INFORMS Annual Meeting Charlotte 2011: Plenaries and Keynotes

The Evolution of Analytics

Presented by: Keith Collins, Senior Vice President and Chief Technology Officer, SAS

Disciplines like operations research, statistics, forecasting and econometrics have distinct roots and continue to develop in their own right - in research and academia. In practice it's increasingly recognized that they have grown together, producing a hybrid known as "business analytics." INFORMS has endorsed and embraced this trend through its actions and in its mission statement. What is propelling the business analytics trend, what problems does it seek to address, and where is it going? Business analytics have evolved steadily over the past several decades, adapting to address changing needs and incorporate new technologies. Collins will step listeners back in time to better understand where we are now. As computational advances and the corresponding data deluge present ongoing challenges, Collins will lead us toward the future with insights into the tremendous opportunities these changes herald for all of us.

Build, Test, Scale: Innovative Strategies that are Transforming the U.S. Healthcare System

Presented by: Susan D. DeVore, President and Chief Executive Officer, Premier healthcare alliance

Susan DeVore is President and CEO of the Premier healthcare alliance, the nation's leading alliance of hospitals, health systems and other providers dedicated to improving healthcare performance. An alliance of more than 2,500 hospitals and health systems and more than 70,000 other healthcare delivery sites, Premier uses the power of collaboration to lead the transformation to high quality, cost-effective healthcare. Premier membership includes more than 40 percent of all U.S. health systems. With the ultimate goal of helping its members improve the health of their local communities, Premier builds, tests and scales models that improve quality, safety and cost of care. Through successful initiatives such as the Hospital Quality Incentive Demonstration with CMS, and QUEST: High Performing Hospitals collaborative, the alliance has driven improvements in evidence-based care and safety, as well as significant reductions in mortality, harm and cost. Premier is a leader in the accountable care movement and recently announced a joint-venture with IBM to develop industry-leading population analytics tools. Under DeVore's leadership, Premier has built an industry leading code of ethics, has been named four times as one of the World's Most Ethical Companies by Ethisphere and has won the Malcolm Baldrige National Quality Award.

Studies in Activity Networks: A Guided Tour

Presented by: Salah E. Elmaghraby, University Professor of Operations Research and Industrial Engineering, North Carolina State University

Activity Networks (ANs) is a model for the planning and (dynamic) control of projects. I shall present a historical perspective of studies of the model from its inception over 50 years ago, and the various issues with which the original model of CPM/PERT has been enriched. These include: portfolio selection, the management of resources (including funds), the acquisition and dispensation of resources, coping with uncertainty (the popular name is "risk management"), and the graphical representation of progress to satisfy certain managerial desires. Of the myriad of issues raised in the field of project management and control, I shall elaborate on three areas of my continued research:

• Assuming deterministic environment (Deterministic Activity Networks (DANs), can one resolve the "optimal crashing problem" under generalized precedence relations (GPRs)? How about the optimal scheduling of the activities subject to resource availabilities (the RCPSP under GPRs)?
• Still under the assumption of DANs, how to schedule the activities subject to limited resource availabilities (including how much to acquire of these resources in the first place) in order to achieve certain objectives.
• Assuming stochastic environment (Probabilistic Activity Networks (PANs), how to model, and cope with, uncertainty relative to specifying the start time of the various activities and the scheduling of the resources required for their execution.

Why Good Simulations Go Bad

Presented by: Barry L. Nelson, Walter P. Murphy Professor and Chair, Department of Industrial Engineering and Management Sciences, Northwestern University

Simulation is a powerful tool to design, evaluate and improve the kinds of systems and processes that concern operations researchers and management scientists. Far from its early role as the brute force method of last resort, simulation now supports decision making both routine and critical. The software, and knowledge about how to use it, are widespread. But simulation frequently involves a large commitment of time, effort and money, and users often do not get everything they paid for; even worse, the results they do get may be seriously misleading. This talk describes common ways that good simulations go bad and how to avoid (or at least recognize) them. No background in simulation is assumed, and lots of examples will be provided to support the technical points.

Barry L. Nelson is the Walter P. Murphy Professor and Chair of the Department of Industrial Engineering and Management Sciences at Northwestern University. He received his BA in mathematics and computer science from DePauw University and his MS and PhD in industrial engineering from Purdue University. Before coming to Northwestern in 1995 he was on the faculty at The Ohio State University. His research is on the design and analysis of computer simulation experiments on models of discrete-event, stochastic systems, with applications to manufacturing, services, finance and transportation. Nelson has published numerous papers and two books, including Discrete-Event System Simulation, 5th edition, which has been adopted by over 60 universities. He is a Fellow of INFORMS and IIE. In 2006 he received the Outstanding Simulation Publication Award from the INFORMS Simulation Society for his work on simulation optimization, and in 2007 and 2010 he was awarded the Best Paper-Operations Award from IIE Transactions. He has also received the Northwestern University Alumni Association Excellence in Teaching Award, and has twice been named McCormick Teacher of the Year in engineering at Northwestern.

OMEGA RHO, the official Honor Society of INFORMS, was founded in 1976 to recognize superior scholarship and encourage leadership in operations research, management science, and related disciplines. The society has 39 active collegiate chapters, more than 5,000 student and faculty members and is a member of the Association of College Honor Societies. In addition to sponsoring OMEGA RHO Distinguished Lectures at INFORMS Annual and International meetings, OMEGA RHO provides financial support to the annual INFORMS Colloquium. Honorary Membership in OMEGA RHO is bestowed upon individuals who provide leadership and extraordinary support for the encouragement of operations research and management science through their professional activities. Prior to delivering the OMEGA RHO Distinguished Lecture, Barry L. Nelson will be inducted as the 45th Honorary Member of OMEGA RHO. Visit OMEGA RHO online: http://omegarho.informs.org