Guest Post: Why we need decision modeling and simulation

---

In my last post, I introduced two problems in evidence-based decision-making. Here they are in a little more detail, along with the beginning of a solution.

Problem 1: How can decision makers identify the data they need to support their decisions?

Decision makers need to know how the outcomes they are trying to achieve depend on the actions they can take. But data seldom directly relates actions to outcomes. On the other hand, analysts and data scientists often have a wealth of decision assets—datasets and models—indirectly related to actions and outcomes, data that can profoundly inform decision making. But how does the decision maker know (a) what this data is and (b) how it relates to the decision they need to make?

But outside of decision intelligence, there is no systematic method for using the statement of a decision to derive requirements and specifications for the data science and analytic work required to support that decision.

Problem 2: How can we use these decision assets to inform our understanding of how decision outcomes depend on the actions taken by the decision maker?

Even when accurate and thoughtfully prepared information is provided to decision makers, they often cannot put it to effective use. This is because decisions are often based on the answer to the question “If I take this course of action, what will the outcome be?” Unless there is a dataset or model that maps action to outcomes, decision assets provided to support decision making do not address the decision maker’s most important question.

Decision support information seldom shows “decision outcomes” vs. “decision actions”, and there is no systematic method for transforming available decision assets to inform this relationship

Solutions

Problem 1 and Problem 2 have solutions that are provided by processes and tools that should be considered by every organization that aspires to making evidence-based or data-driven decisions.

1. Decision Modeling, and
2. Simulation.

Decision Modeling produces Decision Diagrams, like the one above. These are the “blueprints” or visual representations of Decision Models, bridging the gap between decision makers and the analysts and data scientists who support them. These diagrams make explicit the dependency chains that link decision maker’s actions and external factors to the outcomes they produce. Unknown values or relationships in these chains identify exactly the places where knowledge, data science, or other modeling work is required, precisely define what is required, and provide a framework for using the results of data work when it is completed.

Simulations use the Decision Models to create a data set that makes explicit the relationship between the actions available to the decision maker, and the outcomes that are likely to result. This gives the decision maker the ability to explore the consequences of the courses of action available to them, including:

• Which actions lead to favorable, or optimal outcomes?
• How do external factors over which the decision maker has no control affect the outcomes?
• For each course of action, what is the risk of various adverse outcomes occurring?
• How does uncertainty in various sources of information captured in the Decision Model propagate into uncertainty in the outcomes?
• How volatile are the outcomes, that is, how sensitive are they to small changes in the various inputs into the Decision Model.

In my next posts, I’ll show you a case study on Decision Modeling, explain how Decision Intelligence differs from Business Intelligence and why we need both, and show you how decision simulation addresses problem 2, above.

---