Menu
Warren B. Powell
Warren B. Powell

Warren B. Powell

Professor emeritus, Chief Innovation Officer, Executive-in-residence
Princeton University, Optimal Dynamics, Rutgers Business School
Bio

Warren B Powell is Professor Emeritus at Princeton University, where he taught for 39 years, and is currently a co-founder and Chief Innovation Officer at Optimal Dynamics as well as Executive-in-Residence at Rutgers Business School.   He was the founder and director of CASTLE Lab, which focused on developing optimization models for a wide range of applications (freight transportation, energy, health, manufacturing, …) that require making decisions under uncertainty.  These experiences led to his work on sequential decision problems and the universal modeling framework.  This is documented in his latest book with Wiley: Reinforcement Learning and Stochastic Optimization: A unified framework for sequential decisions (2022). He published over 250 papers, five books, and produced 60 graduate students and post-docs.  He is the 2021 recipient of the Robert Herman Lifetime Achievement Award from the Informs Society for Transportation Science and Logistics, and the 2022 Saul Gass Expository Writing Award.  He is a fellow of Informs, and the recipient of numerous other awards.


If You Want to Run a Better {Anything} You Have to Make Better Decisions: The Universal Modeling Framework for Analytical Thinking

Modern optimization tools offer tremendous capabilities for making decisions. However, we often fall into the trap of choosing problems that fit into these modeling frameworks which only work for a tiny fraction of business applications.  My goal is to promote analytical thinking, regardless of whether we eventually use a model. Modeling always starts in English, but you have to ask the right questions.

My Universal Modeling Framework can be used to represent any decision problem.  It addresses any type of decision as they are made over time.  It starts by identifying three dimensions that arise in all applications: metrics, decisions, and uncertainties. This is the starting point for any model, but their identification helps even if a model is never used.

I next help identify the three dimensions.  Metrics should be organized in pyramids, identified as objectives, targets or limits. I offer the first practical definition of risk that applies to all situations. I identify six types of decisions that cover any application.  I provide 12 classes of uncertainty that guide the process of identifying all sources of uncertainty and describe the different behaviors of uncertainty.

Finally, I will address the problem of how to make decisions. Rather than running to some sophisticated tool, I argue that any method for making decisions (known as “policies”) fall into one of four classes, including any method already used in practice.  Policies are characterized by six attributes, covering how well they work, along with flexibility, transparency, computational speed and data requirements.

Essential / Professional / Leadership
Back To Top

The Institute for Operations Research and the Management Sciences

5521 Research Park Drive, Suite 200
Catonsville, MD 21228 USA

phone 1 443-757-3500

phone 2 800-4INFORMS (800-446-3676)

fax 443-757-3515

email

Get the Latest Updates

Copyright INFORMS. All Rights Reserved