OR/MS Today - October 2002|
Operations Research 50th Anniversary
History in the Making
INFORMS celebrates 50 years of problems, solutions, anecdotes and achievement
By Peter Horner
With visions of their fledgling field's success during World War II still dancing in their heads and eager to train their technical talents on peace-time problems in industry and business, 73 individuals from academia, the military and corporate America gathered at the Arden Estate in Harriman, N.Y., on May 26, 1952, to map out plans for a professional organization that would carry the OR banner in the United States. They emerged at the end of the historic day as the founding fathers of the Operations Research Society of America.
Before the year was out, ORSA would publish Vol. 1, No. 1 of its flagship journal, Operations Research, and organize its first national meeting, held at the Bureau of Standards in Washington, D.C. Four hundred members and guests attended. The world seemed full of interesting, complex industrial problems that cried out for operations research-type analysis.
"The early ORSA meetings were attended by a couple of hundred people, almost all of them employed by industry, government or the military," recalls longtime ORSA member Sid Hess. "The best parts of the meetings were late night bull sessions in the cocktail lounge where students, faculty and practitioners argued and exchanged views."
But all was not cozy in the OR community.
"There was a feeling among [some of the members] back then that ORSA was too strongly oriented toward mathematics, that its leaders were just interested in military applications," said Andrew Vazsonyi, one of the first "associate" member of ORSA. "ORSA was dominated by the Phil Morse and George Kimball crowd, people who were always talking about the things they did for the military during World War II. Many of us thought that ORSA needed another approach, something oriented toward management, but we didn't think we could get the Morse crowd to change. It turns out we were right." 
ORSA's early focus on military matters at the expense of management, combined with its restrictive, layered membership policies, prompted several disenfranchised members to consider establishing a new society that catered to their needs. Not long after ORSA was founded, Melvin Salveson organized the first in a series of informal meetings on the West Coast to explore interest in the venture. Almost simultaneously, similar groups with similar designs sprung up on the East Coast. David Hertz, then teaching at Columbia University, hosted one such meeting at his New York apartment.
"There were about 10 of us," Hertz remembered. "Bill Cooper was there. Merrill Flood was there. We spent a long night talking about the creation of [what would become] The Institute of Management Sciences. The big question, with ORSA already up and running, was, Why are we doing this? We were doing it because a lot of people, myself included, were not satisfied with the idea that ORSA didn't seem to be concerning itself with the management world. We believe we could make important contributions in this area by applying the techniques of operations, which we called management science." 
Meeting of the Minds: The head table (rear of the photo) at TIMS' Second Annual Banquet, held in New York City in 1955, featured many of the profession's most notable pioneers including (left to right) Roger Crane, Ezra Glazer, George Kozmetsky, Herbert Simon (vice president), William Cooper (past president), Merrill Flood (president), Melvin Salveson, C. West Churchman, Gifford Symonds, Alex Orden, Andy Vazsonyi and David Hertz. (click here to view a larger version in a seperate window)
Raising Another Banner
Just 18 months after ORSA was founded, a room full of economists, engineers, mathematicians, statisticians, astronomers, philosophers, attorneys and, yes, erstwhile operations researchers met on a cold winter's day at Columbia University to ponder the possibility of joining forces under another banner, this one called "management science." The guest list included four future Nobel Prize winners and one future recipient of the National Medal of Science, George B. Dantzig. They debated where the new organization should go, how it should get there and whether the effort was even necessary, given the existence of ORSA.
At the end of the day, they decided that the effort was indeed worth it, and thus was born The Institute of Management Sciences. The date was Dec. 1, 1953. William C. Cooper was elected the first president of TIMS. Several of the men who had attended the founding meeting of ORSA 18 months earlier traveled to New York for the founding of TIMS. Prominent among them was Churchman, who would become the founding editor of TIMS' flagship journal, Management Science.
For a profession that strives to bring clarity to management decision-making, TIMS got off to a confusing start. For one thing, Salveson reportedly showed up at Columbia having already collected about $60 in dues for an as yet non-existent Institute.
"That really got Dantzig upset," Cooper said many years later. "I remember I got in a big argument with Salveson over the nature of management science. He was looking, in a scientific sense, for laws and regularities that he believed governed management. I thought management science meant developing methodology, improving the way we do things. Looking back, I would have to say I was right." 
