Practical use of Artificial Intelligence: Making Beer

In other articles I have discussed artificial intelligence in games and the cognitive types.  Another practical use of artificial intelligence is to use it to help in brewing beer.  An article (Threapleton and Wilson),, describes the use of neural networks and AI in the creation of beer.  You can try the solution, and if you go to the pages:, you can download some powerpoints about the Neurosolutions product, which has received the Vista Ready Logo!  Good work Neurosolutions! 

How does it work?  Well this is more along the lines of a perceptive software, it makes the use of neural networks and Bayesian equations to emulate the human brain (or so the marketing statement says).  That isn’t true, if you read the article, it is actually a process that uses procedures that a human would use.  This is an example of an AI that does something that is productive, but the AI is not cognitive.

Let’s take a look at the Neurosolutions and creating beer and compare the process to classical AI:


Classical “Cognitive” AI

Neurosolution Beer and AI

All behaviors must be representable in the system.

Analytical inputs were identified and training data was diverse

Therefore, the system should either be able to construct arbitrary automata or to program in some general purpose programming language.

MatLab (sadly) was used for the general purpose language

Interesting changes in behavior must be expressible in a simple way.

Use of Excel or similar spreadsheet was used to express findings for the brewmaster

All aspects of behavior except the most routine must be improvable. In particular, the improving mechanism should be improvable.

A burnt flavor, for example, would be controlled during the brewing process via changes generated by the software program. Isn’t clear on how this works, so I give it a fail (anyone disagree?)

The machine must have or evolve concepts of partial success because on difficult problems decisive successes or failures come too infrequently. The system must be able to create subroutines which can be included in procedures as units. The learning of subroutines is complicated by the fact that the effect of a subroutine is not usually good or bad in itself. Therefore, the mechanism that selects subroutines should have concepts of interesting or powerful subroutine whose application may be good under suitable conditions.

If the analysis is kept to a limited number of flavors, the system is predictive.  However, in general it appears that the neurosolutions fails, but mostly due to the ability for the sensor input as well as chemical control.


So is the Neurosolution a cognitive AI? No, but it is a pretty smart system that allows a beer maker to produce a complex chemical more efficiently.  This conserves resources and creates profit for companies, all a good thing.

Please understand, drinking excessively or drinking and driving is wrong, beer should be drank in moderation.  Beer has long been a part of human society, this article is discussing the use of AI in creating a complex chemical product, do not read this as approval of excessive drinking.