President's Desk
The OR/IT Interface
By Thomas Cook
INFORMS President
tom.cook@calebtech.com
The Edelman Prize has been described as the "Nobel Prize of Operations Research." Those of us who have been involved in the competition over the years recognize that without exception, the finalists in the competition created huge value for their organizations, and that the value was delivered by an enterprise system that, at its core, had an OR module or series of modules that created the vast majority of the value. What we sometimes don't think about is those prize-winning applications could not have been successfully implemented without the surrounding infrastructure that provides the data to fuel our models and the software that provides the user and system interfaces necessary for successful implementation and use.
Several years ago, I was part of a celebration hosted by SNCF (the French National Railroad) to mark the successful completion of a new passenger reservation system and revenue management system. There were more than 100 developers from Sabre who had worked for more than two years on the project at the celebration that was held in a beautiful chateau outside of Paris. After handing out plaques of appreciation to the software engineers on the reservation system project, the senior VP of SNCF, responsible for the project, handed plaques to the seven OR professionals who were responsible for designing and implementing the new revenue management system. "And now we want to thank those few people who made the entire project financially possible," he said. He meant that without "OR inside," the return on investment on the project would not have been sufficient to justify the huge capital expenditure. Obviously, without the IT infrastructure provided by the reservation system, revenue management would not have been possible.
Almost all Edelman finalists are characterized by the successful integration of good operations research and good software engineering. The question is, Why aren't there far more systems with "OR inside" competing for the Edelman Prize each year? Based on my experience, there are at least two major reasons that systems devoid of intelligent OR-based modules are being designed and implemented, creating a huge opportunity cost for our society:
• Lack of awareness. The vast majority of people developing application systems do not know that operations research exists and what our value proposition is. Those who do know we exist are often threatened by us, offended by our arrogance, or simply don't want to be bothered with the added complexity or risk we might bring to their project. There is an obvious lack of demand-pull for OR by the software development community and IT management in general.
• Failure of OR practitioners. Many OR practitioners do not even attempt to sell themselves and their discipline to the IT department within their organization. The IT department in an organization typically has better access to the end-user or client, has a much larger development budget, better access to senior management, and most importantly, owns the data environment. For these and other reasons, I think that an organization's IT department is the ideal sales channel for our discipline. Failure on the part of the OR practitioner to partner with his or her counterparts in IT is a major contributor to why there are far too many systems devoid of "OR inside."
If the lack of a healthy relationship between the OR and IT disciplines is a major reason that we are not achieving our potential, the obvious question is, What can be done to improve the interface between the two disciplines? One thing that can be done is, in fact, now nearing completion — developing a message and communication strategy for the Marketing of the Profession initiative that I have been discussing in this column throughout the year.
The INFORMS Marketing of the Profession Committee has identified the IT community in general and CIOs in particular as among the most important target audiences for the OR message. If we can alert those people who are designing and implementing system solutions to enterprise problems about the OR value proposition, we might be able to create some demand-pull for putting OR inside of future application systems. I am confident that the communication plan that will emerge out of our current INFORMS marketing initiative will have a number of good ideas for reaching and influencing the IT profession.
In addition to trying to create demand pull in the IT community, it is important that many OR practitioners change. The OR practitioner must first realize and appreciate what an organization's IT department brings to the table, quit competing with his or her IT counterparts, perhaps reduce his or her arrogance level, and realize that people in IT can be great partners in delivering real value to a shared client base.
The most successful decision-support systems that I have been involved with have all been characterized by a healthy partnership between the OR and IT people responsible for development.
OR/MS Today copyright © 2003 by the Institute for Operations Research and the Management Sciences. All rights reserved.
Lionheart Publishing, Inc.
506 Roswell Rd., Suite 220, Marietta, GA 30060 USA
Phone: 770-431-0867 | Fax: 770-432-6969
E-mail: lpi@lionhrtpub.com
URL: http://www.lionhrtpub.com
Web Site © Copyright 2003 by Lionheart Publishing, Inc. All rights reserved.