Churchman, meanwhile, squared off with Dantzig over the issue of ethics. "George and I didn't see eye to eye on linear programming," Churchman recalled. "I'm a philosopher. When they said optimize, I thought they meant optimize in an ethical sense. You can optimize the decision-making function in linear programming and find out that it is only optimal in terms of what management perceives to be the monetary benefits minus cost. LP's only constraints are physical or purely financial, not ethical.
"George and I were like any two old professors who recognized that the other SOB was simply off somewhere else. George was a mathematician, I was a philosopher. George was selling LP, and I was trying to sell ethics. Looking back, it's pretty obvious that George did a better job of selling than I did." 
The eclectic cast of characters at Columbia managed to put aside their individual differences long enough to agree to proceed with the formal founding of TIMS. The founding fathers adopted a constitution and bylaws, elected a slate of officers, appointed an editorial board and established a committee to incorporate TIMS. In true Jeffersonian fashion, Constitution Chairman Merrill Flood and company authored an objective statement that holds up as well today as it did 50 years ago: "The objects of the Institute shall be to identify, extend and unify scientific knowledge that contributes to the understanding and practice of management."
The founding fathers of TIMS might have been wise in their ways, but they weren't perfect. In crafting the constitution, they created the position of past president, and bestowed upon the post a number of important duties, not the least of which was the power to open meetings and chair the Council. After adopting a constitution and electing officers, the Institute found itself in a dilemma: Without a past president, the meeting could not officially begin.
About that same time, over in the ORSA camp, the operations researchers found themselves embroiled in a similar debate concerning elections. Should officers be elected by a majority, a plurality or some other means of counting? The hour was late. Everyone was tired. Lengthy discussions had failed to produce an answer. Finally, George Kimball stood up and simply said, "the one who gets the most votes wins."  End of debate.
From the beginning, ORSA and TIMS staked their reputations on using sophisticated mathematical modeling to solve complex operational problems. At the same time, good operations researchers and management scientists know that sometimes all it really takes to solve a problem is a little common sense.
Brothers (Up) in Arms
Like two rival brothers who have much in common yet can't seem to see agree on anything, ORSA and TIMS went their separate ways, only to reunite later in life. In 1954, ORSA appointed its first (and long-serving) business manager, Norvell Miller III, and presented its first award, the Frederick W. Lanchester Prize for the best English language paper in operations research. In 1957, the Case Institute of Technology awarded the first Ph.D. degree in operations research and ORSA became a charter member of the International Federation of Operational Societies, extending its worldwide reach.
TIMS, meanwhile, was busy building its own flagship journal, Management Science (1954), appointing Harold Cauvet as its first executive director (1958) and approving its first college (Planning, also in 1958).
By the end of the 1950s, ORSA membership stood at about 3,000. Membership in TIMS soared to more than 2,600, boosted by a meeting in Paris in 1959 that included membership dues as part of the registration fee. A significant number of people such as Hess, who joined ORSA and TIMS in 1957, maintained memberships in both organizations.
ORSA, the "older brother," was generally considered the stronger of the two, at least in a financial sense. In fact, a series of unsettling events in the early 1960s stopped TIMS' growth in its tracks and brought the Institute to the brink of bankruptcy. The tumultuous period was marked by a revolving business office (three moves to three different cities) and executive directors (four in six years). Things settled down with the 1968 appointment of Mary DeMelim as executive director, a post she held for more than 25 years. (Patricia Morris, whose career closely mirrored DeMelim's, added stability to ORSA as its executive director during much of the same timeframe.) The late 1960s ushered in a second wave of sustained growth that pushed TIMS' membership past 6,500 by the end of the decade. 
MIT professor John D. C. Little, who attended his first national ORSA meeting in 1955, recalled the 1960s as a time when the emphasis in OR was on methodologies, which involved into an increase in the "mathematization" of the field. By the 1970s, Little said, "we returned to more attention to applications in the universities and journals." 
The 1970s also marked a reunion of the long estranged "brothers" ORSA and TIMS highlighted by the joint publication of Interfaces and the first joint U.S. meeting (held in Boston). In 1971, TIMS established the College on the Practice of Management Science and the Management Science Achievement Prize Competition, which was later renamed the Franz Edelman Award for Management Science Achievement. Edelman, the longtime leader of the management science group at RCA, was widely considered the "quintessential" management science practitioner.
