Picture an empty one-hour machine slot. Two small jobs each need twenty-five minutes of processing. The calendar looks almost generous: fifty minutes of work, ten minutes left over.
In this fictional rehearsal, moving from the first job to the second also requires twenty minutes of setup. The machine starts ready for job A. The same machine performs the changeover, so setup cannot overlap either job. Nothing else interrupts the slot.
Processing and changeover occupy seventy minutes together. The sixty-minute opening is ten minutes short.
We would put that missing twenty minutes beside the processing estimate before asking an automated scheduler to fill the calendar. A correct sum of the job times can still leave part of the work outside the estimate.
Write the sequence, not just the total
For the invented slot:
Job A processing: 25 minutes.
Changeover from A to B: 20 minutes.
Job B processing: 25 minutes.
Total machine occupancy: 70 minutes.
Available continuous slot: 60 minutes.
Gap: 10 minutes.
These assumptions matter. If a different sequence changes the setup requirement, that is another case to evaluate. If preparation can happen elsewhere without occupying this machine, record that separately. Neither possibility has been established in the example.
Keep an unknown setup time unknown. Quietly entering zero would turn missing information into a claim that no time is needed.
Give the proposed schedule a question
A useful output says which interval is occupied, what starts ready, and where the setup sits. The reviewer can then examine a different slot or sequence using the operation’s existing planning process.
It is less useful to put a confident finish time on the calendar and leave the next operator to discover that the jobs require different tooling. The calendar entry should retain the assumption that made its estimate possible.
Before choosing a scheduling tool, give it this fictional case. Does its proposed sequence occupy seventy minutes or only fifty? Keep the result with the assumptions above. It tells you how that output handled one case; it cannot establish how the tool will perform across an operation.
Our earlier rehearsal, Count the correction time, applies the same habit to an automation trial: include review and repair in the time you record.
https://www.rebuildshow.com/p/count-the-correction-time
Take one upcoming machine slot and separate setup from processing in the estimate. Review whether the whole sequence fits under its stated assumptions before promising a finish time.
The Rebuild Editorial


