A baseline you can rely on: the 14 quality checks
Abstract
The baseline is what every later progress report and delay is measured against. A walk through the DCMA 14-point assessment — logic, constraints, float, durations and the critical path test — and the mistake each threshold exposes.
The baseline is the yardstick every later progress report and delay claim is measured against. If its logic is weak, everything built on it inherits the weakness. The DCMA 14-point assessment is a widely used checklist for testing one; these are the checks that matter most.
Logic
- Missing predecessors or successors: apart from the project start and finish, every activity should be tied into the network. Threshold: 5%.
- Leads (negative lags): none. They hide a relationship that should be stated plainly.
- Lags: at most 5% of relationships. Real waiting time, such as concrete curing, is better modelled as its own activity.
- Relationship types: at least 90% finish-to-start.
Constraints and float
- Hard constraints (Must Start/Finish On): at most 5%. A hard constraint cuts the logic and makes the critical path artificial.
- High float: activities with more than 44 working days of total float usually point to a missing link; at most 5%.
- Negative float: none in a baseline.
Durations and resources
- High duration: at most 5% of activities over 44 working days — a long activity cannot be controlled well.
- Invalid dates: an actual date in the future or a forecast in the past is an update error.
- Resources: activities should carry resources or cost so the schedule reconciles with the estimate.
Performance tests
- Critical path test: delay a critical activity artificially; the finish date should move by the same amount.
- CPLI and BEI — the critical path length index and the baseline execution index — both at least 0.95.
These numbers are not rules; they are thresholds that show where to look harder. A schedule that passes them can be relied on for control, and defended when a delay has to be analysed.