`x )
eE

Let y {0, 1}|X | be a vector of binary variables such that yx = 1 if the

failure disc is centered at x and yx = 0 otherwise. Now, let z {0, 1}m be a set
of binary variables such that ze = 1 if edge e is operative and ze = 0 otherwise
for any given x X . Variables y and z are related as follows

yx + ze 1, e E | d(e, x) R, x X (YZ.1)
X
yx ze 0, e E. (YZ.2)
xX |d(e,x)>R

Constraints (YZ.1) and (YZ.2) state that, for any x X , an edge e has to be
operative (ze = 1) if is not reached by the disruption disk (x, R). Since a single
disruption disk affects the network, we have that
X
yx = 1. (YZ.3)
xX

Using (YZ.1) and (YZ.2), for a given x X the edge costs lex can be written as
lex = le ze + (1 ze )M, e E. Hence, the MCSFSPP is as follows
n o
= max `x | (YZ.1), (YZ.2), (YZ.3) and (y, z) {0, 1}m+|X | . ()
xX

Problem (), as it is, is non-linear. To linearize it, we will convert the max min
objective into a pure max one; to do so, let us consider the dual of (`x ), which
is given by
`x = max {t s | j i lij zij + (1 zij )M, e : {i, j} E and Rn } . ()

Embedding () into(), we get the next Mixed Integer Programming Formula-

tion (MILP) formulation for the MCSFSPP:

= max t s (MCSF.1)
s.t (YZ.1), (YZ.2) and (YZ.3) (MCSF.2)
j i lij zij + (1 zij )M, e : {i, j} E (MCSF.3)
m+|X | n
(y, z) {0, 1} and R . (MCSF.4)
 5

Note that in our approach we assume that (x, R) can be located in any
point x X without any stochastic characterization. That is, any point x X
is likely to host the center of the failure.
In the proposed setting we assume that if an edge e is disrupted by at least
one failure disk (x, R), then it becomes inoperative. However, one can easily
extend this to a more general case by defining a coefficient de 0 e E
representing the delay on edge e in case of interdiction (in our setting de = M
e E). The MCSFSPP can be redefined by replacing constraint (MCSF.3)
with

j i lij + (1 zij )dij , e : {i, j} E. (MCSF.3b)

The Shortest-Path Network Interdiction problem presented in [8] is very sim-

ilar to the definition of the MCSFSPP using (MCSF.3b) instead of (MCSF.3). In
that problem, edges can be interdicted without any geometrical pattern among
them; instead, they consider interdiction costs so that any feasible disruption
of the network should not cost more than a given interdiction budget. Later we
formally define these concepts and adapt them to our setting.

2.2 The Multiple Failures Case

As described above, in the MCSFSPP only a single failure (x, R) occurs. How-
ever, there are applications in which this characteristic does not hold and,
instead, multiple failures occur simultaneously. More precisely, we now have
that k failure disks (x1 , R), . . . , (xk , R) of radius R are located in X , re-
sulting in an operative network Gxk = (V, Exk ) where Exk = {e E |
minx{x1 ,...,xk } d(e, x) R}. Under these conditions, finding the maximum cost,
across all possible {x1 , . . . , xk } X k , of the shortest s, t-path on Gxk can be
done by modifying MCSFSPP as follows. Instead of (YZ.3), we have
X
yx = k. (YZ.3k)
xX

Besides, constraint (YZ.2) should be now adapted in order to impose that ze = 1

if none of the k failure disks reaches e; the new constraint is
X X
yx ze 1 yx , e E, (YZ.2k)
xX |d(e,x)>R xX

clearly if k = 1, then (YZ.2k) corresponds to (YZ.2). Therefore, the Max-Cost

Multiple-Failure Shortest Path Problem (MCMFSPP) can be formulated as
k = max {s t | (YZ.1), (YZ.3k), (YZ.2k), (MCSF.3) and (MCSF.4)} (MCMF)

Note that in formulation (MCMF) it is assumed that R R is known in

advance.
Maximal Disruption for an interdiction budget Similar as in [3, 8, 7],
let us consider that associated with each point x X there is a disruption cost
6

cx > 0. Assume that the interdictors have a budget B of interdiction resources, so

that they can disrupt the network using several disks (x, R) as long as the total
cost does not exceed B. Formally, the interdiction-budget constraint is given by
X
cx yx B; (IB)
xX

so the Budget Constrained MCMFSPP is formulated as

k = max{s t | (YZ.1), (YZ.2k), (IB), (MCSF.3) and (MCSF.4)} (B)

