POLICY MAKING & MODELING IN A COMPLEX WORLD

profiletestwork
ITS832_Chapter_4.pdf

ITS 832 Chapter 4

Policy Making and Modeling in a Complex World

Information Technology in a Global Economy

Professor Miguel Buleje

Introduction

• Policy Making and Modeling in a ComplexWorld

• Complexity

• Managing Complex Systems

• Modelling for Complex Systems

Complexity

• System composed of multiple interacting elements • Possible behavioral states can combine in ways that are hard to

predict

• Many complex systems in the physical world

• Adaptive capacity of organisms allow for long-term survival in complex systems

• Complex Adaptive Systems (CAS)

• Strong capacity to self-organize

Double Pendulum Example

Common Mistakes in Managing Complex Systems

• Quantification • Policy is biased towards Quantifiable Variables.

• Most often, monetary quantification.

• Most often, solution selected would be favorable to the optimal economic outcome.

• Commonly overlooks important non-quantifiable aspects

• Compartmentalization • Second response by policymakers in trying to simplify complex system.

• Attempts to simplify complex social systems

• Large systems are split into smaller systems

• High RISK to miss interactions between smaller systems

• Spillover effect

Complexity in Policy Making

• Common approaches • Instrumental

• Choosing between a set of possible policies

• Evaluated based on past effectiveness

• Requires • Large enough pool of available strategies

• Effective assessment of effectiveness

• Representational • Little more complication approach

• Series of models

• Each is assessed on its ability to predict observed behavior

Instrumental Approach

Representational Approach

Agent-based Simulation Models • Agent-based simulation is presenting an optimal approach to address

issue around complexity and policymaking. • Agent base models:

• Represents individuals as separate computer models • Each model captures the behavior by each individual

• Agents interact through a network

• Distributed nature allows for realistic interactions, and makes this an attractive alternative.

• SIMSOC

• Simulated Society: use by universities and other groups to teach social sciences.

• Large modeling projects repository, per inventory on the SIMSOC, in our space of policy making.

Summary

• Complex systems are difficult to model • Interactions can be unpredictable • Common mistakes in modeling complex systems

• Quantification • Compartmentalization

• Two common approaches to complex system modeling • Instrumental • Representational

• Agent-based modeling