Transcript ppt - ISMOR

Illicit Trafficking in
Central America: Honduras
Cornwallis XVII, 2 April 2012
©
1
Purpose & Agenda
Purpose:
• Provide on overview of work completed to date on Senior Capstone.
• Receive feedback to incorporate in model development going forward.

Introductions

Problem Definition
2
Modeling

Network Flow Model

Background

Nodal Analysis

Previous Work

Formulation

Problem Statement

Utility

Modeling Approaches



Limitations/Assumptions
Research and Data Analysis

SOUTHCOM Visit

Key Resources

Cocaine Database Analysis

Summary Effects of Illicit
Trafficking


System Dynamics
Summary and Way Forward
Introductions
Errin Helbling


Systems Engineering Major
Zach Price


Operations Research Major
Derek Sipperly



Systems Engineering Major
Academic Internship with USSOUTHCOM J8, Summer
2011
Brad Kelly


3
Engineering Management Major
Background



4
Project initiated by Engineering Research and
Development Center - Construction
Engineering Research Lab (ERDC-CERL)
through the Center for Nation Reconstruction
and Capacity Development (CNRCD) at West
Point in 2010.
Analysis intended to complement ongoing
research at ERDC for the development of
tools and techniques to rapidly prototype
localized agent models for the analysis of
infrastructure and essential service
improvements on the local population.
Aimed at identifying key areas to insert
influence in order to impede illicit trafficking
activity within Honduras.
Previous Work
Develop a Holistic, Systems Understanding of the
Illicit Trafficking Problem in Honduras
Illicit
Trafficking
Problem
exacerbated
by
has
creates
International
Aid
provided
by
fails to
protect
offers
Extremely
High Debt
US
Government
and NGOs
US Policy
Goals
Strengthened
Democracy, Effective
Justice System, Protect
Human Rights, Promote
Rule of Law,Promote
Sustainable Economic
Growth
to assist
Honduran
Government
leads
to
decreases
helps
Local
Gangs
fuels
by
Confidence
in System
Local
Populace
Literacy
Rates
Extremely High Poverty
causes
seek
creates
Education
System
deters
Jobs
increases
Illegal Migration,
Crime, Narcotics
Trafficking,
Trafficking in
Persons, Port
Security
Law
Enforcement
Presence
that
have
lack
of
to
create
Trafficking
Networks
Remote Areas
(landing strips)
as
Improved
Living
Conditions
Reach
Out
from
Ports
Unsecure
Locations
which
protects
Highways
hinders
with
Human
Rights
reduces
Monetary
Aid
Cooperation
List
to
assist
Public
Security
entices
guided
by
5
recruit
hinders
provided
by
NGOs and
other
International
Org
Community
Corrupt
Rule of Law
Intimidating
Presence
requires
SA/Mexican
Cartels and
Producers
allows
utilizes
Illicit
Traffickers
Problem Statement

Develop viable models that can be
incorporated into ERDC tools and techniques,
and offer insights into the allocation of
resources to infrastructure and essential
services to affect illicit trafficking (2011-2012)
Current Modeling
Approaches



Network Flow

Strengths: Can accurately reflect drug trafficking network,
well-matched for military/anti-terrorism scenarios.

Weaknesses: Limited data resources to quantify and “solve”
the network problem. Network is ever-changing.
System Dynamics

Strengths: Grounded in causal relationships between many
different aspects of society. Can be quantified in many different
values. Extensive studies already in illicit trafficking and crime.

Weaknesses: Relies heavily on empirical data support to
develop behavioral relationship equations.
Network Science

Strengths: Identifies network’s “key players” and anticipates
their actions through organizational structure analysis
(communication, activity coordination, and decision making).

Weaknesses: Requires abundance of data resources.
7
Constraints, Limitations,
Assumptions
Constraints:


Time, only two 40 lesson semesters.
Limitations:


Readily available data sources and subject matter experts.


Limited availability of specific data relating local populace to illicit
trafficking.
Language barrier with Honduran local newspapers and other primary
source documents.
Assumptions:



8
Data available will be adequate to develop a holistic, systems
understanding of illicit trafficking throughout the region.
General modeling approaches based on data available
can be useful when applied to local conditions.
SOUTHCOM Visit
Purpose: Provided USSOUTHCOM an overview of a Department of
Systems Engineering at West Point Cadet Capstone, while
• gaining an improved understanding of illicit trafficking from
USSOUTHCOM subject matter experts
• receiving feedback on current approach
• exploring collaboration opportunities going forward
Outcome: Valuable trip resulting in
• understanding where our capstone fits into the problem and what
exactly we are looking for
• subject matter expertise in formulating nodes and arcs of our network
flow model
9
Key Resources and Data
Sources

10
Current Primary Resources
 Cocaine Movement Trends Update End of Year 2010 – Office of National
Drug Control Policy
 Provided analysis of the total cocaine flow through Central America
 Cocaine Movement Trends Update Jan – Mar 2011 – Consolidated
Counterdrug Database
 Provided data points for analysis in our Network Flow Analysis
 Woodrow Wilson International Center for Scholars, September 2011
 Provided general information and tactics, techniques, and procedures
on illicit trafficking through Honduras and Central America
 United Nations Office on Drugs and Crime – Bi-Annual Seizure Reports
 World Bank Data
Cocaine Trafficking From South
America to North America
“DTOs (Drug Trafficking Organizations)
are businesses. Their objective is to
limit costs and maximize profits. They
do this by trying to minimize the
number of participants, borders
crossed, authorities they have to bribe.”

