From a cursory search there don’t appear to be any good, fleshed-out video games for training in the trades. There are a couple that look promising at first glance (HVAC simulator on Steam) but then you look a little deeper and they don’t have any content. Funding issues, most likely. Everybody agrees it’s a good idea to teach the youth, but when you pass the hat around it always comes back empty because there’s crusades and there’s business, and the only legal business in America is employee surveillance. Hyperbole, but 80% accurate.
It occurs to me the best use of this little side project, other than appreciating the pure autism of it, would be to encode its logical representation of machines and components into a video game engine for simulating any possible assembly, installation, and maintenance task. Well, anything that can be represented by logical operations like “weld this”- actual welding and other “touchy” tasks could be mini-games separate from the main game logic. The most obvious choice of overall UI would be a point-and-click adventure using something like Adventure Game Studio, because something like this procedure maps very easily:
(Each of those steps is pretty intricate and has a flowchart of its own.)
I figure the only way a plan like this would work is to make it easy to expand, do a cool but very small proof of concept, and let it out into the wild. That’s already well beyond my time and abilities probably, but it’s a thought.
Aeoli Pera 
The difficult part is the gamification of solving the puzzle. How do you turn HVAC into the new Myst?
I’ve been a tech and very much enjoy the troubleshooting; so it can be virtualized. Doing virtual twins now; real time physics of HVAC is pretty simple compared to a shooter. There are a lot fewer elements in the simulation.
>How do you turn HVAC into the new Myst?
I was figuring a simple life management game where completing jobs gets you money and screwing up costs money (e.g. refrigerant spills cost you the EPA fee). That aside, I believe we’re entering a time when people will be proactively looking for fast, cheap ways to learn and practice handyman skills, so gamifying it won’t be strictly necessary. Low trust means cars breaking down, electricity going out, computers failing, so you need to be a bit of a mechanic, electrician, repair tech, etc.
If you search for “video game teach programming” you can find a million hits but for the trades there’s always some half-assed Unity VR shit with one level and nothing else.
Do it like an aircraft simulator then. Make a list of faults, randomly select fault, tradesman trouble-shooting (with or without process); buying parts to fix it costs money…that’s lost, you make money fixing the problem and selling parts that fix the problem.
Digital twin the various systems and there’s your game.
The troubleshooting trees are boring and don’t allow people to explore the system.
The aircraft sim thing is a good idea.
>The troubleshooting trees are boring and don’t allow people to explore the system.
I’m not a big fan of sandbox games in general, but that aside the concept doesn’t benefit from exploration elements at all. But it’s pretty easy to add tension to a maintenance job by putting it on a timer, which is also pretty realistic. The main difference between a game and real life is you can get so fast at the individual steps it gets you into a megalomaniacal flow state.
Each level can have increasingly complex HVAC systems and potential faults lists.
Papers Please could be a good spearhead for designing games like that. It’d work especially well in this way if you could get away with depicting the various functions with pixel art, since that’s cheap to make.
Automotive mechanics have a par time to do any job. They get paid the same regardless of how fast or slow they work. If the game had a fixed length, say 10min, and each fault had a fixed par time to repair; then it would encourage faster troubleshooting. It would not encourage following the troubleshooting tree.
That’s fine, the main thing that needs training is the mechanical steps of, say, reclaiming refrigerant or replacing a head gasket.
Use Renpy with 2D assets. Designing 3D models for everything will cost a ton.
Oh, it would be pixel art for sure.