In 1974, ORSA established the Kimball Medal for service to the society and the profession, and the John von Neumann Theory Prize. From the narrow military focus of its early years, operations research broadened its scope to include a dizzying array of theoretical research and applications in the industrial and service sectors. By the end of the decade, the fuzzy line that had historically distinguished ORSA and TIMS, and OR from MS, had all but disappeared, prompting the first serious talks of merger.
"In the 1950s and '60s, the competition between the two organizations was almost unfriendly," Hess recalls. "Gradually those of us who lived and worked in the overlap whatever that was helped encourage cooperation."
Winds of Merger
Merger seemed inevitable, but it required many more years of joint meetings, joint journals, joint councils, joint proposals, joint counter-proposals and good, old-fashioned lobbying before ORSA and TIMS would officially tie the knot. In the meantime, both organizations enjoyed impressive growth in terms of their journals, subdivisions and meetings. At the first ORSA national meeting, all of the technical papers were delivered in a single session. By 1990, the joint national meetings were running more than 40 parallel session tracks. (Whether the quality kept pace with the quantity of sessions is another question for another day.) The April 1988 TIMS/ORSA National Meeting in Washington D.C., chaired by Saul Gass, drew 2,879 total registrants a record that still stands.
Logic says that the growth of ORSA and TIMS should have been driven to a large extent by the development of the computer, the increased access to data and the creation of landmark theories, algorithms, methodologies, applications and publications. The accompanying timeline compiled by Saul Gass lists many of these "great moments in OR history." Certainly Dantzig's work on linear programming and the simplex method in the late 1940s, for example, jump-started the profession.
History, however, often defies logic. Sometime in the 1970s, right about the time Apple Computer started making personal computers for the masses, the growth rate of membership in ORSA and TIMS began to slow down. In the 1980s, a decade when Narendra Karmarkar's interior-point LP solution method marked the first fundamentally new LP algorithm in nearly 40 years and the OR group at American Airlines pioneered the concept of yield/revenue management that revolutionized how airlines and other industries do business, membership growth slowed even further. In the 1990s, at a time when computer power was growing exponentially and the business world was floating in a sea of data generated by virtually every desktop in the world, membership growth stopped altogether.
What's wrong with this picture?
It's not a question of what went wrong, but where the demand for OR/MS-type skills went. The classically trained OR professional was a generalist, but the market began demanding specialists. Many companies that had historically supported separate OR groups or departments began to disperse their analytical talent throughout the enterprise. An individual armed with an advanced OR degree who ended up in the financial department, for example, began to think of herself in terms of finance rather than operations research. She was still doing OR-type work and using OR-type tools, but over time she became more in tune with the folks in finance rather than those in OR, and perhaps she changed her professional affiliation as well. That might explain why the U.S. Bureau of Labor Statistics keeps saying there are 50,000 to 80,000 (depending on the year and the definition of "operations research") OR analysts running around the country, but only a fraction of them called ORSA, TIMS or now INFORMS home. In other words, the number of people "doing OR" is almost certainly increasing; they just aren't calling it OR (or joining INFORMS).
It didn't help that traditional management science courses started losing ground in business schools. It turns out that MBAs are more interested in making money than making mathematical models. Go figure. The net result: many business schools responded to their customers' (i.e., students) demands by reducing or eliminating the once-required management science component of their curriculums. That, in turn, further dampened the profile of OR/MS in the corporate world and no doubt led some analysts to reconsider their professional affiliation.
Finally, the so-called "bowling alone" syndrome (from the Robert Putnam book of the same title and the subject of a recent column by INFORMS President Mike Trick ) negatively impacted INFORMS membership just as it did many other associations. The premise of Putnam's book is that over the last 25 years, people have become less interested in building "social capital" than they once were. As Trick notes in his column, summing up Putnam's book, "We sign fewer petitions, are less likely to volunteer for a political party, are less likely to invite people to dinner and are less likely to join organized sports" (like bowling leagues, hence the title of the book). We're also less likely to join professional associations like INFORMS. If it's any consolation, INFORMS is not alone in this regard. The vast majority of professional associations have experienced declining membership in recent years.
