Forum
Scientists Tailor-Made for Today's World
By E. Andrew Boyd
Rumor is that Las Vegas won't have them any more. As a group they don't gamble enough. When they do gamble, they don't lose enough money. When they play Blackjack, one of the few games where a player can actually turn the odds against the house, pit bosses watch warily for any of the telltale signs that they're mercilessly taking advantage of an irregular run of cards.
The group is INFORMS, the Institute for Operations Research and the Management Sciences. Not every city is capable of handling their annual meeting, which hosts an average of 3,000 attendees. The problem isn't just the number of people but the fact that most of them will take the podium at one time or another. Over a period of four days during the 2005 meeting, 16 time slots were set aside during which 50 simultaneous sessions took place, each session hosting an average of three to four speakers.
What do these people talk about when they get together? A sample of the presentations gives some hint: "Mathematical modeling of dynamic breast screening policies," "A staffing decision methodology for TSA security checkpoints at airports," "Capacitated production planning with price-sensitive demand and general concave revenue functions," and "Robust Wardrop equilibria." Topics range from the exceedingly practical to the intensely theoretical.
With such a wide variety of interests, subdivisions have formed. In all there are 37, including subdivisions devoted to aviation, e-business, health, railroads, telecommunications, manufacturing, probability, optimization, the military and even sports. One of the fastest growing subdivisions is the Revenue Management and Pricing Section. Spawned by the early success of the airline industry, the section serves as home not only for airline revenue management but for the many pricing applications sprouting up in new industries.
What binds this eclectic group is a common tie to operations research, or simply O.R. for short. Arguably the worst name ever adopted by a group of professionals, it's not exactly about operations or research. The fact is the discipline has never been able to escape its history and come up with a name that evokes a sense of what it's about. "Philosopher," "electrical engineer," "economist" and most other professional titles conjure up an immediate sense of what a person does. "Operations researcher" doesn't even slip easily off the tongue. The name management science is sometimes used, but is considered too restrictive by many since the discipline's focus isn't limited to managerial problems. And as a marketing firm charged with raising the visibility of INFORMS recently concluded, "operations research" carries more brand equity than "management science."
The name is only the beginning of the discipline's difficulties. Describing what operations researchers do is equally difficult. On one hand, operations researchers are fundamentally problem solvers. They like to take complex problems, reduce them to their basic components and come up with optimal solutions. Unfortunately, while this description is fundamentally correct, it's far too vague. Anyone who works or runs a household for a living would likely describe himself as a problem solver, yet very few actually practice operations research.
Sometimes operations researchers avoid describing what they do and simply point to the results of their work. When checking in at the airport, renting a car or mailing a package at the post office, we usually stand in a single long line and go to the first available server when we reach the front. This wasn't always the case. Lines used to form behind each individual server as they still do at grocery stores. The single line reduces the average waiting time per customer substantially, and was due to the work of operations researchers. The problem with such descriptions is that they don't convey any sense of what lies in the background. A listener may well walk away with a sense that operations researchers are efficiency experts and, as I was once informed, do work that is "dangerously close to common sense."
The reality is that operations research is a form of mathematics. While problem solving is a vast topic, it turns out that the mathematical tools employed to solve most quantitative business problems aren't as numerous as might be imagined. Random events are an important part of many problems, whether they correspond to the length of time people spend waiting in line, the number of casualties sustained in a military action or the number of people buying plane tickets. In the parlance of mathematics, these types of random events fall under the heading of stochastic processes, and comprise one of the two major branches of operations research.
The other major branch is optimization, which deals with finding the best solution to well-defined problems. A good example is that of finding the shortest route between any two points in a city. Given a list of the streets and their length, there are optimization methods that can find the shortest route between any two points in a mere fraction of a second. If a computer actually had to try all the different routes and compare them, then for any city of even moderate size today's computers would take longer solving the problem than the estimated age of the universe.
Operations researchers receive formal training in both optimization and stochastic processes, normally including some training in statistics, forecasting and computer simulation of complex systems. They also receive training on how to use these tools to solve real problems.
The fact that a common set of mathematical tools can be used to address such a wide variety of practical problems is nothing short of remarkable. Abstraction is the key, recognizing that orders arriving at a warehouse and people calling to purchase airline tickets can be described in fundamentally the same way. While the actual business problems are quite different, when abstracted they can be studied together.
Yet the efficiency provided by abstraction is one of the greatest challenges facing the discipline. A truck dispatcher for a delivery company isn't likely to see the connection of his work with designing printed circuit boards or doctors using radiation to destroy tumors. There should be people who study truck dispatching, and they should belong to a truck dispatching professional society, just as printed circuit board designers and radiation treatment professionals should belong to their own societies. To everyone who isn't an operations researcher, problems are defined by problems, not a common, abstract mathematics. This is why most operations researchers develop a close affiliation with a profession other than operations research. More than a few leave the discipline behind. Still, many cling to the mathematical foundation that helped draw them into operations research to begin with, and this is why they return year after year to a common meeting place.
The allure of mathematics, however, is a two-edged sword. While mathematics attracts people to the discipline of operations research, it also creates a barrier between those who develop mathematical tools for activities like pricing and those who actually use the tools. Operations research is tailor-made for today's world, a world filled with data and technical challenges. But getting industry to accept operations research requires communication in non-mathematical terms; it requires communicating what the discipline is and, at a deeper and more important level, communicating the discipline's value to a world that isn't mathematical. The extent to which scientific pricing will be embraced by industry depends on operations researchers' ability to tackle the many non-mathematical pricing challenges they face with the same passion they devote to the mathematics of pricing.
|
E. Andrew Boyd is chief scientist and senior vice president of Science and Research at PROS, a revenue management and pricing software firm. He received his Ph.D. in operations research from MIT in 1987.
Source:
-
E. Andrew Boyd, "The Future of Pricing: How Airline Ticket Pricing Has Inspired a Revolution," 2007, Palgrave Macmillan, reproduced with permission of Palgrave Macmillan.
OR/MS Today copyright © 2007 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 2007 by Lionheart Publishing, Inc. All rights reserved.