By solving (B) we can know how vulnerable the network is if the interdictors are
able to optimally use their resources to disrupt it. Models as the one presented
in [8, 7] are particular cases of (B) in which X coincides with the midpoint of
every edge e E and R =  ( being infinitesimally small).
Minimum Simultaneity for Complete Vulnerability: Critical k One
might be interested in knowing the minimum number of failures (the critical
k or k c ) that should occur simultaneously in order to have at least one set
(x1 , R), . . . , (xk , R) that damages the network so that s and t cannot be con-
nected anymore or the shortest length between them is greater than a threshold
.
The value k c and the corresponding collection {x1 , . . . , xkc } will enable a
decision maker to perform more general preventive actions to endure the network
not in a single but a in several areas. In many practical contexts, the possibility
of multiple and synchronized failures might be the rule, so knowing k c might
play a strategical role. Clearly, for a given R, the larger k c is the more robust
the network is. Mathematically, one can formulate the search for k c as
kc = min {k | (YZ.1), (YZ.2k), (YZ.3k), (MCSF.3), (MCSF.4) ,
s t and k Z0 } (kc )

If = M , then (k c ) aims at finding the minimum k such that allocating k disks

produces a disconnection between s and t. A similar model is presented in [7] in
the context of interdiction in stochastic networks.
If instead of k c one is interested in knowing the minimum cost needed to pro-
duce a damage represented by , model (k c ) can P be easily modified by replacing
the objective function of (k c ) with C c = min xX cx yx .

3. Computational Results

3.1 Instance Benchmark and Solver Setting

Instance Benchmark For our experiments we consider three sets of in-

stances: ND, US and Bangladesh.
In the first set, the instances are generated as follows: (i) n points are ran-
domly located in a unit Euclidean square; (ii) a minimum spanning tree connect-
ing all points is calculated; (iii) n additional edges are added to the network
 7

(a) ND with n = 500. (b) Network of US. (c) Network of Bangladesh.

Fig. 2. Representation of the instances used for computations.

such that an edge is added if lij (euclidean distance) satisfies lij / n and the
planarity of the network is still preserved; (iv) the set X is created by randomly
located K points within the are area defined by points (x1 , y1 ), (x2 , y1 ), (x1 , y2 )
and (x2 , y2 ).
For experiments we have considered n {500, 1000}, = 1.5, = 1.6,
(x1 , x2 , y1 , y2 ) = (0.3, 0.7, 0.0, 1.0) (X1 ) and (x1 , x2 , y1 , y2 ) = (0.1, 0.9, 0.1, 0.9)
(X2 ), and K = 100.
In Figure 2(a) it is shown an example of an instance with 500 nodes and X
contained in (0.3, 0.0), (0.7, 0.0), (0.3, 1.0) and (0.7, 1.0).
In the case of groups US and Bangladesh we consider the geographical co-
ordinates of the most populated cities in each case [see 13] to define the set V .
Then, we used an approximation of their highway and interurban road system
with the information available in [6] to approximate the set of edges E. The set
X is created by randomly located K points within the are area defined by points
(x1 , y1 ), (x2 , y1 ), (x1 , y2 ) and (x2 , y2 ). In Figures 2(b) and 2(c) we show the
networks used to generate the instances US and Bangladesh respectively. In the
case of US, the area X is given by placing 100 points in the so-called south area.
With this we intend to represent possible cases of failure produced by hurricanes
and other natural disasters. For the Bangladesh instances, we have created X
by placing 100 points in squared area in the very center that covers around the
15% of the total area.
In the case of instances ND, nodes s and t are selected as those with the
longest euclidean distance. In the case of instances US we have used s
{NY:New York, CH:Chicago} and t {LA:Los ngeles, HS:Houston}; likewise,
in the case of instances Bangladesh we have used s = Rajshahi and t = Silhat.
Solver Setting Models (MCSF.1)-(MCSF.4), (MCMF) and (k c ) were solved
using CPLEX 12.5 (all CPLEX parameters were set to their default values).
The experiments were performed on a Intel Core i7-3610QM machine with 8 GB
RAM.
8

