01
Establish the baseline with explicit counting
If approval arrives on Monday, the three production days are Tuesday, Wednesday and Thursday. The job completes at the end of Thursday under the supplied rule, leaving Friday before the collection deadline. Describe that as one working day of schedule buffer. It is not a measured guarantee about a real shop. The calculation depends on the next-working-day start rule and the absence of holidays. Keep those assumptions beside the dates so another reader can reproduce the count without silently using a different definition of three days.
02
Move approval and identify the changed consequence
If approval arrives on Tuesday, production runs Wednesday, Thursday and Friday. Completion now falls at the stated collection deadline, so the model has no working-day buffer. If approval instead arrives on Wednesday, production runs Thursday, Friday and Monday, finishing after the planned collection. These scenarios change only the approval day. They do not establish which approval date is likely or whether a shop would offer expedited work. Record the consequence as conditional on the supplied duration rather than describe a possible delay as an event that has already occurred.
03
Turn the scenario into a reviewable dependency record
The reusable record contains deadline, approval status, production duration, start rule, working calendar, completion date, buffer and excluded activities. Add a named person responsible for confirming the approval date in a real project. If shipping or rework becomes relevant, revise the model rather than assume it fits inside the existing buffer. A useful review question is whether the duration and collection cutoff have been confirmed. This keeps the worksheet focused on a specific dependency and avoids converting a simple schedule illustration into an unsupported overall risk score.