Member-in-Chief Memo

Irrevocably allocating resources

Ed Kaplan
INFORMS President
president@informs.org

Ed Kaplan

As of early September when this memo was written, the race for the White House had heated up to become one of the hottest contests in memory. By the time many of you read this, the election could be history, Americans will have made their decision, and the next president of the United States will have been chosen. Flying beneath the radar of cable television and social media, another election is in play at present, namely, the race for president-elect and 2018 president of INFORMS. Again by the time you read this, the newest elected leader of INFORMS will be known.

Both of these elections are examples of collective decisions, and both reflect the by now classic O.R. definition of “decision” introduced by Ron Howard in his 1966 paper “Decision Analysis: Applied Decision Theory.” Howard’s definition: “A decision is an irrevocable allocation of resources, irrevocable in the sense that it is impossible or extremely costly to change back to the situation that existed before making the decision. Thus for our purposes a decision is not a mental commitment to follow a course of action but rather the actual pursuit of that course of action.” Indeed, once the president of either INFORMS but especially the United States is chosen, it would be extremely costly and difficult to deliberately roll back the clock and return to the state where those elected were no longer so!

It is this commitment to and implementation of action that distinguishes how operations researchers think about decisions compared to, say, historians and psychologists. And, it is exactly this focus on decisions-as-resource-allocation that is the hallmark of so many of our research contributions over the 60+ years our professional association has existed, dating back to the founding of ORSA in 1952.

Indeed, we are the experts of decision-aiding technology. That is reflected in the terminology of even our most technical contributions. Optimization models encode choices via the use of decision variables. The entire field of decision analysis à la Howard Raiffa, Ron Howard, Ralph Keeney and others has evolved within our professional world to the point where INFORMS boasts both its own Decision Analysis Society and supporting journal titled, appropriately enough, Decision Analysis. And, in the idiosyncratic work of our members across many disparate areas – communications, e-commerce, health care, manufacturing and service operations, military operations, policy modeling and analysis, transportation – models of varying complexity are constructed to capture the most salient features of whatever is studied in order to help evaluate the consequences of alternative decisions en route to identifying the best course of action.

Paraphrasing one of the four goals in the INFORMS strategic plan adopted during our Winter 2016 board meeting: Decision-makers will have access to and use our approaches to transform visions/tasks into better choices to achieve better outcomes. While members surely agree that this is a worthy goal, simply saying this is so does not make it so. Simply put, INFORMS could use your advice regarding how to better reach out to decision-makers to provide better awareness of and access to ready-to-go decision-aiding tools.

Some of you have already provided wonderful examples of what can be achieved by creating websites from which your models can be deployed, either via download or in real time online. As an example, consider the RealOpt© suite of models developed by Professor Eva Lee and her colleagues at Georgia Tech to help public health systems plan and allocate resources for all hazards emergency response. Developed for the Centers for Disease Control, the system is available to any public health system. Perhaps INFORMS could create an online library of links to functioning models (and how-to users guides) that could then be publicized by INFORMS as a resource for decision-makers in different areas.

INFORMS already offers various educational programs and focused conferences across our areas of expertise. Perhaps there is also room for a conference showcasing our best decision-aiding tools for a carefully targeted audience of leading decision-makers. Beyond the usual goals of our meetings, this meeting would have the tightly focused goal of technology transfer: have decision-making tools, will help make decisions!

Election season reminds us what is at stake when high-profile decisions must be made. And make no mistake – decisions will be made, with or without our input. Members of INFORMS, let’s do all that we can to make sure our decades of research and expertise in decision-making are used when consequential decisions are made, rather than merely provide a lens for outside commentary after the fact. Indeed, promising approaches to reaching decision-makers with the best we have to offer deserve the irrevocable allocation of INFORMS resources.