1 of 13

Towards cross-validation of C2-harmonisation Hypotheses: comparing ELICIT simulation and Network Synchronisation modelling approaches to Command and Control

Mitchell Kiely (Presenter) & Alexander Kalloniatis

Defence Science and Technology Group

2 of 13

Overview: an introduction to a new model

  • How we construct mathematical models for the decision making process
  • The insight we gain from doing so
  • Limitations on existing models:
    • Constant network structure
    • Time-dependent structure with predetermined interactions
  • Our novel model:
    • State dependent organizational structure

​

​

​

3 of 13

ELICIT

  • Measures the behavior of a network in an information sharing scenario.
  • Agents are given a who/what/when/where problem that they must solve by sharing information.
  • Variable parameters:
    • Agent’s traits
    • C2 configuration

​

​

  • Allows for extensive testing of different scenarios (D Alberts 2011)

“Elicit – The experimental laboratory for investigating collaboration, Information-Sharing & Trust” M Ruddy (2007)

4 of 13

Network synchronization

  • Permits a unique insight into the synchronization of the decision makers

 

  • Static networks (A Kalloniatis, 2008)

​

  • Battle-Rhythm: Dynamic networks with predetermined structure

and pattern of interaction (A Kalloniatis, 2009)

​

5 of 13

Network synchronization

 

 

Individual decision making speed

 

R(t) measures how synchronized the decision makes are

The Kuramoto model:

Influence of connected individuals

 

 

 

 

 

 

6 of 13

NATO-143 SAS scenario

  • The scenario involves three entities; Executive cell (green), Operations cell (Orange), and a Communications cell (Blue).
  • Individuals within these entities interact with other individuals as well as external factoids.
  • Mission has four phases; Awareness, Planning, Execution, and Completion.
  • The network configuration varies depending on the stage of the mission

1)

2)

3)

Awareness Phase

Thank you Ken Teske and David Alberts for providing us with the data

7 of 13

Phase Dependent Kuramoto model

 

  • The state of the decision maker

​

​

 

  • The decision making speed of the individual

​

​

 

 

  • The tightness of the organization

​

​

 

  • The dynamical organizational structure

​

​

 

  • The type of interaction between decision makers

​

​

  • The influence of information

​

​

 

 

  • A restriction that limits the progression of individuals

​

​

8 of 13

Results

 

Varying decision making speed

 

 

High harmonization

Low harmonization

Mission complete

 

 

 

 

 

Varying interaction strength

Varying importance of information

 

 

9 of 13

Results

Average result over many simulations

 

 

 

 

 

 

 

10 of 13

Future works

  • An extensive sweep of all parameters
  • Application of algorithm to experimental data
  • Engaging with ELICIT to work towards a cross-validation of results

11 of 13

Q&A

Mitchell.kiely@dst.defence.gov.au

12 of 13

2)

3)

4)

5)

Phase-dependent Kuramoto model

1)

The algorithm:

1. Find the smallest Ɵ

​

 

 

 

5. Iterate

 

13 of 13

Results

Varying coupling strength

Varying factoid influence

Weak coupling strength -> low harmonization

Strong coupling strength -> high harmonization

Negligible factoid influence -> high harmonisation

Large factoid influence -> low harmonization