Journal
The cycle-time board and what the line actually did
Boards at the line often show a cycle time that was timed on a good hour, on a different SKU, or before a guard was added. Operators work to the board because that is what the last audit asked for. Output for the shift then looks like a failure of people rather than a failure of the number.
Time thirty consecutive cycles on the SKU that currently fills the orders. Discard the first few after a restart. Write the median and the slowest without the genuine stoppages. If the board is faster than the median, the board is a wish. If the board is slower, someone has padded it to protect a bonus or to hide a known wait.
Throughput for the factory manager is pieces to the warehouse, not cycles at the loudest machine. A fast cycle feeding a blocked packer is not a fast line. Walk the piece from the constraint to the pallet. The minutes on the floor between those two points belong in the same conversation as the board.
When we draw this for a visit, we put the board number, the timed median, and the packed output on one strip. The argument then has three facts instead of a slogan about ‘discipline’.