Schedule Development and Critical Path Management
Schedule questions test network logic, float arithmetic, and compression choices. Most misses come from confusing total float with free float, or from crashing a path that is not critical.
Core rules
- Precedence diagramming dependencies: finish-to-start (most common), start-to-start, finish-to-finish, start-to-finish; leads and lags modify them. A lag is waiting time, not an activity, and needs justification.
- Forward pass gives early start and finish; backward pass gives late start and finish; float = LS minus ES, also LF minus EF. The critical path is the longest path through the network and has zero total float in a single-project network.
- Total float is shared along a path; free float is how long an activity can slip without delaying the successor's early start.
- Crashing adds resources to critical-path activities, costs money, and shows diminishing returns; fast tracking overlaps activities and adds rework risk. Both shorten the project only when applied to the critical path.
- Resource leveling can extend duration and change the critical path; smoothing uses existing float and should not.
- Milestones have zero duration. The schedule baseline is the approved version used for variance analysis.
Worked example
A (3 days) precedes B (4 days) and C (2 days) in parallel; D (5 days) needs B and C; E (1 day) follows D. Forward pass: A 0-3, B 3-7, C 3-5, D starts at max(7,5) = 7 and ends at 12, E 12-13, so duration is 13 days. Backward pass: E 13/12, D 12/7, B 7/3, C 7/5, A 3/0. Float on C is 5 minus 3, or 2 days. Critical path is A-B-D-E. Crashing C buys nothing; crashing B by two days shortens the project, though C becomes critical once its float is used up.
Traps
Crashing the cheapest or shortest activity rather than a critical one. Reporting total float as free float. Adding lags to paper over logic errors. Forgetting that compression shifts the critical path.