n = 500 n = 1000
X1 X2 X1 X2
R % t[sec] R % t[sec] R % t[sec] R % t[sec]
0.01 2.17 38.92 0.01 0.00 32.93 0.01 2.48 115.39 0.01 0.14 144.66
0.02 3.93 46.46 0.02 1.64 36.83 0.02 2.93 153.58 0.02 0.69 215.16
0.03 3.93 65.63 0.03 1.64 49.73 0.03 5.38 235.95 0.03 1.52 240.88
0.04 5.15 80.79 0.04 1.64 63.06 0.04 7.31 258.85 0.04 1.52 265.00
0.05 5.15 103.01 0.05 1.64 79.67 0.05 7.17 395.46 0.05 1.52 259.80
0.10 5.15 97.83 0.10 1.64 112.76 0.10 8.69 917.78 0.10 3.87 373.61
0.15 - 53.70 0.15 10.64 111.65 0.15 9.93 587.27 0.15 6.19 843.45
Table 1. Solutions for the MCSFSPP considering different values of R (Instances ND)

3.2 Vulnerability Assessment of Spatial Networks: Solutions

From the operative perspective, the value of R corresponds to the intensity of a

disruption. If we consider the MCSFSPP or the MCMFSPP we would expect that
a vulnerable network is such that increases quickly (up to M ) when R increases
marginally. On the other hand, a reliable network is such that the cost of the
shortest s, t-path does not change too much even if R increases considerably.
In Table 1 we report solutions for the MCSFSPP for instances of group ND
considering different values of n, different compositions of set X and different
values of R (columns 1, 4, 7 and 10). In columns % is reported the relative
increase of , for a given X and a given R, with respect to cost of the shortest
s, t-path without any failure. In this column, "-" means that all paths have been
disrupted. In columns t[sec] are reported the running times in seconds needed to
reach optimality. One can observe from this table that when the area where the
failure can occur, X , is such that covers a stripe on the network (as X1 ) then it is
more vulnerable (see the values % for different R) than a network in which
the failure area, although larger, still leaves corridors where s, t-paths can be
constructed, as for X2 . In a warfare context, if we were to be the enemies, this
analysis would suggest us that is better to concentrate our resources in a narrower
area potentially spanning a complete stripe of the network than in a larger area
(which might be more expensive) that does not properly covers the network. On
the other hand, who wants to protect the network should concentrate the efforts
in protecting at least one corridor connecting s and t.
In Tables 2 and 3, results for (MCMF) and (k c ), respectively, are reported.
The analysis is similar as for Table 1. From Table 2 we can see that the increase
of % (due to a larger k), is greater for X1 than for X2 . Along the same lines,
we see from Table 3 that the minimum resources needed to disconnect s and t
(see columns k c ) are greater for X2 than for X1 . In Table 3, when results for a
given R are not reported (e.g., R = 0.01 for n = 500 and X1 ) is because not even
|X | failure disks are enough to make the s, t connectivity collapse. This applies
for all the remaining Tables.
 9

500 1000
X1 X2 X1 X2
R k % t[sec] R k % t[sec] R k % t[sec] R k % t[sec]
0.01 1 2.17 24.15 0.01 1 0.00 22.51 0.01 1 2.48 88.16 0.01 1 0.01 88.64
2 3.93 25.07 2 0.00 21.96 2 3.03 94.80 2 0.09 96.35
3 4.74 25.02 3 0.00 22.11 3 3.03 93.90 3 1.63 93.57
4 5.56 24.87 4 0.00 22.21 4 3.03 95.08 4 1.63 95.08
5 6.79 24.52 5 0.00 22.14 5 3.03 94.15 5 1.63 94.4
0.1 1 5.15 59.94 0.1 1 1.64 69.61 0.1 1 8.68 889.61 0.1 1 3.87 222.36
2 - 26.43 2 13.82 142.93 2 17.89 3448.93 2 8.16 1012.06
3 - 98.05 3 13.92 297.95 3 21.24 75319.9 3 12.45 5434.01
4 - 526.47 4 - 63.01 4 28.14 26123.3 4 21.99 13130.3
5 - 147.39 5 - 149.54 5 - 16576.5 5 - 742.63
Table 2. Solutions for the MCMFSPP considering different values of R and k (In-
stances ND)

