I'm putting together a dataset to use at a lithography equipment manufacturer and need to figure out what kind of data they might be collecting. Do they typically collect data on equipment that's installed at customer sites? Would they have access to it? I was told that the typical dataset has 10-20K observations and 30 or so columns. Rows correspond to parts (which I'm guessing means IC devices, although I suppose it could also be test structures). I was also told that the columns include failure mode, part, part type, supplier (what besides photoresist? coater manufacturer? Would other folks build subassemblies inside the steppers?). I suspect they may also be collecting inline and offline data, but I'm not sure how for installed equipment. Any input would be appreciated.