The issue? A really long dead loop time.
"The time delta between the moment the strategy opens a given loop's valve [and] starts the pump+furnace and the moment a temperature increase is measured can sometimes be on the order of an hour," tohox notes.
So how can overshoot and oscillations be minimized, or more fundamentally, is a PID loop the best control option in a case like this?
Tohox also asks whether inputs must come from analog I/O in the system, or whether they can come from another source, like floating point variables within a table.
Take a look at the answers and expand your knowledge about PID loops with this real-world example from the OptoForum.