-4

Proposal: Operations Research and Analytics

If the forum title were "Operations Research and Proposed Analytics" rather than "Operations Research and Analytics", it would be clearer that the forum is not about choosing the best shades or fonts for analytics dashboards and similar things often going under the Analytics moniker. Prescriptive analytics might be more or less synonymous with Operations Research. But analytics encompasses many things which would make the forum too diffuse.

Edit: As I stated in comments, I interpret "and" in the forum title to be union, not intersection. So everything in Operations Research is covered. Also, as elaborated in the comments, I view Operations Research widely, not as just being a bag of technical tools. Therefore it includes most of the "good" stuff often promoted under the Analytics banner, to address the end to end analysis process.

  • 1
    I reckon that the title "Operations Research and Analytics” embrace the whole operative chain from data collection, unveiling of constraints, choice of KPIs / objectives, making of optimization/simulation/machine learning models, use of software for such models, and interpretation/presentation of solution, as well as academic studies herein. Indeed broad, but I don't have a problem with that. – Henrik A. Friberg Apr 1 at 8:55
  • 1
    @Henrik A. Friberg I claim that Operations Research ecncompasses all those things - at least it should, and when I do it, it does.. It is not just a bag of technical tools. – Mark L. Stone Apr 1 at 12:13
  • 2
    @JamesJenkins: The Discussion Zone is different -- read what you link: "For more extensive deliberations and discussions regarding the proposal as a whole, we have a separate Discussion Zone". We are supposed to chat to figure out what we want "Operations Research and Analytics" to include and exclude! ^_^ – Henrik A. Friberg Apr 2 at 7:51
  • Repeating my comment below "Also, and most importantly, simulation is considered within scope of Operations Research. So the forum name definitely covers that.". I.e., I interpret "and" in the forum title to be union, not intersection. So everything in Operations Research is covered. And my interpretation of what Operations Research is or should be includes everything listed in the first comment above by @Henrik A. Friberg . I will add to his list, following through to assessment/refinement of implementation of solution. Throwing a briefing or software over the transom is usually inadequate. – Mark L. Stone Apr 2 at 12:27
17

My opinion is that we should leave it as "Analytics". We (OR/MS/IE people) do more than just prescriptive analytics -- for example, simulation is descriptive analytics, demand forecasting is predictive analytics, etc.

I also am taking a page from the INFORMS web site, which uses the phrase "Operations Research and Analytics" a number of times.

I do agree that "analytics" is being used WAY too broadly, but I think "prescriptive analytics" is too narrow.

But -- I think this is an important question and a big one, and I'm curious to hear others' thoughts on it.

4

Sometimes less is better: I’d prefer “Operations Reserach” and that’s it.

  • So would I. My suggestion was already meant as a compromise vs. what I really wanted.And don't get me started on my opinion about CAP (INFORMS trying to erect a toll booth).. – Mark L. Stone Apr 6 at 12:57
3

I was checking the area51 and found a proposal, dedicated to data science, that has already finished the beta stage (the operational version is available from here). This might limit our chances of seeing this proposal through the process.

  • 1
    It looks like datascience.stackexchange.com/questions is starting to catch on a bit. For many years, it had very little activity, and many questions which seemed like they might be suitable for it were instead asked, and answered, on stats.stackexchange.com , which is very active, and has a significant overlap with what could be on an Operations Research/Analytics site. – Mark L. Stone Apr 1 at 15:24
  • 2
    I don't think the existence of the data science site limits our chances of being successful with the OR site. The guidelines say that it makes sense to include site X within site Y if "Almost all X questions are on-topic for site Y," which I don't think is true if X = OR and Y = data science or stats. – LarrySnyder610 Apr 1 at 17:20
  • 2
    On the other hand, the existence of the data science site might have an effect on our decision with regard to the question of "Analytics" in our title. – LarrySnyder610 Apr 1 at 17:20
  • 1
    @Larry Reminds me of when Stanford had 3 "O.R.", more or less, departments: Operations Research (mine, which was the best), Engineering-Economic Systems, Industrial Engineering and Engineering Management (IE was considered old-fashioned, like O.R. is now). Huge overlap, even if some differences in philosophy & emphasis. But basically 3 different names for roughly the same thing. But founded and populated by 3 rival groups of faculty members, each preferring their own department name. Eventually, after a series of shotgun weddings, the 3 were combined into "Management Science and Engineering". – Mark L. Stone Apr 1 at 18:18
  • 1
    @LarrySnyder610: I was referring to the analytics part. OR is probably safe as it is a multi-disciplinary field with various theoretical and applied aspects. This makes it difficult for OR to be a subset of any existing field on StackExchange. – Ehsan Apr 1 at 21:14
1

I don't love the proposed title Operations Research and Analytics however I think the alternate title of Operations Research and Prescriptive Analytics is quite a bit worse.

As you know, Operations Research (OR) is a broad field. While many facets of OR are prescriptive in nature, many of the most used OR techniques are inherently predictive. I understand the OP is suggesting "OR and Prescriptive Analytics" this is still too misleading.

-2

I agree (stochastic, a.k.a. Monte Carlo) simulation should be in scope.However, even though simulation may sometimes be viewed as predictive rather than prescriptive, I've seen it described as being part of prescriptive analytics. Simulation is often used to support decision making, which would place such use at least very nearby, if not in, prescriptive analytics. And of course, whether formally or informally done, simulation is often used to optimize or improve the (system) phenomenon being simulated, which clearly is prescriptive.

So my proposed use of "prescriptive analytics" would be a broad tent view, certainly encompassing simulation.

  • "Simulation is often used to support decision making" -- isn't that true for every sort of analytics? – LarrySnyder610 Mar 31 at 21:48
  • No, it's not, Anyhow, I strongly dislike the term analytics, and I did not make up the delineations, descriptive, predictive, and prescriptive. And of course there are no clean boundaries to any of this. I'd rather not have the word "analytics" in the forum name at all, but view having "prescriptive analytics" as a reasonable compromise. – Mark L. Stone Mar 31 at 21:51
  • OK fair enough. I'll step back and let some other folks weigh in. – LarrySnyder610 Mar 31 at 22:00
  • Also, and most importantly, simulation is considered within scope of Operations Research. So the forum name definitely covers that. – Mark L. Stone Mar 31 at 22:06
-2

I'm with Shakespeare on this:

"What's in a name? That which we call a rose
By any other name would smell as sweet."

  • 1
    What's in undesired posts due to name choice? If that didn't matter, you'd be fine with the old OR-X forum being SPAMMED to the hilt. So unlike name of a rose, the board name could matter. – Mark L. Stone Apr 1 at 21:57
  • Ironically, the original quote is in the context of ignoring real-world limitations on what can be done by a given group or individual, due to the connotations of the name of that group/individual. Words mean things, so the name matters a lot. – Nij May 12 at 19:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .