PMP Schedule Management: Critical Path, CPM & Exam Qs

I will be honest: schedule management was the knowledge area where I had to rewire my instincts the most. Before the PMP I treated schedules as Gantt charts in a tool. After studying pmp schedule management properly I realised the Gantt is just the visible tip of a deep iceberg of activity definition, sequencing, estimation, network analysis, and continuous control.

In this guide I want to share what I learned about schedule management in a practical way. We will walk through the six processes, work through a critical path method (CPM) example by hand, calculate float, look at schedule compression techniques, and discuss resource levelling. I will also give you ten exam-style questions with worked answers and call out the traps that catch most candidates.

By the end you should be able to draw a network diagram, find the critical path, calculate total and free float, and confidently answer the schedule-related situational questions that show up in roughly one in seven PMP exam questions.

Schedule Management is the set of processes that ensure the project is completed in a timely manner. It covers everything from defining individual activities to monitoring the schedule and managing changes to it.

The knowledge area has five planning processes and one monitoring and controlling process. There is no executing process - because executing the schedule is part of Direct and Manage Project Work in Integration Management.

I think of schedule management as the choreography of work. It is not just about when things happen; it is about how dependencies, durations, resources, and constraints interact to produce a deliverable schedule.

The flow is largely linear during planning, with iteration as estimates change and constraints become clearer.

This process defines how the schedule will be developed, managed, executed, and controlled. The output is the schedule management plan, a subsidiary of the project management plan.

The schedule management plan addresses:

• Project schedule model development (which methodology, which tool)
• Release and iteration length for adaptive projects
• Level of accuracy and units of measure
• Organisational procedures links
• Control thresholds (e.g., 10% SV variance triggers escalation)
• Rules of performance measurement (e.g., earned value rules)
• Reporting formats

A good schedule management plan eliminates a hundred small arguments later.

Define Activities decomposes work packages from the WBS into the specific activities required to produce deliverables.

Outputs:

• Activity list - all schedule activities required for the project
• Activity attributes - extended description of each activity (predecessors, successors, resources, constraints, assumptions)
• Milestone list - significant points in the project (zero duration)

Tools include decomposition, rolling wave planning, and meetings.

Remember: WBS work packages are decomposed into activities here. The exam tests this transition often.

Sequence Activities identifies and documents relationships among project activities. The output is the project schedule network diagram.

Logical relationships:

• Finish-to-Start (FS) - most common; predecessor finishes before successor starts
• Finish-to-Finish (FF) - predecessor finishes before successor finishes
• Start-to-Start (SS) - predecessor starts before successor starts
• Start-to-Finish (SF) - rarest; predecessor starts before successor finishes

Dependency types:

• Mandatory (hard logic) - inherent in the work (you must pour the foundation before framing)
• Discretionary (soft logic) - preferred logic, often based on best practice
• External - relationship with non-project activities (e.g., regulatory approval)
• Internal - within project team's control

Leads and lags:

• Lead - successor starts earlier; written as FS-3 (start 3 days before predecessor finishes)
• Lag - successor starts later; written as FS+5 (wait 5 days after predecessor finishes)

This process estimates the number of work periods needed to complete each activity. Key techniques:

• Analogous estimating - top-down, based on historical similar projects (fast, less accurate)
• Parametric estimating - statistical relationship between variables (e.g., 100 sq m at 5 hours/sq m = 500 hours)
• Three-point estimating - triangular ((O+M+P)/3) or beta/PERT ((O+4M+P)/6) using optimistic, most likely, pessimistic values
• Bottom-up estimating - aggregate from detailed estimates of components (slower, more accurate)
• Data analysis - reserve analysis (contingency for known unknowns)

I default to bottom-up for high-stakes activities and analogous for early-phase rough estimates.

Three-point example. Activity X has O=4 days, M=6 days, P=14 days.

• Triangular: (4+6+14)/3 = 8 days
• PERT/Beta: (4 + 4*6 + 14)/6 = 42/6 = 7 days
• Standard deviation: (P-O)/6 = (14-4)/6 = 1.67 days

Develop Schedule analyses activity sequences, durations, resource requirements, and constraints to create the schedule model. Tools include:

• Schedule network analysis - the umbrella technique
• Critical path method (CPM) - calculates early/late start and finish, identifies critical path
• Resource optimisation - levelling and smoothing
• Data analysis - what-if scenarios, simulation (Monte Carlo)
• Leads and lags
• Schedule compression - crashing and fast-tracking
• Project management information system (PMIS) - tools like MS Project, Primavera, Smartsheet
• Agile release planning - for adaptive projects

Outputs include the schedule baseline, project schedule, schedule data, and project calendars.

Control Schedule monitors status, manages changes to the schedule baseline, and prevents unauthorised changes. Tools include data analysis (earned value analysis, iteration burndown chart, performance reviews, trend analysis, variance analysis, what-if scenario analysis), critical path method, project management information system, resource optimisation, and leads and lags.

Schedule changes route through Integrated Change Control. Control Schedule identifies the variance; Integrated Change Control approves the response.

Let us work through a CPM example. Consider seven activities:

 

Step 1: Forward pass (calculate early start (ES) and early finish (EF)).

• A: ES=0, EF=4
• B: ES=4, EF=10
• C: ES=4, EF=7
• D: ES=10, EF=15
• E: ES=7, EF=11
• F: ES=max(15, 11)=15, EF=17
• G: ES=17, EF=20

Project duration = 20 days.

Step 2: Backward pass (calculate late finish (LF) and late start (LS)), starting from the end.

• G: LF=20, LS=17
• F: LF=17, LS=15
• E: LF=15, LS=11
• D: LF=15, LS=10
• C: LF=11, LS=8
• B: LF=10, LS=4
• A: LF=4, LS=0

Step 3: Total float = LS - ES (or LF - EF).

• A: 0
• B: 0
• C: 8-4 = 4
• D: 0
• E: 11-7 = 4
• F: 0
• G: 0

Critical path: A-B-D-F-G with zero float. Total duration: 20 days.

The critical path is the longest path through the network, and any delay on it delays the project.

Three types of float to know:

• Total float = LS - ES (or LF - EF). The amount of time an activity can be delayed without delaying the project end date.
• Free float = ES of next activity - EF of current activity. The amount of time an activity can be delayed without delaying any successor's early start.
• Project float = imposed end date - calculated end date. The amount of time the project can be delayed without missing the customer's deadline.

In the above example, C has total float of 4 days. Its EF is 7. The successor F has ES of 15 (driven by D, not C). So C's free float = 15 - 7 = 8 days? Not quite. C's immediate successor is E (since E depends only on C). E's ES = 7. So free float for C = 7 - 7 = 0 days. Total float of 4 means it can be delayed without delaying the project, but doing so delays E.

When the schedule must be shortened, two techniques are available.

Crashing adds resources to critical path activities to reduce duration. It always increases cost. Use it when:

• The activity is on the critical path
• Additional resources can meaningfully reduce duration (not all work scales)
• The cost trade-off is acceptable

Fast-tracking performs activities in parallel that would normally be sequential. It often increases risk and may add rework. Use it when:

• Activities have discretionary (not mandatory) dependencies
• Rework risk is manageable
• Cost increase is undesirable

Exam tip: if the question emphasises cost containment, fast-track first. If quality and risk are paramount, crash first.

Both are forms of resource optimisation, but they differ in priority.

Resource levelling adjusts the schedule to balance resource demand against supply. It typically extends the schedule because some activities cannot proceed without their required resources. Use when resources are constrained.

Resource smoothing adjusts activities within their available float so resource limits are not exceeded, but the critical path is not extended. Use when the end date is fixed and small reshuffles fit within float.

A typical workflow on my projects:

1. Develop the initial schedule using CPM
2. Identify resource over-allocations
3. Apply resource smoothing first (uses float)
4. If smoothing is insufficient, apply resource levelling (may extend schedule)
5. Re-baseline if necessary

Q1. What is the critical path?

A. The most expensive path
B. The longest path through the network with zero float
C. The path with the most activities
D. The path with the most risk

Answer: B.

Q2. If activity A has ES=5, EF=10, LS=8, LF=13, what is its total float?

A. 0
B. 3
C. 5
D. 10

Answer: B. Total float = LS - ES = 8 - 5 = 3.

Q3. Adding resources to a critical path activity is called:

A. Fast-tracking
B. Crashing
C. Resource levelling
D. Resource smoothing

Answer: B.

Q4. Which dependency type is rarest?

A. Finish-to-Start
B. Finish-to-Finish
C. Start-to-Start
D. Start-to-Finish

Answer: D.

Q5. Three-point PERT estimate with O=5, M=8, P=17 is:

A. 10
B. 9
C. 8.5
D. 11

Answer: B. PERT = (5 + 32 + 17)/6 = 54/6 = 9.

Q6. Resource levelling typically:

A. Shortens the schedule
B. Has no effect on the schedule
C. Extends the schedule
D. Increases cost only

Answer: C.

Q7. The project schedule network diagram is produced by which process?

A. Define Activities
B. Sequence Activities
C. Develop Schedule
D. Estimate Activity Durations

Answer: B.

Q8. Free float is the time an activity can be delayed without delaying:

A. The project end date
B. The successor's early start
C. The predecessor's late finish
D. The next milestone

Answer: B.

Q9. A lag is best described as:

A. Waiting time between activities
B. An activity starting earlier than predecessor finish
C. An activity that is late
D. A critical path activity

Answer: A.

Q10. Which estimating technique uses optimistic, most likely, and pessimistic values?

A. Analogous
B. Parametric
C. Three-point
D. Bottom-up

Answer: C.

The traps I see most often:

• "Crashing always works" - false. It only works on critical path activities and only when adding resources reduces duration.
• "Fast-tracking is free" - false. It increases risk and may require rework.
• Confusing total float with free float - total = project; free = successor.
• Assuming the critical path never changes - it can change as activities slip or estimates update.
• PERT vs triangular - triangular is simple average; PERT/beta weights the most likely.
• Treating milestones as activities - milestones have zero duration.
• Schedule baseline updates without change control - never. All baseline changes route through Integrated Change Control.

Let me share three real scheduling lessons that reinforced these concepts for me far better than any textbook.

Story 1: The disappearing critical path. On a regulatory programme I led, the critical path ran through compliance certification by an external body. We planned a 60-day certification window. Three weeks in, the certifier responded faster than expected. Suddenly our critical path shifted from certification to user training. Because I had not been tracking near-critical paths, we missed two weeks of preparation time and ended up crashing training at significant cost. The lesson: always monitor near-critical paths, because today's float is tomorrow's critical activity.

Story 2: The fast-tracking that backfired. During a banking platform launch we fast-tracked UI development in parallel with API specification. The API specification then changed twice, requiring extensive UI rework. The total effort cost more than if we had sequenced the work. The lesson: fast-track only when downstream change risk is genuinely low. Soft logic dependencies are soft for a reason.

Story 3: The estimate that was right. A junior PM on my team was pressured by a sponsor to commit to an aggressive timeline based on analogous estimating. She held firm, asked for a week to do bottom-up estimating with the team, and produced a defensible estimate 30% longer than the sponsor's hope. The sponsor was unhappy initially, but the project delivered on time. The lesson: estimating discipline buys the right to push back. Bottom-up estimates give you the data to defend a realistic schedule.

These stories share a theme: schedule management is not about producing a Gantt chart, it is about producing trustworthy information. Treat that as your north star and the exam concepts become intuitive.

PMBOK 7 emphasises that schedule management techniques must be tailored to project context. Predictive (waterfall), adaptive (agile), and hybrid lifecycles all need schedules, but they look different.

Predictive lifecycles rely on detailed upfront scheduling using CPM, PERT, and Gantt charts. Changes are formally controlled.

Adaptive lifecycles rely on release plans (multi-iteration roadmaps), iteration plans (sprint backlogs), and information radiators (burndown charts, cumulative flow diagrams). Velocity-based forecasting replaces critical path analysis.

Hybrid lifecycles combine predictive milestones (regulatory dates, integration windows) with adaptive execution windows (development sprints). I use hybrid more than pure predictive or pure adaptive on most enterprise projects.

The PMP exam tests all three. When a question describes a stable, regulated environment, expect predictive answers. When it describes high uncertainty and frequent change, expect adaptive answers. Hybrid answers are common in real-world scenarios.