April 2000 Forum We Live in the City, Not in a Study By Tom Koch While considering a jaunt to Korea for INFORMS-KORMS Seoul 2000, I've been thinking about the Philadelphia conference last November. Will the conference in Korea be sufficiently interesting or useful (not always the same thing) to justify the time and, for me, considerable expense it involves? Thinking of INFORMS, past and future, I've also been remembering an incident at the Fourth International Symposium on the Analytic Hierarchy Process (AHP) held in Burnaby, B.C., Canada, in 1996. At that meeting a Japanese scholar spoke to a large audience about his use of AHP to determine the best strategy for winning baseball games [1] The Japanese have a passion for the game and this OR strategist applied his preferred tool with fervor. He interviewed coaches, constructed his hierarchy, weighed the results of a pair-wise analysis, and ranked the needs of the baseball team that sought to win. In halting English, he slowly explained that hierarchically, fielding, catching and pitching were most critical to the game. After he spoke there was polite applause and a moment of silence in which he waited for praise, and perhaps, a polite question or two. A woman in the back row raised her hand and when she was recognized asked, politely, "What about batting and hitting home runs?" In his model he had ranked everything but power hitting, the thing really needed to win. It was so obvious and so critical that the coaches he had talked to apparently hadn't thought to list it. And, good researcher that he was, he had not thought to ask. "Oh yes," the Japanese scholar said, somewhat flustered. "Very important. Home Runs." As Peter Horner says of the humor he found at the Philadelphia INFORMS meeting, "maybe you had to be there." [2] The Philadelphia meeting was filled with similar moments, although most were less extreme. In considering models developed for clients, in the report on projects yet unfunded but within the OR vision, practitioners seemed bent on forgetting the critical and the obvious. The problem wasn't simple ignorance or oversight, but a perspective that assumed if only the system could be tweaked and the model improved that everything else would fall into line. So in lectures on transportation planning, nobody talked about the general U.S. failure to provide adequate funding for a public system, while assuring average if hidden governmental subsidies of more than $2,500 per year per licensed automobile. [3] Nobody talked about how this disadvantaged those who, like me, cannot drive because of physical or visual impairments, or the poor whose travel is thus restricted. Nobody talked about what it means to give over as much as a third of the city's space to the automobile, or what more than 50,000 fatal injuries a year in automobile accidents means to the health care system. Not What the Client Asked For The context was too big, and after all, it's not what the clients asked for. "What do you think of this as a citizen?" I asked several times. "That's an interesting question," I was told, "but not within the study's parameters." Alas, we live in the city and not the study, and it's the former context that defines our lives. The later is merely what we may be paid to do . . . jobbers for the world. In my current field of health care, I heard how one might streamline the delivery of intensive care services, but not how to improve both general and home care services for folk who are sped through the system. It wasn't within the parameters of the study, of course. A presenter arguing means of increasing the efficiency of emergency room treatment of non-critical persons did not think of what this meant for those who might have to seek these services elsewhere. The study was about making money-losing ERs more efficient, not about the necessary services provided by what is, for many Americans, the only health provider they can find because it is the only thing they can afford. [4] Ruth Malone is right, the ER is the last alms house of the Americas. It's where people go for treatment, and sometimes for a quick meal and a change of discarded clothes. And if the alms house function is inefficient, well, OR has the answer: Get rid of it and don't worry about the result on those who will be harmed. The problem was rarely ignorance — although that was present, at times, as it always is at a large conference where people explore new ideas — but attitude. The OR practitioners whose lectures I attended have, by and large, taken a manufacturing model of production and applied it to areas affecting social policy, and more importantly, social life. By doing this they've allowed themselves too often to avoid the consequences of their work upon the greater society, the greater import of their small-scaled work upon the greater field. "Widgets are us," should be the OR devotee's and modeler's motto. The failure of this general perspective was forcibly brought home to me in a session on health care in Philadelphia where I asked each speaker about these greater links. "What does this mean to you as a citizen?" I would ask. "Gee, I don't know," came the refrain, "it's not in the study's parameters." But whatever our expertise, we are all citizens first and foremost. If the OR mystique blinds practitioners to that, then it blinds them to the world their work may indeed affect. In that event, research contracts in hand, we become what we've been taught to despise: amoral bureaucrats without a care for the world we inhabit. Perhaps former U.S. Secretary of Defense Robert MacNamara, the man who applied system analysis to the Vietnam War, should be our honorary mascot. At the least, he should be asked to speak at a plenary in Korea. At a health care session an OR practitioner argued this perspective with me. My concerns were irrelevant, he insisted, to the problem at hand. He kept speaking of the product, and the customers. I kept talking about the service, and the patients. The difference was — and is — more than semantic. When social services become products