Resolution of Ambiguity through HUMINT An M&S Methodology - - PowerPoint PPT Presentation

Resolution of Ambiguity through HUMINT

Briefing to ISMOR

29 August 2007

An M&S Methodology

16 August 2007 2 HUMINT brief to ISMOR

Topics

• Project Overview.
• Ambiguity on the Battlefield.
• Project Scope.
• Project Approach.
• Ambiguity/HUMINT Methodology Overview.
• Summary & Status.
• Information.

16 August 2007 3 HUMINT brief to ISMOR

Project Overview

• Project Purpose: Develop model methodologies that can be used to better

represent ambiguity and the contribution of HUMINT in resolving that ambiguity, where HUMINT is defined as: – “The collection by a trained HUMINT collector of foreign information from people and multimedia to identify elements, intentions, composition, strength…” (FM 2-0, Intelligence)

• Project Result: The resulting methodology prescribes, through requirements

definition and model representations: – Multiple sides with varying and dynamic allegiances. – Perceived truths that include enemy intent. – Intelligence source reliability and cooperation factors. – HUMINT collection functionality. Problem: M&S generally do not portray the ambiguity that influences the

• perational decision making on the battlefield. Additionally, HUMINT, a

source to help resolve ambiguity, is only portrayed in a gross, implicit way. The methodology provides a means for M&S to adapt to and represent the dynamic nature of today’s battlefield. The methodology is currently being implemented in both analytic and training models and helping revise doctrine.

16 August 2007 4 HUMINT brief to ISMOR

Ambiguity on the Battlefield

I m p a c t The project’s intent is to represent in M&S, the ambiguity a commander faces in making decisions, the actions he takes to resolve ambiguity (e.g., HUMINT), and their consequences in terms of the impacts on ambiguity and operations. I n f l u e n c e Produce Actions

• Maneuver
• Fires
• Etc.
• Human Intel

Collection

• Civil Military

Operations

• Humanitarian

Relief

• Psychological

Operations

• Coercion
• Security

New No Longer Ground Truth

• What’s the enemy doing?
• What and where is his next

action?

• Whose information can I

rely on and to what degree?

Cdr’s Decisions Battlefield Conditions

• Enemy Locations
• Enemy Composition
• Enemy Activity
• Terrain
• Environment
• Etc.
• Allegiances
• Enemy Intent

Ambiguity

Operational Impacts

16 August 2007 5 HUMINT brief to ISMOR

Project Scope

The team scoped the project to achieve credible representation while focusing on high-payoff factors.

• Project focused on:

– Elements of ambiguity related to information quality and quantity as well as the commander’s perception of the information. – Only active HUMINT collection, vice passive, where active is a collection effort initiated to ascertain specific information. – Representation of processes, capabilities, and effects at the tactical and

• perational levels.

– Development of processes for combat simulation models, not performance or engineering models.

• Resulting methodology adaptable for:

– Current and future operational environments, with Threat considerations represented in the data. – Aggregate and entity-level processes. – Deterministic and stochastic models. – Constructive and human-in-the-loop (HITL) models. – Analytic and training models.

16 August 2007 6 HUMINT brief to ISMOR

Project Approach

Problem Decomposition

• 1a. Develop

Taxonomy & Conduct Literature Review

• 1b. Develop

Key Threads

• 1c. Develop

Real-World Process Description

• 2. Develop

Requirements

• 4. Implement as

Proof-of- Principle & Test

Requirements Development Methodology Development

1 2 3

• Military HUMINT

processes.

• Elements/causes
• f ambiguity.
• Information

provided by HUMINT to resolve ambiguity.

• Motivating factors
• f human sources.
• Role of HUMINT in

the decision making process.

• Real-world

process description.

• Existing HUMINT M&S methodologies.
• Model

requirements.

• User

requirements.

• Methodologies to

implement requirements. – Stochastic. – Deterministic. – Entity. – Aggregate. – Constructive. – HITL.

Inputs Outputs Legend:

The approach consists of four principal stages:

• Key aspects
• f real-world

processes.