In 2009, an estimated 380 metric tons
of cocaine was shipped from South
America (source) to North America
(sink).

DTOs ship cocaine as a commodity
flowing through a network of nodes
and arcs.

DTOs ship to minimize cost.

In the 1980s and 90s, Caribbean
routes were preferred until improved
airspace and maritime monitoring and
intelligence sharing made arc costs
prohibitive.

In the past decade, Central America
has become the preferred
transshipment for cocaine moving
North.
Source: Wilson Article
Trafficking Corridors to North America
11
Source: USSOUTHCOM Presentation
Source: UNODC, World Drug Report 2011
Consolidated Cocaine
Database Findings
Year of Seizure vs. Total Number of
Seizures by Mode of Transport
400
# of Cocaine Seizures
350
300
250
Maritime
200
land
150
Air
• # total seizures continues to
climb but land seizures are low.
• General upward trend of seizures
drives further question.
• Is drug trafficking as a whole
increasing as a business?
• Are methods of interdiction
improving?
100
50
0
2004
2005
2006
2007
2008
Year
2009
2010
2011
2012
• Data from cocaine database (above) is
measured from 2005 through Quarter 1 of
2011.
Source: CCDB 2005 - 1st QTR 2011
12
2005
2006
2007
2008
2009
2010
2011
Totals
Maritime
98
62
49
144
281
335
142
1111
land
Air
26
6
8
4
8
8
1
61
14
11
37
51
84
97
55
349
Consolidated Cocaine
Database Findings
Size of Cocaine Seizure by Year
and Mode of Transport
Seizure Amount (kg of cocaine)
450000
400000
350000
300000
250000
Maritime
200000
Land
150000
Air
100000
50000
0
2004
2005
2006
2007
2008
Year
2009
2010
2011
2012
• Data of size of seizure (kg) by type or
transportation and year from cocaine
database (above) is measured from 2005
through Quarter 1 of 2011.
Source: CCDB 2005 - 1st QTR 2011
13
• Slope of two plots are very
similar.
• Do traffickers simply prefer to
traffic drugs via maritime
because more can be
transported at once or is that all
we target?
• Land interdictions remain low
while maritime and air trends
rise.
2005
2006
2007
2008
2009
2010
2011
Totals
Maritime
Land
Air
(kg)
(kg)
(kg)
122290 5216.273 12119.56
89074
118
20240
60625
1058 40263.06
193905 1526.62 56655.3
332967.3
6462
75991
394627.5
4241.9 118098.5
258413.5
72 51289.88
1451902 18694.79 374657.3
Interdictions by
Origin Country
• Pie chart
represents
percentages of
interdictions
based on origin
country.
• Most cocaine
destined for
Honduras
comes from
Colombia and
Venezuela.
Source: CCDB
2005 - 1st QTR
2011
14
Event Type
by Origin Country
• Interdiction type based on
frequency and origin country in
transit to Honduras.
• Denotes importance of maritime
interdiction, especially in Atlantic
coast corridors.
• Shows that Honduras is more of
a transshipment destination
rather than a supply or demand
destination.
Source: CCDB 2005 - 1st QTR 2011
15
Drug Amount
vs. Origin Country (kg)
Sum of Drug Amunt (Kg)
Pareto Chart of Drug Amount (kg) v. Origin Country
180000
170000
160000
150000
140000
130000
120000
110000
100000
90000
80000
70000
60000
50000
40000
30000
20000
10000
0
100
163472
90
80
70
60
50
40
53309
30
32737
15500 12472 10400
9390
20
6409
3290
2600
100
92
Sum of Drug Amount (kg)
Cum. Percent
10
0
Origin Country
• This graph represents the total drug amounts (kg) interdicted by origin country.
• Can help us make key assumptions as we formulate the network flow model as to specific
drug amounts coming from source nodes to/through Honduras.
• Exponential decay: Shows countries’ importance in developing source nodes – Columbia,
Venezuela, and Panama are responsible for the majority of cocaine in-flow to Honduras.
Source: CCDB 2005 - 1st QTR 2011
16
The Effect of Trafficking
Through Honduras
Cocaine Seizures vs. Homicides
100
90
80
70
60
50
40
30
20
10
0
2000




2002
dg cocaine seized
2004
2006
2008
2010
Time (yrs)
Homocide Rate (%)
(Murders per 100,000 people)



