Chemists use many sources of information to do this.
Part of being trained as a chemists is learning, in general terms, what kinds of reactions are known and what their best uses are. That training can go 5-10 years beyond a bachelors degree and is very intensive.
Beyond what you can keep in your head, one of the most important information sources is the "primary literature", consisting of:
- Journal publications
As an example, see this patent about ethanol manufacture.
Alternatively, see this very recent journal publication dealing a new catalytic reaction.
Having all of this information out there is a start, but it's a vast collection spread out over many publishers and 100 years or more. To find what you need, it's essential to have a good indexing system. This allows you to search large databases of reactions and molecules for references that exactly match, or which are close to what you need.
This last part - being able to reason by analogy - is critical to being able to get the most out of the primary literature. Maybe I don't find an exact match for the exact reaction I want to do. But I find something pretty close. Here's where my training and reasoning ability come in to either decide to adapt what I've found or keep looking for something better.
An example of one of these search systems is Chemical Abstracts Service, but there are others.
As a professional chemist, you're expected to keep up with all of the main journals in your field so you stay up to date. You do this by reading articles, talking to people, giving talks, and going to conferences.
There's another category of information consisting of 'reviews'. These are articles that summarize a large area of chemistry - for example, recent large-scale methods for the production of ethanol.
The discovery/invention of new reactions is a very active area of current chemical research.