There is a time for some things, and a time for all things;

a time for great things, and a time for small things. ...

But all in good time.

-- Miguel de Cervantes, Don Quixote

 

Table of Contents

Abstract

Acknowledgements

Table of Contents

List of Figures

List of Tables

List of Symbols

Chapter 1 Introduction

1.1 Background

1.2 UNIFY -- An Overview

1.3 Requirements for Effective MRM

1.4 Claims and Contributions

1.5 Evaluation

1.6 Outline

Chapter 2 Related Work

2.1 MRM Applications

2.1.1 Multi-Resolution Graphical Modelling

2.1.2 Hierarchical Autonomous Agents

2.1.3 Blackboard Systems

2.1.4 Cache Coherence

2.1.5 Abstract Data Types and Object Inheritance

2.1.6 Views in Databases and Integrated Environments

2.1.7 Nested Climate Modelling

2.1.8 Integrated Molecular Modelling

2.1.9 Multi-Level Computer Games

2.1.10 Battlefield Simulations

2.1.11 MRM Applications Summary

2.2 Multi-Model Execution

2.2.1 Selective Viewing

2.2.2 Aggregation-Disaggregation

2.2.2.1 Full Disaggregation

2.2.2.2 Partial Disaggregation

2.2.2.3 Playboxes

2.2.2.4 Pseudo-Disaggregation

2.2.3 Variable Resolution Modelling

2.3 Maintaining Consistency among Concurrent Representations

2.4 Chapter Summary

Chapter 3 Foundation

3.1 Model

3.2 Interactions

3.3 Multi-models

3.3.1 Cross-model Relationships

3.3.2 Mapping Functions

3.3.3 Time-Steps

3.4 Evaluation

3.5 Assumptions and Rationale

3.6 Chapter Summary

Chapter 4 Fundamental Observations

4.1 Problems with Aggregation-Disaggregation

4.1.1 Mapping Inconsistency

4.1.2 Chain Disaggregation

4.1.3 Transition Latency

4.1.4 Thrashing

4.1.5 Network Flooding

4.1.6 Cross-Level Interactions

4.1.7 Summary of Problems with Aggregation-Disaggregation

4.2 Fundamental Observations

4.2.1 Fundamental Observation 1

4.2.2 Fundamental Observation 2

4.2.3 Fundamental Observation 3

4.2.4 Fundamental Observation 4

4.3 Chapter Summary

Chapter 5 Multiple Representation Entities

5.1 Description of an MRE

5.2 Challenges

5.3 Rationale

5.4 Execution of an MRE

5.4.1 Maintaining Consistency

5.4.1.1 Temporal Consistency

5.4.1.2 Mapping Consistency

5.4.2 Resolving Concurrent Interactions

5.4.3 Storing Attributes in a Core

5.4.4 Comparing against Alternative Approaches

5.4.4.1 Comparing against aggregation-disaggregation

5.4.4.2 Comparing against selective viewing

5.5 Benefits of MREs

5.6 Limitations of MREs

5.7 Chapter Summary

Chapter 6 Consistency Enforcers

6.1 Constructing an Attribute Dependency Graph

6.1.1 Assigning Nodes to Attributes

6.1.2 Assigning Arcs to Dependencies

6.1.3 Assigning Semantics to Dependencies

6.1.3.1 Cumulative and Distributive Dependencies

6.1.3.2 Interaction and Modelling Dependencies

6.1.3.3 Selecting Dependencies

6.1.3.4 Properties of Dependency Classes

6.1.3.5 Examples of Dependency Classes

6.1.3.6 Dependency Weights

6.1.3.7 Interaction Semantics

6.1.4 Summary of Attribute Dependency Graphs

6.2 Selecting Mapping Functions

6.3 Traversing an ADG

6.3.1 Algorithm for Traversing an ADG

6.3.2 Cyclic Dependencies

6.3.3 Unplanned Dependencies

6.3.4 Traversal Path

6.4 Possible Implementations of a Consistency Enforcer

6.4.1 As-Is

6.4.2 Spreadsheets

6.4.3 Attribute Grammars

6.4.4 Mediators

6.4.5 Constraint Solvers

6.5 Chapter Summary

Chapter 7 Interaction Resolvers

7.1 Interactions

7.2 Serialization

7.3 Abandoning Isolation

7.4 Switches -- A Simple System

7.4.1 Unconstrained System

7.4.2 Constrained System

7.4.3 Dependent Concurrent Interactions

7.4.4 Complexity

7.5 A Taxonomy of Interactions

7.5.1 Properties of a Taxonomy of Interactions

7.5.2 Interaction Characteristics and Classes

7.5.2.1 Request and Response

7.5.2.2 Certain and Uncertain

7.5.2.3 Combining Characteristics

7.5.3 Evaluating the Taxonomy

7.5.4 Resolving Effects of Concurrent Interactions

7.5.5 Policies for Resolving Effects of Interactions

7.6 Constructing an Interaction Resolver

7.6.1 Operation of an IR

7.6.2 An Example IR

7.7 Chapter Summary

Chapter 8 Applying UNIFY : A Process

8.1 Guidelines for MRM Designers

8.2 Using UNIFY with a Specification Methodology

8.3 Process for Effective MRM

Chapter 9 Evaluation

9.1 Evaluating UNIFY in terms of MRM Requirements

9.1.1 Multi-Representation Interaction

9.1.2 Multi-Representation Consistency

9.1.3 Cost-Effectiveness

9.1.3.1 Assumptions

9.1.3.2 Consistency Cost

9.1.3.3 Simulation cost

9.1.3.4 Expected Costs

9.1.3.5 Experimental Costs

9.1.3.6 Summary of Cost-Effectiveness

9.1.4 Summary of Evaluation in Terms of MRM Requirements

9.2 Applying UNIFY to Existing Models

9.2.1 Military Models

9.2.2 Autonomous Agent Model

9.3 Limitations

9.4 Chapter Summary

Chapter 10 Conclusions

10.1 Contributions

10.2 Future Work

Appendix A Examples of Multiple Representations

Appendix B Joint Task Force Prototype

Appendix C Joint Precision Strike Demonstration

Appendix D Real-time Platform Reference

Appendix E Hierarchical Autonomous Agents

Indexed Glossary

References