Implementation and Testing

4

• 3. Develop

Methodologies

16 August 2007 7 HUMINT brief to ISMOR

Ambiguity/HUMINT Methodology Overview

Decision, Action & Evaluation Collection BC Planning and Asset Allocation

• Commander’s

critical information requirements (CCIR)

• Collection team

tasks

• Intelligence reports
• Tactical action
• Non-kinetic action

Updating Perception of Conditions

• Allegiances
• Faction perceptions
• Commander’s

influencers

• Enemy Information
• Unit
• - Activity
• - Disposition
• - Composition
• Intent
• Politics, State
• Sides, Allegiances
• Non-combatants
• Factions
• Para-military
• Unknowns

Ambiguity -- What do we know, what don’t we know?

The Ambiguity/HUMINT methodology follows the real-world process. Identification of Conditions

16 August 2007 8 HUMINT brief to ISMOR

Identification of Conditions

• Operational Environment
• Commander’s Perceptions
• Sides, Factions, Allegiances
• Relationships between Sides
• - Cooperation Level
• - Reliability of Source
• - Quantity of Intel
• Collection behavior (ROE)

Identification of Conditions

Commander Influences Conditions Commander Develops/Updates CCIR

BC Planning & Asset Allocation

• Civil Military Operations
• Humanitarian Relief
• Psychological Operations
• Fire and Effects
• Coercion
• Security

Commander Evaluates Intel

Decision, Action & Evaluation

The methodology requires starting conditions that introduce ambiguous conditions and allow for the commander’s influence of those conditions.

Not portrayed in most models HUMINT-Related Information Items

• Intent
• Intent-What
• Intent-Where
• Intent-When
• Intent-Why
• Intent-How
• Intent-On Whom
• Intent-By Whom
• Allegiance/Side
• Activity
• Time
• Disposition
• Direction
• Speed
• Orientation
• Location
• Composition
• # Personnel
• # Equipment

16 August 2007 9 HUMINT brief to ISMOR

%

B R G P Y B

100 75 30 30

R

100 15 60 1

G

60 15 100 40 30

P

1 75 30 100 1

Y

1 1 25 1 100

Cooperation Matrix

Receiver of Intel Provider of Intel

This range of percentages are used by the SME to establish the start condition or initial Cooperation Matrix. Level Range Band

Always Cooperative

99.9 to 100% 81 – 99.8% 61 - 80% 41 - 60% 21 - 40% 1 - 20%

Never Cooperative

0 - .9%

Very Cooperative Cooperative Marginally Cooperative Uncooperative Very Uncooperative

Green provides 75% of its knowledge to Blue; while Blue only provides 60% to Green.

In the developed methodologies, the perceived cooperation levels between sides can be represented as percentages in a matrix.

Identification of Conditions

16 August 2007 10 HUMINT brief to ISMOR

BC Planning and Asset Allocation

Commander Develops/Updates CCIR Collection Manager Determines Intel Domain Collection Manager Determines Asset Collection Manager Tasks Collector Collection

• Signals
• Comms Intel (COMINT)
• Electronics Intel (ELINT)
• Imagery Intel (IMINT)
• Measurement & Signatures Intel (MASINT)
• Human Intel (HUMINT)

Staff Develops IR

HUMINT

HCT SFOD-A SMU LRSD Scouts

HCT

Identification of Conditions (Recognition of Ambiguity) The starting point of all intelligence collections are the CCIR that focus available intel assets (human and mechanical sensors) on those collections. Commander/Staff Determines Need For More Intel

Decision, Action & Evaluation

16 August 2007 11 HUMINT brief to ISMOR

Collection

Collector Executes Mission Commander Evaluates Intel

Decision, Action & Evaluation

Collection Manager Tasks Collector

Collection Team Behaviors

• Move to Area of Interest
• Terrain Reasoning
• Subject to Attrition
• Avoid hostile forces
• Conducts self-defense
• Disengage/break contact
• Time at Area of Interest
• Collect Intel

Collector Reports Findings

• Direct to Cdr
• To Intel Fusion

Collection Manager Determines Asset BC

Planning and Asset Allocation

Subject to mission requirements, collectors of intel behave in much the same manner as any typical maneuver element (e.g., subject to attrition, detection) and report their findings for analysis and evaluation.

