Idle time
The idle time at a transition matrix is used to make the calculation of setup and changeover time dependent. When calculating the transition for a batch, the next batch in the sequence is taken into consideration only if the time between the batches is not greater than the idle time.
If the time between the batches is greater than the idle time, the next batch in the sequence is ignored. The calculation is performed as if there was no next batch. The transition value NONE in the transition matrix is used.
If you want the idle time to be ignored by the transition calculation, specify an idle time of 0:00. Consequently, the transition calculation depends only on the sequence of batches. The time between the batches is not taken into consideration.
The time between two consecutive batches is calculated based on the end of the outflow of the first batch, and the start of the inflow of the second batch.
Example
At a paint factory, cans of paint are filled at the filling station. Normally, when switching between colors, flushing the filling station with the new color for one minute suffices. However, if the filling station has been idle for 30 minutes, the paint inside the station is starting to dry. 15 minutes are then necessary to flush all the paint out. Therefore, the changeover is not only dependent on the sequence of the batches, but also on the time between two consecutive batches.
This example can be modeled with a transition matrix with these parameters:
- The transition value from one color to the next is one minute.
- The transition value from a color to NONE is 15 minutes.
- The idle time is 30 minutes.
If 30 minutes or less elapse between two batches, the transition time of one minute is used. However, more than 30 minutes elapse between two batches, the transition is considered as a transition to NONE, and the time of 15 minutes is used.
If you use a solver set to minimize changeover, the solver tries to keep batches within 30 minutes from each other, in order to prevent the 15-minutes penalty time. In general, though, the objective span or distance to start work area is a better way to achieve this solution, because time-dependent changeovers are more difficult for the solver.