n X R kc t[sec] n X R kc t[sec]
500 X1 0.10 2 170.76 1000 X1 0.10 5 628.76
0.15 1 43.09 0.15 2 805.32
X2 0.03 10 39.05 X2 0.02 19 219.38
0.04 9 109.70 0.03 13 621.60
0.05 7 161.06 0.04 11 933.85
0.10 4 274.20 0.05 8 1759.32
0.15 3 162.47 0.10 5 1277.24
0.15 3 1214.14
Table 3. Solutions of (kc ) considering different values of R (Instances ND)

From the algorithmic point of view, we can notice in Tables 1, 2 and 3 that
the search for an alternative path in a disrupted network is not for free. In all
cases we see an increase of the algorithmic effort (time) needed to find such a
path (if exists). This is due to the high combinatorial nature of the problem
when more edges are subject to be interdicted (when R increases and/or when
k is either greater than 1 or when it is a variable).
In the case of USA Instances, we report in Table 4 results of the MCSFSPP
considering different pairs of s and t and different values of R. In this case, we
can see that different combinations of s and t yield to different levels of vulner-
ability in the system. For instance, the network is considerably more vulnerable
when it is intended to host a path from Chicago to Los ngeles than when the
path should be established from Chicago to Houston. This is due to the fact
that, in our instance, the system of roads connecting the north of the Midwest
with the south of the West Coast is composed by relatively few elements. Hence,
a single disruption disk (that is optimally placed) is enough to interrupt the
communication between the cities. In this case the values of % are particu-
10

{N Y, LA} {CH, LA} {N Y, HS} {CH, HS}

R % t[sec] R % t[sec] R % t[sec] R % t[sec]
0.01 9.00 17.22 0.01 0.00 17.43 0.01 13.00 17.25 0.01 0.00 17.22
0.02 10.00 21.42 0.02 20.00 25.05 0.02 13.00 26.83 0.02 2.00 20.12
0.03 10.00 24.77 0.03 20.00 33.17 0.03 15.00 22.25 0.03 2.00 35.27
0.04 10.00 23.57 0.04 20.00 31.29 0.04 18.00 23.07 0.04 2.00 28.31
0.05 10.00 25.30 0.05 20.00 35.51 0.05 19.00 25.04 0.05 8.00 36.15
0.10 30.00 34.16 0.10 - 33.67 0.10 45.00 42.31 0.10 - 30.67
0.15 - 40.22 0.15 - 28.41 0.15 - 29.69 0.15 - 28.17
0.20 - 28.00 0.20 - 48.14 0.20 - 27.16 0.20 - 37.13
Table 4. Solutions for the MCSFSPP considering different values of R (Instances USA)

larly important from the tactic point of view; if it is up to the decision maker to
decide where to establish both the source and the target of the transportation
system, then it might preferable to have New York - Houston than, for instance,
Chicago - Los ngeles. However, this analysis is valid only when a single failure
occurs. For an approximate equivalence to real distances in kilometers, R should
be multiply by 1700.
In Figure 3(a) we show the solution of the shortest path problem between
New York and Houston when there is no disruption. In Figure 3(b) is shown the
solution of the MCMFSPP when 5 disruption disks with R = 0.01 are optimally
located. In Figure 3(b) is shown the solution of the MCMFSPP with k = 1 and
R = 0.10. These figures show how different the optimal s, t-paths can be when
the network is disrupted by failures of different magnitude.

(a) k=0 (b) k = 5 and R = 0.01 (c) k = 1 and R = 0.1

Fig. 3. Solutions for the MCMFSPP for different k and R (Instances USA)

Finally, in Table 5 we report results for the Instances Bangladesh. From

the solutions of the MCSFSPP (reported in columns 1-3) we can see that the
relatively dense road system of this country is able to resist (small values of
%), reasonably well the optimal location of a single failure disk up to R =
 11