16 August 2007 12 HUMINT brief to ISMOR

Commander/Staff Determines Need For More Intel Unit Conducts Opns Commander Evaluates Intel Commander Influences Conditions

Updating Perception of Conditions

Commander /Staff Conducts MDMP Conditions Change Collection Manager Determines Int BC

Planning & Asset Allocation

Collector Reports Findings Collection Once evaluated, the intel may give the Commander reason for action in the form of tactical commitments or operational influencers.

Decision – Action - Evaluation

16 August 2007 13 HUMINT brief to ISMOR

Potential Actions for an Example Intent

If the following information is known: Use Terrain as a Weapon

• Contaminate Food
• r Water
• Rubble Buildings
• Blow Dam
• Start Fire
• Generate Pollution
• Start Refugee

Movement

• Where
• By Whom or

How

• Where
• By Whom
• Location (By

Whom)

• By Whom

(PSYOPS)

• On Whom or

Where (CMO/HRO) The following minimal action can be taken: Action Taken: Secure the ‘Where’ Defend the ‘Where’ Attack the ‘By Whom’ Conduct Non- Kinetic Operation ‘By Whom’ or ‘On Whom’

The developed methodologies identify various actions to be taken depending upon the information gleaned.

The ‘By Whom’, in this example provides the commander with the size force to defend against or attack.

Decision, Action & Evaluation

16 August 2007 14 HUMINT brief to ISMOR

Summary & Status

• TRAC led the development of a first-of-its-kind methodology to provide the

M&S community a framework for introducing ambiguity and HUMINT collection. – Serves as the foundation from which methodologies can be effectively distilled and implemented across all M&S domains. – Provides avenue from which to evaluate the effectiveness of HUMINT collection in force-on-force models and could answer force structure questions regarding HUMINT collector needs and levels. – Informs doctrinal development (e.g., prior to this effort, no single document

• utlined the real-world process).
• TRAC and others in the M&S community are currently implementing the

methodology in both analytic and training models: – Advanced Warfighting Simulation (AWARS) – Spring 08. – Objective OneSAF – Spiral 1 demo Oct 07. – JLCCTC – Spiral 1 validation Aug 08.

• Joint Non-kinetic Effects Model (JNEM)
• WARSIM Intelligence Module (WIM)
• Maneuver models (WARSIM/CBS/JCATS)

16 August 2007 15 HUMINT brief to ISMOR

Questions

16 August 2007 16 HUMINT brief to ISMOR

Information

• Briefer information:

– Mrs. Lynn Leath – Phone: (913)684-9137; DSN 552 – E-mail: lynn.leath@us.army.mil

• To obtain copy of Ambiguity/HUMINT report, contact:

– Mr. John Hughes – Phone: (913)684-7593 – E-mail: john.hughes4@us.army.mil.

16 August 2007 17 HUMINT brief to ISMOR

BACKUPS

16 August 2007 18 HUMINT brief to ISMOR

ELEMENTS OF AMBIGUITY AMBIGUITY RESOLUTION THRU OTHER INTs FUSION Refine HUMINT

HUMINT CONFIDENCE PERCEPTION

iCOP

MOTIVATING FACTORS

TRUTH Red Blue Others

HUMINT ROLES & DOCTRINE

HUMINT OPERATIONS

HUMINT RESOURCES

HUMAN SOURCE

DETERRING FACTORS

AMBIGUITY RESOLUTION THRU HUMINT CAUSES HUMINT C2 HOW/WHAT PASSED

COMMS CHANNELS

TIMELY Model Worthy? Model Change? Model Improvements? Data? PREPARE

MDMP

PLAN

Intent/CCIR/PIR

EXECUTE

CA/PSYOPS HUMANITARIAN

Effect on Force Level Outcomes INTELLIGENCE PROCESS OPERATIONS PROCESS AMBIGUITY HUMINT

To HUMINT

JIM ASSETS

Learning CCIR/PIR Actionable Intel HUMAN FACTORS

Other Assets SOF OPS

16 August 2007 19 HUMINT brief to ISMOR

Ambiguity Causes

• Poor comms.
• Deception (includes media deception).
• Decoys.
• Terrain.
• Environment (weather, clutter, etc.).
• Sensor error.
• Information (too little/too much).
• Dynamic sides - changing allegiances.
• Media.
• Cultural shading.

16 August 2007 20 HUMINT brief to ISMOR

Ambiguity Elements

• Threat information.

– Unit.

