The INFORMS Transportation Science and Logistics Society recently recognized Warren Powell as the 2021 recipient of the Robert Herman Lifetime Achievement Award. The award recognizes fundamental and sustained contributions to transportation science and logistics and has influenced the field through her or his writings, teaching, service, and nurturing of young professionals. The transportation society is a subdivision of INFORMS, the Institute for Operations Research and the Management Sciences.
Warren Powell taught in the Department of Operations Research and Financial Engineering at Princeton University from 1981 to 2020. He specialized in models and algorithms for decisions under uncertainty, motivated by problems in freight transportation and logistics (spanning truckload and less-than-truckload trucking, rail and supply chain management), as well as energy, health, finance, e-commerce, and other applications.
He wrote the first interactive optimization tool for network design for LTL trucking, SuperSPIN, which was adopted by the entire LTL industry. His lab produced the first commercially successful real-time load matching tool for truckload trucking and the first successful optimization model for locomotive management (at Norfolk Southern Railway).
He also established and directed PENSA (Princeton Laboratory for Energy Systems Analysis), which produced an array of energy models, spanning large scale grid models for simulating high penetrations of wind and solar, as well as a large library of energy storage models which produced new insights into general inventory problems.
"We could not be more proud of our friend and colleague (and father!)," said Daniel Powell, Co-Founder and CEO of Optimal Dynamics. "Warren's work in transportation science and logistics is quite literally the bedrock of our organization, as we have built our platform off of his research. Perhaps more importantly, we strive every day to build products that are an extension of his work and that continue to solve deep problems."
He has written three books. The first, Approximate Dynamic Programming: Solving the curses of dimensionality, was the first to present models and algorithms for the high-dimensional problems that arise in complex resource allocation problems such as truckload trucking.
The second, Optimal Learning (coauthored with Ilya Ryzhov), addresses the problem of the efficient collection of information, a problem that arises in both field and laboratory experiments, spanning recommending offers to customers to finding the best price or the best supplier for a product.
The third, Reinforcement Learning and Stochastic Optimization: A unified framework for sequential decisions, is the culmination of a career thinking about a wide range of sequential decision problems under uncertainty. The book presents a universal framework for modeling and solving any problem involving the sequencing of decisions and information, ranging from sequential experimentation up to the management of fleets of trucks or supply chains. This book lays the foundation for a new field he is calling sequential decision analytics that provides practical, scalable tools for strategic, tactical, and real-time planning of a wide range of sequential decision problems.
Based on 40 years of academic research, Optimal Dynamics delivers a new class of decision and optimization technology to the transportation and logistics industries. Optimal Dynamics CORE.ai platform augments decision-making for carriers and shippers to automate and optimize strategic, load acceptance and dispatching decisions. CORE.ai has an easy-to-use visual interface that integrates with data software to drive more resilient and self-healing operations. Learn more at www.OptimalDynamics.com.