3

AI algorithms, specifically in industries, can be used to work and manipulate the data to estimate or learn this data to achive a solution or prepare trained data to feed that into an optimization model.

On the other hand, the simulation techniques, especially discrete-event and Mont-Carlo, use almost the same data, e.g. number of failures in a specific station in the production line, to either estimate a function to robust some parameters as a final solution or feed these parameters into an optimization model. And already investigate the behavior of the complex system in a dynamic scheme with an appropriate feedback.

Now my question is, what exactly is the main difference between an AI method and a simulation technique for example to estimate the failures or production utilization?

A.Omidi
  • 8,832
  • 2
  • 13
  • 49
  • 5
    The main difference is hype. ""When we raise money it’s AI, when we hire it's machine learning, and when we do the work it's logistic regression."" https://twitter.com/daniela_witten/status/1177294449702928384 – Mark L. Stone Aug 28 '23 at 10:42
  • 1
    Dear @MarkL.Stone, Thanks for your reply. Would you please, add your comment as an answer. – A.Omidi Aug 29 '23 at 04:53

1 Answers1

5

The main difference is hype.

Here is a quote tweeted by University of Washington Statistics Professor Daniela Witten in 2019, attributed to source unknown, which captures the spirit:

When we raise money it’s AI, when we hire it's machine learning, and when we do the work it's logistic regression.

Basically, most of O.R. could be considered to be AI.

Mark L. Stone
  • 13,392
  • 1
  • 31
  • 67
  • 1
    Dear @MarkL.Stone, thank you so much. – A.Omidi Aug 29 '23 at 10:37
  • 2
    Here is something (renewal paradox, a.k.a. inspection paradox, a.k.a. length-biased sampling.) which would elude the average AI or Computer Science person doing simulation. https://or.stackexchange.com/questions/6723/significant-bias-introduced-into-simple-simulation/6724#6724 . Perhaps we need ASI or AST (Artificial Stochastic Intelligence or Artificial Stochastic Thinking), which seems to be absent from the AI now being used in the real world. ASI or AST would be cognizant of these phenomena. – Mark L. Stone Aug 29 '23 at 15:02
  • I saw this post, but this point of view and actually ASI and AST are fascinating. Thanks for mentioning that. – A.Omidi Aug 30 '23 at 05:14