• Activity – digging in, mobile, etc.
• Disposition – location, orientation, posture, etc.
• Composition – size, personnel, equipment, etc.

– Morale. – Tactics. – Intent. – ROE.

• Politics, state.
• Sides and allegiances.

– Non-combatants. – Factions. – Para-military.

• Media.

16 August 2007 21 HUMINT brief to ISMOR

Gross and Specific Intents

• Conduct political / economic / social disruption.

– Plot assassination. – Kidnap or hold hostage. – Terrorize. – Blackmail. – Extort. – Conduct refugee movement / displacement. – Conduct strikes or demonstrations.

• Conduct tactical / operational maneuvers (e.g.

attack, defend, secure).

• Conduct irregular logistics.

– Establish serve and relocate cache network. – Establish serve and relocate safe houses. – Fund logistics efforts.

• Impact allegiance / manipulate culture.

– Conduct PSYOPS. – Conduct CMO. – Conduct HRO. – Coerce. – Provide security.

• Disrupt info / communications or

infrastructure. – Conduct jamming. – Conduct IO (CNA). – Disrupt power grid. – Interdict LOC / staging.

• Use Terrain as a weapon / obstacle.

– Contaminate water, food supply. – Rubble buildings. – Blow dams. – Start fire. – Pollute. – Conduct refugee movement / displacement.

• Bomb.
• Riot.
• Acquire weapons.

– Purchase weapons. – Manufacturing weapons. – Steal/hijacking weapons.

Scenarios may drive the addition of other intents. The major bullets are considered aggregate, the lesser bullets are for entity level models.

Identification of Conditions

16 August 2007 22 HUMINT brief to ISMOR

HUMINT Asset Collection Criteria

Permissive Environment Availability Mission Proximity to Source Permissive/ Proximity Priority Info Collection Inside BDE AOR Inside/outside AOR Info Collection & Tactical Mission 1 (Highest) 2

HCT

Full to semi Yes/No

*Special Team

(e.g. SFOD-A, SMU, LRSD & Scouts) All Yes/No

Criteria Collector

* Distinguishing characteristics of special collection teams include the associated equipment, personnel and capabilities. Assumes Special units are similar enough to represent as a whole.

Once the HUMINT domain is chosen in the M&S to collect the information attributes, then the developed methodologies identify the selection criteria for choosing the appropriate HUMINT asset to conduct the mission.

BC Planning and Asset Allocation

16 August 2007 23 HUMINT brief to ISMOR

The methodology provides two alternatives to represent the collection and transfer of HUMINT-related information from the source side to the collection

• team. The first alternative is mathematical in nature.

B G75 Y30

Green gains 30% (5 items) of the information items from Yellow

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 8 9 10 11 12

Blue gains 75% (12 items)

• f the information items

from Green

9 10 11 12

Information Items are 75% Reliable Information Items are 30% Reliable

2 7 13 17

Source Reliability Carries Forward

12 13 14 15 16 17 1 2 3 5 6 9 13 17 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 3 5 12 17

Blue gains 75% (4 items)

• f the information items

from Green about Yellow Items in gray are false.

Collection

Collection/Transfer of HUMINT Information

(Mathematical Alternative)

16 August 2007 24 HUMINT brief to ISMOR

The second alternative, ambiguity states (AS), simplifies the method of collecting/transferring HUMINT information through static look-up tables.

• Ambiguity states (AS):

– Each unit (or entity) has an AS based on its cooperation level with each

• ther unit (entity).

– The higher the cooperation level, the higher the state number (ground truth is AS5) and the less ambiguous.

• Information collected when units are in proximity:

– All (true) intent-related attributes from AS5 units. – Unknown or misleading ‘intent-what’ and other selected intent-related attributes from AS1-AS4 units. – True, time-based, non-intent-related attributes (e.g., composition, speed) regardless of AS.

• Example: Blue unit encounters

green unit that has low cooperation level (10% from coop matrix); Blue: – Cannot determine ‘intent-what.’ – Collects only the correct information regarding the ‘by-whom.’