it is easy to exclude elements of care to assure efficiency and give the customer — normally the agency that has employed the practitioner — the limited answers he or she wants. Cut here, trim there and tweak the system, please. The real "customer" — the health care patient or the transit rider — becomes nothing more than a cipher on the demand curve that sits comfortably distant from the life and work of the OR practitioner and his or her employer. Management experts rarely experience the horrors they might cause in the name of efficiency. It's one of the real boons of the job. If, as Douglas Samuelson insists, an OR analyst solves "big complicated problems with numbers in them," this is not the way to solve anything. If that is what operations research is about, the practitioner has a dual obligation in seeking even a limited solution to an assigned problem. He or she must see it first as a citizen involved in the social context and then as a specialist who will insist upon addressing the problem and its context together. We are first and foremost citizens of societies, or we are nothing much at all. Our first obligation is therefore not as practitioners working for wealthy client companies, but as citizens who see those problems in the context of our social world. The Vietnam War impressed this lesson upon my student colleagues and me in the 1960s. Early system analysts defined the war as an abstract problem in a world apart from the one we inhabited. They sought a winning solution and forgot what we then students of draft age knew: The war was not just a strategic problem but the story of our lives and deaths, and of course, those of the Vietnamese. And so even as we studied system theory we were citizens first, arguing with the analysts whose tools said "acceptable losses here" would mean a more efficient war, a better outcome. A dialog about the sometimes conflicting roles of analyst and the citizen is rarely read in these pages, or heard at INFORMS meetings. There were almost no social policy-based sessions at the Philadelphia meetings, and the few that were scheduled were buried at odd times and in small rooms. Who cares? These aren't the issues set forth by the client. And yet, issues of citizenry and responsibility undeniably bear upon the success or failure of the health, transit, and other programs modeled by OR practitioners. Self-esteem and self-interest, if not a sense of civic membership, should therefore demand a greater sense of social integration in these meetings. We who are patients, transit riders and supply-side ciphers will not conform easily to the abstraction of a Markov chain analysis. Ignore that truth and the elegant theories of OR modeling will never translate from computer spreadsheet or modeled graph to the world. These differences of scale and perspective were, I believe, at the heart of a debate in this journal over the modeling of distribution of transplantable human livers. I criticized simulation modelers and others who I believed sought to maximize the efficiency of organ supply and distribution without attention to a broad range of social and political factors. [5] I modeled the problem at a courser scale, focusing upon broad social restrictions and inequities that define the general context. [6] In arguing at this geographical and political scale, however, I was criticized in turn for misunderstanding the technical elements of the problem and the model. [7] Kids Playing With Uzis I doubt I'll go to the Korea meeting. It's a wonderful country with a history, art and culture worth knowing. But going to INFORMS in Philadelphia was rather like watching children play with Uzi machine-guns. Powerful tools in the hands of those who don't think beyond their immediate effect, who see no farther than the next battle or project or tactical situation, are tools turned to disaster and not to the heart of the problem itself. OR practitioners have access to incredibly powerful analytic weapons whose use affects us just as much as the product of this or that armory. Until we remember we're citizens first, people asked to consider specific problems within a social context, we're no better than the kid whose brigade leader gives him a clip of bullets and sends him or her out to fight and die for a cause. I don't have to go to Seoul, Korea, to see that. I can watch it tonight on the evening news. References 1. Proceedings of the Fourth International Conference on the Analytic Hierarchy Conference, Simon Fraser University, Burnaby, B.C., July 12-15, 1996. 2. Horner, Peter, 1999, "Highlights and Low-brow Humor," OR/MS Today, December 1999. 3. See, for example, "Pollution Probe. The Costs of the Car: A Preliminary Study of the Environmental and Social Costs Associated with Private Care Use in Ontario, Toronto," Ontario: Pollution Probe, 1991. Other studies abound and will be familiar to transportation planners. 4. Malone, Ruth E., "Whether the alms house? Overutilization and the role of the Emergency Department," Journal of Health Politics, Policy, and Law, 1998: Vol. 23, No. 5, 795-832. 5. Koch, Tom, 1998, "Transplantation: Fairness vs. Efficiency," (letter to the editor), OR/MS Today, June 1998. 6. Koch, Tom, 1999, "The Organ Transplant Dilemma: What is Fair? What is Equitable?" OR/MS Today, January 1999. 7. Pritsker, Alan and Koch, Tom, "Organ debate Œtransplanted' to pages of OR/MS Today: Different Scales, Modes, and Methodologies," OR/MS Today, June, 1999; and Koch, Tom Koch, "Transplantation: Fairness vs. Efficiency (letter), OR/MS Today, October, 1998. Tom Koch is adjunct professor of gerontology at Simon Fraser University and the author of "The Limits of Principle: Deciding Who Lives and What Dies" (Praeger, 1998). His URL is: wwwa1net.com/koch/. Table of Contents OR/MS Today Home Page OR/MS Today copyright © 2000 by the Institute for Operations Research and the Management Sciences. All rights reserved. Lionheart Publishing, Inc. 506 Roswell Street, 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 1999, 2000 by Lionheart Publishing, Inc. All rights reserved.