Decision Making

ITS 832 Chapter 5

From Building a Model to Adaptive Robust Decision Making Using Systems Modeling

Information Technology in a Global Economy

Professor Michael Solomon

Introduction

• Systems modeling • Focus on decision making abilities

• Legacy System Dynamics (SD) modeling • Recent innovations • What the future holds • Examples

Systems modeling

• Dynamic complexity • Behavior evolves over time

• Modeling methods • System Dynamics (CD) • Discrete Event Simulation (DES) • Multi-actor Systems Modeling (MAS) • Agent-based Modeling (ABM) • Complex Adaptive Systems Modeling (CAS)

• Enhanced computing supports model based decision making • Modeling and simulation has become interdisciplinary

• Operation research, policy analysis, data analytics, machine learning, computer science

Legacy System Dynamics Modeling

• 1950s – Jay W. Forrester • Primary characteristics

• Feedback effects – dependent on their own past

• Accumulation effects – building up intangibles

• Behavior of a system is explained • Casual theory – model generates dynamic behavior

• Works well when • Complex system responds to feedback and accumulation

Recent Innovations

• Detailed list of individual innovations • Deep uncertainty

• Analysts do not know or cannot agree on • Model

• Probability distributions of key features

• Value of alternative outcomes

• Two primary evolutions • Smarter methods (Data Science)

• Usability/accessibility advances

What the Future Holds

• Better models • More data (“Big Data”) • Social media • Advanced capabilities for

• Hybrid modeling

• Simultaneous modeling

Modeling and Simulation

Examples

• Assessing the Risk, and Monitoring, of New Infectious Diseases

• Simple systems model with deep uncertainty

• Integrated Risk-Capability Analysis Under Deep Uncertainty

• System-of-systems approach

• Policing Under Deep Uncertainty • Smart model-based decision support system

Summary

• Modeling has long been used with complex systems • Recent evolutions have advanced modeling

• Increase computing power

• Social media and Big data

• Sophisticated analytics

• Multi-method and hybrid approaches are now feasible • Continued move into interdisciplinary study

• Advanced modeling for complex systems