Prof. Retsef Levi of MIT
Putting math to work for efficiency’s sake
People behind the science
As U.S. hospitals grapple with how to keep down costs in the face of insufficient federal reimbursements, two prominent American hospitals have turned to operations research, analytics, and operations management for answers.
Prof. Retsef Levi from MIT’s Sloan School of Management - who just completed a year as a Feinberg Foundation Visiting Fellow at the Weizmann Institute’s Department of Computer Science and Applied Mathematics - is spearheading major projects at Massachusetts General Hospital and Beth Israel Deaconess Medical Center in Boston that are changing their system design and streamlining everyday operations to improve patient care and make it more efficient.
Prof. Levi’s work at these two Harvard University teaching hospitals addresses core system design challenges at those places and is potentially applicable to many other similar centers throughout the U.S. More broadly, his research addresses far-ranging real-life problems faced by companies, organizations, and societies concerning supply chains, resource allocation, and the interplay between optimization and learning. At the Weizmann Institute, he took part in collaborations involving operations research and computer science that aim to introduce new methods in operational efficiency with Prof. Robert Krauthgamer, postdoctoral fellow Dr. Michael Dinitz (now a faculty member at John Hopkins University), and Chen Attias, an MSc student in the Department of Computer Science and Applied Mathematics.
The surgical 'supply chain'
In the past eight years, Prof. Levi and his MIT colleagues conducted a system-level analysis at Mass General of surgical patient flow, and the hospital has implemented several of the study’s recommendations. As a result, wait times for patients have been shortened, more beds have been freed up, and the foundation has been laid for better operating room utilization. About a third of surgeons changed their regular surgery day to improve patient flow and resource utilization through the week. The overall outcome of this collaboration is a creative redesign of the hospital’s surgical ’supply chain’, which has saved the hospital over $1 million annually, improved patients’ access to care, and pertinently demonstrates that improving the quality of health care can actually go hand-in-hand with lowering costs.
Prof. Levi is now developing an improved approach for intensive care unit safety and risk management at the hospital. “I believe we are going to redefine the way in which safety and risks are being perceived in the healthcare environment, moving from focusing on ‘safety events’, which require us to enumerate all the possible mishaps - a feat that can never be truly exhaustive - to the consideration of ‘risky states’. Risky states are fundamental conditions of the environment, the system, and the people in that system that might increase the likelihood of harm to occur and the magnitude of the risks.”
Prof. Levi and his team at MIT are creating statistical and learning models that will indicate which states really give rise to increased risks, and will later use these to formulate practical measures. “I believe this is a more fundamental approach to tackling a system-level problem because rather than taking care of individual events, you are changing the quality of the system and therefore its properties in a way that mitigates all harm rather than one specific harm," he says.
Prof. Levi has also worked on developing new models and algorithms to address core challenges of the U.S. Air Force in the areas of logistics and supply chain management. In Israel, he has been working with multiple industrial and governmental organizations, including developing a new distribution model for the Yedioth Media Group (publisher of Yedioth Ahronoth newspaper), which was implemented successfully. That project was awarded the Daniel H. Wagner Prize for Excellence in Operations Research Practice.
His highly applicative field of research makes him an unlikely visiting scientist on the Weizmann Institute campus. But his deep interest in the theoretical questions that arise from modeling complex problems provides common ground with Prof. Krauthgamer and others. Their joint effort, together with a postdoctoral fellow, brings together two timely topics in operations research and computer science: investigating process and supply-chain design flexibility, where the goal is to design supply chains that can quickly adapt a set of products to variable demands. Their project leverages modern techniques from the areas of network flows, graph expanders, and efficient algorithms and optimization, to develop new theories and practical methods to design robust supply chains.
Born and raised in Kfar Saba, Prof. Levi took an unusual route into academia, reflecting a family tradition of creativity and achievement; his brother Reshef is a well-known comedian and filmmaker and another brother, Yannets, is the author of a popular children’s book series. Like many of his Weizmann Institute counterparts, Prof. Levi’s path has always been directed by a deeply rooted curiosity and wish to expand intellectually. “I was an intelligence officer in the army for almost 12 years. Only around the age of 29, without so much as an academic degree, I started my studies, not guided by a clear plan but rather by what stimulated my interest,” he says. He focused on mathematics at Israel’s Tel Aviv University, earning a Bachelor’s degree, summa cum laude, in 2001, and then a PhD in Operations Research from Cornell University in 2005.
At the Weizmann Institute, Prof. Levi was joined by his wife Anat - a professional medical translator and interpreter at Boston hospitals - and their six children. He says he tremendously enjoyed working and living at the Institute. “The Institute was very generous to provide world-class logistical support to allow the smoothest transition of my family to Israel,” he says. “The intellectual environment is very stimulating and teaching and interacting with the students was a delight. I really hope to maintain long-term relations with individuals and the Institute.”