R = 0.01 R = 0.1
R % t[sec] R kc t[sec] k % t[sec] k % t[sec]
0.01 3.76 8.72 0.05 3 10.06 1 3.76 7.69 1 24.17 8.11
0.02 2.93 7.44 0.10 2 28.52 2 5.48 8.75 2 - 38.28
0.03 3.76 8.99 0.15 1 24.01 3 5.48 8.42 3 - 26.75
0.04 3.76 9.53 4 5.48 8.41 4 - 28.32
0.05 4.78 19.03 5 5.48 7.64 5 - 10.55
0.10 24.17 36.97
0.15 - 11.19
Table 5. Solutions for MCSFSPP, MCSFMPP and kc , s =Rajshahi and t =Silhat
(Instances Bangladesh)

0.05. For greater values, the network can be dramatically damaged. This later
observation is reinforced by the results reported in columns 4-6 in the same table:
a critical k can be found only if R 0.05. When looking at the results of the
MCSFMPP (columns 7-9 for R = 0.01 and 10-12 for R = 0.1) we can see that
the network resists well (% 5%) several failures with R = 0.01; however,
if R = 0.1 then the network collapses even if k = 2.

4. Conclusions and Future Work

We have presented a collection of combinatorial optimization problems that in
combination allow to measure the vulnerability of a network. Vulnerability is
represented by the relative increase of the cost of a s, t-shortest path when part
of the network is disrupted. By analyzing the solutions of these problems for
different instances, we have highlighted how different aspects of both the failure
and the network yield to different levels of vulnerability.
Two main paths of future work can be identified. First, we should consider
the case in which X is not given by a discrete set of points, but rather as
continuous area. Second, at the light of the large computational effort needed to
solve some of the instance considered here, we think it is important to design
and implement more sophisticated algorithmic techniques such as decomposition
approach in order to be able to consider larger and more complex instances.

Acknowledgements This research was supported by Fondecyt Project #1121095,

CONICYT, Ministerio de Educacin, Chile. Eduardo lvarez-Miranda thanks the In-
stitute of Advanced Studies of the Universit di Bologna from where he is a PhD
Fellow.
 Bibliography

[1] N. Assimakopoulos. A network interdiction model for hospital infection

control. Computers in Biology and Medicine, 17(6):413 422, 1987.
[2] R. Church, M. Scaparra, and R. Middleton. Identifying critical infrastruc-
ture: The median and covering facility interdiction problems. Annals of the
Association of American Geographers, 94(3):491502, 2004.
[3] K. Cormican, D. Morton, and R. Wood. Stochastic network interdiction.
Operations Research, 46(2):184197, 1998.
[4] D. Fulkerson and G. Harding. Maximizing the minimum source-sink path
subject to a budget constraint. Mathematical Programming, 13(1):116118,
1977.
[5] B. Golden. A problem in network interdiction. Naval Research Logistics
Quarterly, 25(4):711713, 1978.
[6] Google. Google Maps, 2013. URL https://maps.google.com/.
[7] R. Hemmecke, R. Schltz, and D. Woodruff. Interdicting stochastic net-
works with binary interdiction effort. In D. Woodruff, editor, Network In-
terdiction and Stochastic Integer Programming, volume 22 of Operations
Research/Computer Science Interfaces Series, pages 6984. Springer US,
2003.
[8] E. Israeli and R. Wood. Shortest-path network interdiction. Networks, 40
(2):97111, 2002.
[9] L. Khachiyan, E. Boros, K. Borys, K. Elbassioni, V. Gurvich, G. Rudolf,
and J. Zhao. On short paths interdiction problems: Total and node-wise
limited interdiction. Theory of Computing Systems, 43(2):204233, 2008.
ISSN 1432-4350.
[10] A. Murray. An overview of network vulnerability modeling approaches.
GeoJournal, 78(2):209221, 2013.
[11] C. Phillips. The network inhibition problem. In Proceedings of the Twenty-
fifth Annual ACM Symposium on Theory of Computing, STOC 93, pages
776785. ACM, 1993.
[12] J. Salmeron, K. Wood, and R. Baldick. Worst-case interdiction analysis of
large-scale electric power grids. IEEE Transactions on Power Systems, 24
(1):96104, 2009.
[13] United Nations Statistics Division. UNSD Statistical Databases, 2013. URL
http://millenniumindicators.un.org/unsd/databases.htm.
[14] R. Wollmer. Removing arcs from a network. Operations Research, 12(6):
934940, 1964.