3.3 Schedule Management
Key Takeaways
- The critical path is the longest path through the network and sets the minimum project duration; its activities have zero total float
- Total float is delay tolerance versus the project end date; free float is delay tolerance before disturbing a successor's early start
- Crashing adds resources and raises cost; fast-tracking parallels activities and raises risk
- Finish-to-Start is the most common dependency; Start-to-Finish is the rarest and most confusing
- PERT expected duration = (O + 4M + P) / 6; PERT standard deviation = (P - O) / 6
The Six Schedule Processes
Schedule management is the most calculation-heavy area of the CAPM exam. Master the process flow, then the math.
| Process | Process group | Key output |
|---|---|---|
| Plan Schedule Management | Planning | Schedule management plan |
| Define Activities | Planning | Activity list, milestone list |
| Sequence Activities | Planning | Network diagram |
| Estimate Activity Durations | Planning | Duration estimates |
| Develop Schedule | Planning | Schedule baseline, project schedule |
| Control Schedule | Monitoring & Controlling | Schedule forecasts, change requests |
Logical Relationships (Dependencies)
| Type | Meaning | Example |
|---|---|---|
| Finish-to-Start (FS) | B starts only after A finishes | Pour foundation, then build walls |
| Start-to-Start (SS) | B starts only after A starts | Start coding, then start documenting |
| Finish-to-Finish (FF) | B finishes only after A finishes | Finish writing, then finish editing |
| Start-to-Finish (SF) | B finishes only after A starts | New shift starts, then old shift ends |
FS is by far the most common; SF is rarest. Dependencies are also classified as mandatory (hard logic, required by the work), discretionary (soft logic, preferred practice), external (e.g., a permit), or internal (within team control). Leads pull a successor earlier (FS minus 2 days); lags push it later (FS plus 2 days).
Critical Path Method
The critical path is the longest path through the network; it sets the project's minimum duration. Its activities have zero total float, so any slip delays the whole project. A project may have multiple critical paths, which raises risk.
Worked example:
| Path | Activities | Duration |
|---|---|---|
| 1 | A → B → D → F | 3 + 5 + 4 + 2 = 14 |
| 2 | A → C → E → F | 3 + 2 + 6 + 2 = 13 |
| 3 | A → B → E → F | 3 + 5 + 6 + 2 = 16 (critical) |
Critical path = 16 days. Total float of Path 1 = 16 − 14 = 2 days; Path 2 = 16 − 13 = 3 days.
Float (Slack)
| Type | Definition |
|---|---|
| Total float | Delay allowed without delaying the project end date |
| Free float | Delay allowed without delaying the early start of any successor |
| Project (positive) float | Slack between the imposed deadline and the critical-path finish |
Total float = Late Start − Early Start = Late Finish − Early Finish. A forward pass computes early dates; a backward pass computes late dates.
Duration Estimating
| Technique | How it works | Best when |
|---|---|---|
| Analogous | Scale from a similar past project | Little detail; quick top-down |
| Parametric | Rate × quantity (e.g., 10 hrs/unit × 50) | Quantifiable, repeatable work |
| Three-point | Combine optimistic, most likely, pessimistic | Uncertainty is high |
| Bottom-up | Estimate each work package, then sum | Detail available; most accurate |
Three-point math (memorize):
- PERT (beta): Expected = (O + 4M + P) / 6
- Triangular: Expected = (O + M + P) / 3
- PERT standard deviation: σ = (P − O) / 6
Example: O = 4, M = 8, P = 18 → PERT = (4 + 32 + 18) / 6 = 9 days; σ = (18 − 4) / 6 ≈ 2.3 days.
Compression and Resource Optimization
| Technique | Mechanism | Trade-off |
|---|---|---|
| Crashing | Add resources to critical-path tasks | Higher cost |
| Fast-tracking | Run sequential tasks in parallel | Higher risk, possible rework |
| Resource leveling | Smooth over-allocation | May extend the critical path |
| Resource smoothing | Adjust within available float | Cannot extend critical path |
Exam tip: Both compression techniques target the critical path. Crash when budget allows risk-aversion; fast-track when schedule pressure outweighs rework risk.
Forward Pass, Backward Pass, and Reading Floats
The network calculation has two halves. The forward pass moves left to right, adding durations to compute each activity's early start (ES) and early finish (EF); it tells you the soonest each task can happen. The backward pass moves right to left from the project finish, subtracting durations to compute late start (LS) and late finish (LF); it tells you the latest each task can happen without delaying the project. Total float for any activity is then LS − ES (equivalently LF − EF). Critical-path activities, by definition, have ES = LS and EF = LF, so their float is zero.
A frequent exam wrinkle: an imposed deadline earlier than the calculated finish creates negative float, signaling the schedule is already infeasible and must be compressed.
Estimating Pitfalls and Reserves
Watch the difference between effort and duration: 40 hours of effort for one person is roughly a one-week duration, but two people might compress it — subject to whether the work can be parallelized. Padding individual estimates is discouraged; instead, identified schedule risk is handled with contingency reserve (buffer for known risks, inside the baseline) and unknown risk with management reserve (outside the baseline).
Three-point estimating gives you a defensible range rather than a single optimistic guess, and the PERT standard deviation lets you express confidence — roughly 68% of outcomes fall within one σ of the expected value under a normal approximation.
Schedule Network Diagram Conventions
Most modern scheduling uses the precedence diagramming method (PDM), also called activity-on-node, where boxes are activities and arrows are dependencies. This is what supports the four relationship types (FS, SS, FF, SF) and leads and lags. An older approach, activity-on-arrow, supports only finish-to-start. Milestones are zero-duration markers used to flag key checkpoints such as a phase gate or a customer review; because they consume no time, they never appear on the critical path as duration but can still anchor dependencies.
Control Schedule in Practice
During execution, Control Schedule compares actual progress to the schedule baseline, often using SPI from Earned Value, and produces schedule forecasts and change requests. If the project falls behind, the manager evaluates crashing versus fast-tracking, models the trade-offs, and routes any baseline change through integrated change control — the schedule baseline is never edited informally. Remember that resource leveling can lengthen the critical path because it respects resource limits, whereas resource smoothing stays within available float and therefore cannot.
Three paths exist: Path A = 18 days, Path B = 22 days, Path C = 20 days. What is the critical path, and what is the total float for Path A?
Using PERT with Optimistic = 4 days, Most Likely = 8 days, and Pessimistic = 18 days, the expected duration is:
Which compression technique raises project RISK rather than cost?
Which dependency type is the MOST commonly used in scheduling networks?