Collection

Collection/Transfer of HUMINT Information

(“Ambiguity States” Alternative)

16 August 2007 25 HUMINT brief to ISMOR

Influencer Conducted by Change to

• Coop. %*

Positive (+) or Negative (-) Influence Red or Blue 10% 10% 10% 40% 30% 10%

* Percentages are default values and can be adjusted based on SME input at scenario generation. Factors are applied at the time of event (once every 24 hours for aggregate level or every 6 hours for entity level.

Red or Blue + Influencer to Influenced

• Influencer’s Opponent both as:

Receiver of Influenced & Provider to Influenced + Influencer to Influenced

• Influencer’s Opponent both as:

Receiver of Influenced & Provider to Influenced + Influencer to Influenced

• Influencer’s Opponent both as:

Receiver of Influenced & Provider to Influenced

• Influencer to Influenced

+ Red (Influencer) to Influenced

• Blue (Influencer’s Opponent) both as:

Receiver of Influenced & Provider to Influenced Red or Blue Combatant sides Red Only Blue Only + Blue to Influenced

• Red (Influencer’s Opponent) both as:

Receiver of Influenced & Provider to Influenced CMO HRO PSYOP F & E Coercion Security

Influencing Operations

The methodology identifies the positive or negative effects of the influencing operations.

Decision, Action & Evaluation

16 August 2007 26 HUMINT brief to ISMOR

Influencing Operations

(Impact on Perceived Cooperation Levels)

% B R G P Y B

100 75 30 30

R

100 15 60 1

G

60 15 100 40 30

P

1 75 30 100 1

Y

1 1 25 1 100

Baseline Cooperation Levels (Established by SME) + Influencer to Influenced (10% gain)

• Influencer’s Opponent both as:

Receiver of Influenced (10% loss) Provider to Influenced (10% loss) Example: Blue conducts HRO with Green Revised Cooperation Levels (after Influencing Operation)

%

B R G P Y B

100 75 82.5 30 30

R

100 15 13.5 60 1

G

60 15 13.5 100 40 30

P

1 75 30 100 1

Y

1 1 25 1 100

Influencer \ Receiver of Intel Influenced / Provider of Intel The methodology identifies the effects of the influencing

• perations on the perceived cooperation levels.

Decision, Action & Evaluation

16 August 2007 27 HUMINT brief to ISMOR

Ambiguity and Effects

Ambiguity Elements

• Threat Information
• Unit
• - Activity
• - Disposition
• - Composition
• - Morale
• Tactics
• Intent
• ROE
• Politics, State
• Sides, Allegiances
• Non-combatants
• Factions
• Para-military
• Unknowns
• Media
• Elements in red not

addressed with methodology

• Elements in box already

addressed in AWARS to some degree

HUMINT Effects on Ambiguity

• Comparison of perceived vs.

ground truth (what is known, what we could have known)

• # IEDs avoided/discovered
• Increase/decrease in casualties
• Allegiance changes (good and

bad)

• HVT identified earlier/later
• % HUMINT Info transferred to

COP

• # IR/CCIR satisfied
• Actionable HUMINT to BC (to

include mission changes due to HUMINT info)

• Amount of HUMINT info

corroborated by other INTs

• Increased/Decreased Situational

Awareness & Situational Understanding Operational Effects

• # Targets acquired
• # Targets destroyed
• % Fratricide
• # Casualties
• # HPT killed
• Impact on OPTEMPO
• Synchronization of Forces
• Expenditure of Resources
• Success/Failure of Mission
• Etc.

16 August 2007 28 HUMINT brief to ISMOR

Addressing FALSE Information

• For stochastic models: Use random seed to determine which

attributes are false before conducting collection algorithms.

• For deterministic models: Use pre-determined number sets,

which adjust after every use, to determine which attributes are false. Again, determine false attributes before conducting collection algorithms or applying ambiguity states.

IPT (SMEs) developed false information approach, though agreed not the best, it is a viable method for applying falseness to information attributes.

16 August 2007 29 HUMINT brief to ISMOR

Timeline for Collecting Information

CoRe Matrix Value (source as provider to collector as receiver) 0 to 20% 21 to 40% 41 to 60% 61 to 80% 81 to 100% Time Delay for collection activity 90 min 75 min 60 min 45 min 30min

USAIC says should allow for plus-ups based on:

• Cultural awareness, ice breakers (pleasantries) + 30 min
• Mission type; general vs. specific collection
• Intent type; details specific to intent can vary.