Back in 2004 the Boston Globe newspaper ran a story about “the most influential academic discipline you’ve never heard of”. What was the discipline? Operations research.

Operation everything

It stocks your grocery store, schedules your favorite team’s games, and helps plan your vacation. A primer on the most influential academic discipline you’ve never heard of.

This is rather ironic for those of us in Constraint Programming, since CP deals with many of the same applications as Operations Research, and is, if anything, less well known.

Part of the problem unquestionably lies in the name. “Operations Research”, what is that? The study of hospitals? And the alternative, “Mathematical Programming”, is even worse. Is it mathematics? Is it coding? The only worse name for a field? “Constraint Programming”.

I once had a fellow from a company that I was trying to interest in working with my lab tell me he liked the idea, but ask me if there was some other name for what I did, because “constraints” had such a negative connotation, it turned other people in his company off. Of course, we do have some very positive terms associated with the field: “satisfaction”, “relaxation”, “optimization”. But if I said “I use relaxation for optimal satisfaction”, people might think I was some kind of new age guru.

Indeed even “optimization” is an imperfect standard bearer. I once had a fellow at a commercially-oriented conference tell me that “optimization” could turn his potential customers off. Astonished, I asked why, and he explained that they worried that it meant that the computer would make all the decisions and they would be cut out of the loop. Of course, it doesn’t have to mean that, but perhaps such fears help to explain the current popularity of terms like “human-centred” and “human-aware” as modifiers to “Artificial Intelligence”.

In any case, if we are to compete with such media-friendly terms as “artificial intelligence” and “deep learning” we need to work at getting the message out about what we can do. Actually Operations Research is currently doing a pretty good job of that for their field. See “O.R. & Analytics Impact Everything” at the INFORMS website, and the INFORMS Success Stories.

The ACP has its Success Stories as well, of course, and there are other resources, but we can always do more. I’ve started work on a new page, Sample Applications of Constraint Satisfaction, to make a small contribution towards “getting the word out”.