Increase in Mexican drug wars pushed
drug trafficking into Central America
Ousting of President Zelaya
(June 2009)
Honduras homicide rate doubled in
past decade
59% Hondurans below poverty line
Assassinations of government officials
and news reporters
Gangs control large territories of land
Latin America highest levels of youth
violence in world
“[Drug Trafficking] is the main
single factor behind the rising
levels of violence in the region”
World Bank 2011
17
Node Analysis:
San Pedro Sula
Cocaine Corridors (Geographical Sites)
Inflow: From the coast, Tela (maritime into Tela); and
from the Southeast, Yoro (air tracks into Yoro)
Outflow: Overland, west towards Guatemala
Political: “Governance”—many areas
controlled by gang activity
Military: “Security”; “Rule of Law”; 1000
police officers; impunity rate extremely high
(unsolved serious crimes); 60% of murders
linked to drugs
Economic: Manufacturing, Industrial center
of Honduras; transportation hub; light
industry and commercial production in
coffee, bananas, beef, sugar cane, tobacco,
rd
forestry;
18 2/3 of countries GDP
Social: 2nd largest city; 900k people, 1.2m in
metro area; many higher education
opportunities; most bilingual schools in
CENTAM
Infrastructure: major transportation hub;
international airport; private university, rail
lines (Tela, Puerto Cortes); largest bus
station in CENTAM; city in quadrants,
abundant public transportation
Information: highest murder rate in the
world
Trafficking Through Honduras
U.S.
Trujillo
Tela
San Pedro
Sula
Puerto
Lempira
Yoro
Catacamas
Santa
Rosa
De
Copan
Tegucigalpa
Choluteca
S.A.
(Col.)
19
Formulating the Network
Sink
U.S.
NA
C8NA
Node
Trujillo
3
C78
Node
Tela 7
C9NA
CSA3
C37
C23
San Pedro
Node 8
Sula
C36
Puerto
Node 2
Lempira
Node
Yoro6
Santa
Rosa
Node 9
De
Copan
C1NA
C68
C24
Catcamas
Node 4
C45
C59
CSA2
Tegucigalpa
Node 5
C15
Each arc has a cost, “Cij”
Choluteca
Node 1
20
CSA1
Source
S.A.
(Col.)
SA
Formulating the Network
Sink
NA
C8NA
Node 3
C78
Node 7
C9NA
CSA3
C37
C23
Node 8
C36
Node 2
Node 6
C68
C24
Node 4
Node 9
C45
C59
C1NA
CSA2
Node 5
C15
Each arc has a cost, “Cij”
Node 1
21
CSA1
Source
SA
Illicit Trafficking
Min Cost Network Flow
C78
Node 8
Node 7
C68
C37
Node 3
C36
C8NA
CSA3
C23
Node 6
Sink
NA
CSA2
Node 2
C9NA
C24
Node 4
CSA1
Node 9
C5-9
C45
C15
Node 5
C1NA
22
Node 1
Source
SA
Min Cost Network Flow
Formulation and Assumptions

xij = number of units of flow sent from node i to node j through arc (i,j)

bi = net supply (outflow – inflow) at node i

cij = cost of transporting 1 unit of flow from node i to node j via arc (i,j)

Lij = lower bound on flow through arc (i,j). If no lower bound, Lij = 0

Uij = upper bound on flow through arc (i,j). If no upper bound, let Uij = infinity

The first constraint means that the net flow out of node i must equal bi. This
is the flow balance equation. The second constraint ensures that the flow
through each arc satisfies the arc capacity restrictions.
23
Cost/Capacity
Indicators



Now that we have
established our
network's arcs and
nodes…
Indicators of how a
population lives
depending on cocaine
flow amount need to be
identified
Analyze these indicators
in order to model
cost/capacity as a
function of variables
related to whole of
government actions.

Ex: If we increase school
enrollment in one node,
what are the effects to the
network in terms of
cocaine traffic and
homicide rates?
24
Cocaine Seizures vs. Homicides
100
90
80
70
60
50
40
30
20
10
0
2000
2002
dg cocaine seized
Political
FDI
Military
Military Expenditure
Economic
Poverty Headcount
GDP per capita
2004
2006
Time (yrs)
Homocide Rate (%)
2008
2010
Social
Life Expectancy
School Enrollment
Children out of School
Health Expenditure
Infrastructure
Electricity Produced (kWh)
Information
Internet Users
Way Forward:
System Dynamics
Study Structure
1.
Problem Articulation
2.
Dynamic Hypothesis
3.
Formulation
4.
Testing
5.
Policy Formulation &
Evaluation
Possible Spheres of Interaction:
1. Primary: Drug Trafficking
Sphere
2. Government Policy Domain
3. Economic Influences Sphere
4. Drug Market Violence and
Crime Sphere
5. Civilian Population Interaction
Sphere
25
Summary and
Way Ahead
Summary:
• Developed thorough problem understanding
• Incorporated research/stakeholder input
• Performed data analysis of available resources
Developing a final network flow modeling approach
•
Way Ahead:
•
•
26
Developing System Dynamics technique, targeting specific areas of
influence and interdiction opportunities and incorporating effort into
Network Flow approach.
Final Project Presentation May 2
Questions
?
27