There was a documentary titled (IIRC) "Space Shuttle Garage", shown in 2003 2 or 3 months prior to the loss of Columbia. Gone a bit fuzzy on which channel, I'm fairly certain it was either The History Channel or Discovery Channel.

It was all about the process the Shuttles went through between flights. I didn't get to see all of it. The two parts I saw covered how replacement tiles were made and a crew *failing* to get either of the special wheel nut torque wrenches to work properly.

They didn't keep doggedly working at it until they got the wheel nuts torqued properly, they stuck a note in the huge procedure manual for the next shift. At the time I had expected they would've upgraded to something like laptops with all the documentation digital for easy searching and the ability to add notes which would automatically alert other shifts about problems and what needed to be done. Nope, still using big honking binders full of paper that would take ages to flip through to find information.

As for making the tiles, that was also an all analog affair. I'd expected that by then they would have digitized all the tile shapes for carving them out using CNC milling machines. Nooooo.
First step was to identify the damaged tile, either by reading one of the several ID numbers stenciled on each tile, (several copies of the number in case of erasure from damage) or by referencing an unreadable tile from ones around it.
Second step was to pull the tile's index card from a very old school wooden card catalog cabinet, exactly like the ones libraries had mostly done away with in the late 1990's.
Third step, take the card to the tile template room where rows an rows of shelves held full scale physical models of every tile for every Orbiter, no two identical, and retrieve the correct one.
Fourth step was to use a bandsaw to rough cut a block of the silica foam to size.
Fifth step was mounting the block inside a chamber with a vacuum hose to collect dust. Outside the chamber was a cylindrical probe with a rounded end, attached to a manually operated 3D pantograph. On the other end of the pantograph, inside the chamber, was a matching cylinder, coated with diamond abrasive, spun by a high speed motor. The operator would run the probe all over the template to carve the foam.
Sixth step, after passing visual inspection, was hand dipping the face of the tile into a white ceramic coating, followed by firing in a kiln. If the tile was black it got a second coat of the black ceramic.
Seventh step was getting the number stencil out of a second wooden card catalog then using it to airbrush the tile number onto the ceramic.

I figured I'd see the show again because the channel it was on reran most of their shows many times. But apparently the loss of Columbia prompted shoving this one down the "memory hole". AFAIK it was never shown again, likely because it showed that the high tech Shuttle Orbiters were serviced with a process frozen circa 1980. That procedure binder, if someone dropped it and lost some of the bookmarks and notes, eek! That could set back processing a Shuttle who knows how long, because everything would have to be double checked. I was also thinking of how much time could be saved making replacement tiles if they were 3D digitized and the foam could be carved by putting a block of the foam into a CNC mill then selecting the tile number and clicking GO. Even putting the template room card catalog into a database would've saved a significant amount of time VS having to physically look for and pull cards to find the template locations in the template warehouse. Put that computer at the warehouse door so people could poke in the tile number and it'd tell them the section, row, shelf etc.

If NASA wants money they could sell sets of the tile templates, the catalog cards and the stencil cards. Put them into nice presentation boxes for $1500 a set. How many thousand sets would they have?