Schedule baseline change control and when to rebaseline
The baseline schedule is the target that performance is measured against, so changes to it go through a documented change control process. GAO limits baseline changes to new or reduced scope or formal replanning, and says a rebaseline should be rare.
Baseline versus current schedule
GAO describes the baseline schedule as the agreed plan: the sequence of events, resource assignments and acceptable dates for key deliverables. The current schedule is updated each period with actual progress. Performance is measured by comparing the two.
Both are subject to change control. GAO warns that without a documented, consistently applied process, staff may keep revising the schedule to match performance. That hides the project's true performance, gives stakeholders inconsistent versions of the plan and produces inaccurate variance reports.
Statusing is not revising
Statusing updates the schedule with actual data: actual start and finish dates and progress. Revising changes future work, for example resequencing or adding activities. GAO says the two should not be confused. A revision is handled as replanning or rebaselining, depending on the program's change control approach.
How much of the schedule is under change control depends on size, scope, risk and complexity. A simple project may control only key milestones. On a complex, tightly controlled schedule, even network dependencies may need approval. GAO cites a case where corrections of known logic errors went through change control.
When rebaselining is right
Rebaselining restores management control when the remaining baseline is no longer realistic enough for performance measurement. It normally wipes historical variances so new ones show quickly. GAO says it should be rare. A program rebaselined often may not understand its scope, or may be unable to produce realistic estimates, and that is what to examine before rebaselining again.
Once management approves the current schedule, give it a version number and archive it, so every stakeholder works from the same version.
Where the exam catches people
Believing late activities justify moving baseline dates to match: that hides true performance; baseline changes go through change control.
Believing resequencing unstarted work is part of the monthly status update: that is a revision, handled as replanning or rebaselining.
Believing repeated rebaselining fixes a slipping schedule: GAO says frequent rebaselining suggests poorly understood scope or unrealistic estimates.
Believing obvious corrections can skip change control: on a tightly controlled schedule, even logic fixes can require approval.