Lecture B3 (2022-01-27): Chapter 2, Introduction to Feedback Systems Thinking (Morecroft, 2015)

In this lecture, we cover chapter 2 from Morecroft (2015), which introduces feedback systems thinking -- a shift of mind away from an event-oriented perspective to a more endogenous viewpoint that takes into account the (often long-term) side effects of solutions as they feed back into future problems. This discussion is motivated by a traffic congestion example. The discussion provides the opportunity to discuss Causal Loop Diagrams (CLD's) and the conventions/rules for drawing them that make them useful for either finding systems archetypes that predict behavior or coming up with dynamic hypotheses to test later with simulation. That provides an opportunity to bring up Behavior Over Time (BOT) plots and how to use them in real-world data as compared to simulation data to test and refine plausible causal hypotheses about dynamic behavior.

Lecture B2 (2022-01-25): Causal Loop Diagrams in Vensim

In this lecture, we introduce (via hands-on tutorial) the program Vensim PLE by Ventana Systems and how to use it to draw Causal Loop Diagrams (CLDs). We then review how to draw and edit CLD's in general and introduce a homework assignment where students will have to draw and annotate complex CLD's in Vensim.

Lecture B1 (2022-01-20): Introduction to Causal Loop Diagrams

In this lecture, we pivot from our introduction to system dynamics modeling in general to discuss the broader endogenous perspective, where macroscopic observables of a system are a function of internal, not external, causes. That motivates the rest of the lecture, which introduces the "Causal Loop Diagram" (CLD). A CLD includes a network of links (representing positive or negative causal relationships) that may form loops (that are positive or negative). We can use CLD's to build stock-and-flow diagrams (dynamical systems models) or to gain insights into the trajectories produced by existing stock-and-flow diagrams.

Lecture A3 (2022-01-18): Chapter 1, The Appeal and Power of Strategic Modeling (Morecroft, 2015)

This lecture discusses topics from Chapter 1 of the course textbook by Morecroft (2015). The focus of the chapter is an introduction to strategic modeling, particularly in the case of strategic system dynamics models. We initially discuss the modeling spectrum -- from analog to metaphorical -- and when it is beneficial to leave aspects out of models. We then pivot to discussing Forrester's Limits to Growth/World Dynamics model and its limitations. We then finish by showing how a simple dynamical systems model of a fishery can lead to actionable suggestions about regulation that can help restore a fishery to sustainability.

Lecture A2 (2022-01-13): Introduction to Modeling

In this lecture, we dig more deeply into the definition of a model and the types of models. We focus on how anything can be used as a model so long as it is being used to answer a "What If" question. In other words, models are defined by how they are used and not what they are made of. When then discuss the differences between mental and quantitative models (pros and cons of each). We finish with a description of the different ways that quantitative models are used, the different types of quantitative models, and the different types of simulation.

Lecture A1 (2022-01-11): Course Introduction

In this first lecture of the semester, we discuss the course policies, motivation, and topic outline. We then start to discuss modeling in general -- leading up to a simple definition of a "model" as a tool for answering a "What If" question. The unavoidable modeling degrees of freedom may be open for misuse, even inadvertent. That said, "reality" is not the goal of modeling. All models are wrong, but good models (even if wrong) are crafted to be illuminating and useful; their lack of reality can actually make them more generalizable.