Whether people were, in fact, wandering off the OR/MS reservation to affiliate with other fields or simply opting to "bowl alone," concern over dwindling membership in both ORSA and TIMS was one of the factors that brought the issue of merger to a head in the mid-1990s. Merger was widely seen as a logical next step for two kindred organizations that already shared about half of their budgets and activities. Along with streamlining some obvious internal inefficiencies, the merger was sold as a means to gain more external clout with a united front while creating "a big tent" organization that would foster the growth and independence of existing subdivisions and perhaps attract new, related societies.
During the summer of 1994, members of ORSA and TIMS voted overwhelmingly in favor of merger. On Dec. 31, 1994, ORSA and TIMS ceased to exist. On Jan. 1, 1995, the Institute for Operations Research and the Management Sciences was born. John D. C. Little, a former president of both ORSA and TIMS and one of the primary architects of the merger, was elected to serve as the first president of INFORMS. Alfred Blumstein, another past president of both ORSA and TIMS, was voted to the post of president-elect.
As INFORMS carries the banner of OR in America into a second half-century, the Institute has much to celebrate, as President Mike Trick noted in that same "bowling alone" column. INFORMS continues to expand and improve its products and services. The Institute boasts 10 respected titles in its portfolio of journals. All of them are ranked at or near the top in their specialties and the two flagship publications, Management Science and Operations Research, were recently named to a list of the 12 most influential business journals in the country. INFORMS has created a strong Internet presence: Informs Online receives thousands of hits a day and the successful online publication program has opened up an important new stream of revenue. Meanwhile, INFORMS' national conferences continue to draw well, the new practice-oriented meetings have quickly established a reputation for quality and the Institute is on solid financial ground.
Despite all of its successes, however, two key problems still haunt the Institute. The merger did little, if anything, to boost membership. After reaching a high-water mark of nearly 15,000 (combined ORSA and TIMS) in the late 1980s and early 1990s, membership has slowly retreated to about 10,000 today. Even more perplexing is "the name issue." Despite countless attempts at "branding" its product, the Institute still has not found an effective way to clearly define, let alone market, what its members do and why the world should care. Fifty years after the founding of ORSA and 49 years after the founding of TIMS, the terms "operations research" and "management science" carry virtually no meaning outside the walls of the OR/MS community.
Born in the crucible of World War II with the fate of a free Europe depending on its success and buoyed in the post-war era of rapidly expanding industrial and management opportunities, is OR/MS destined to be forever bewitched by a world that doesn't appreciate its many contributions?
Nonsense, said David Hertz, who hosted that first TIMS planning meeting in his New York apartment 50 years ago. "In any long journey, detours are inevitable," Hertz said upon the 40th anniversary of TIMS. "Membership grew, we almost went broke. TIMS set out to bring the kind of science represented by operations research to the management world. Its members did so in many, many ways and I think it was successful.
"In a way, growth is somewhat of a handicap. Today, you hear a lot of complaining from our members about a lack of recognition. That's a bunch of bull. I don't have any sympathy for those who complain. The people who do exceptional work are always recognized with full professorships at major institutes, with major jobs in industry, with big appointments in government, with research grants.
"I think the problems is there are too many people who like to call themselves management scientists. There are 500,000 lawyers in this country. Are they all recognized? I believe a great deal can be done, but it's not going to be done by complaining. In my opinion, we're crying over the wrong issue. Recognition is not the issue. Good work is. You've got to understand the world of management, the managed and managing." 
"Clarity is supposed to be the objective of science. I disagree. I think the objective of science is confusion, because confusions carries you into problems."
After 50 years of combined history, INFORMS still finds itself knee-deep in confusion. Plenty of problems have been solved... and plenty of problems remain.
Peter Horner is the editor of OR/MS Today. This article (including some passages and attributed quotes) is based on published accounts of ORSA, TIMS and INFORMS that have appeared in OR/MS Today. during the past 10 years, most notably the 40th anniversaries of ORSA and TIMS in 1992 and 1993, respectively, and the merger debate that led to the creation of INFORMS (1994-95). Horner thanks Saul Gass for his comments and suggestions.
OR/MS Today copyright © 2002 by the Institute for Operations Research and the Management Sciences. All rights reserved.