Skip to content
PMPCAPM

Critical Path Method (CPM)

The Critical Path Method (CPM) is a schedule network analysis technique that calculates early and late start and finish dates for every activity to find the longest path through the project, the minimum project duration, and the float on every other path.

Share:

Know it well enough for the exam? Questions on Critical Path Method (CPM) are scenario-based, not definitional.

Practice it free

Explanation

CPM is used in the Develop Schedule process once activities have been sequenced into a network diagram and their durations estimated. It is not a one-off calculation: the schedule is recalculated whenever progress is reported or a change is approved. The results drive real decisions: the finish date in the schedule baseline, where to apply crashing or fast tracking, which activities get the closest monitoring, and how much a delay or change request will actually move the end date.

CPM rests on three assumptions: durations are fixed, dependencies are known, and resources are unlimited. Each assumption is why CPM is paired with another technique. Three-point estimates or Monte Carlo simulation handle uncertain durations, resource leveling handles limited people and equipment, and the critical chain method adds buffers to protect the resource-constrained path.

In hybrid environments CPM still matters. Agile teams rarely calculate a critical path inside a sprint, but programs that combine agile software teams with hardware, procurement or regulatory work often use a milestone-level CPM schedule to coordinate the dependencies between them.

CPM in plain language

A schedule is a set of activities joined by dependencies. Some chains of activities run in parallel; the project cannot finish until the longest chain finishes. That chain is the critical path, and CPM is the bookkeeping that finds it and tells you, for every other activity, how much it can slip before it starts to matter.

It answers three questions a sponsor actually asks: What is the earliest we can finish? (the length of the critical path), Which work can't slip? (the zero-float activities), and Where do we have room? (everything with positive float). You use it while building the schedule in the Develop Schedule process, and again every time the schedule changes, because a delay on a non-critical path can quietly make it critical.

Worked example: forward pass, backward pass and float

A small software release has six activities. Durations are in working days, and every dependency is finish-to-start with no leads or lags.

ActivityDurationPredecessors
A — Requirements3—
B — Build backend4A
C — Design UI6A
D — Integrate5B
E — Build UI2C
F — Test & release3D, E

Step 1 — Forward pass (earliest dates)

Start the first activity at ES = 0 and work left to right. For each activity, EF = ES + duration. A successor's ES is the largest EF among its predecessors, because it can't start until all of them are done.

  1. 1A: ES 0, EF 0 + 3 = 3.
  2. 2B and C both follow A, so both start at 3. B: EF 3 + 4 = 7. C: EF 3 + 6 = 9.
  3. 3D follows B: ES 7, EF 12. E follows C: ES 9, EF 11.
  4. 4F needs both D (EF 12) and E (EF 11). Take the larger: ES 12, EF 15.

The project's earliest finish is day 15.

Step 2 — Backward pass (latest dates)

Set the last activity's LF to the project finish (15) and work right to left. LS = LF − duration. A predecessor's LF is the smallest LS among its successors, because it must finish before the earliest of them has to start.

  1. 1F: LF 15, LS 15 − 3 = 12.
  2. 2D and E both feed F, so both have LF 12. D: LS 7. E: LS 10.
  3. 3B feeds D: LF 7, LS 3. C feeds E: LF 10, LS 4.
  4. 4A feeds B (LS 3) and C (LS 4). Take the smaller: LF 3, LS 0.

Step 3 — Float

Total float = LS − ES (or LF − EF). Free float = the earliest ES of the successors − this activity's EF.

ActivityESEFLSLFTotal floatFree float
A030300
B373700
C3941010
D71271200
E911101211
F1215121500
A · 3dTF 0ES 0 · EF 3LS 0 · LF 3B · 4dTF 0ES 3 · EF 7LS 3 · LF 7C · 6dTF 1ES 3 · EF 9LS 4 · LF 10D · 5dTF 0ES 7 · EF 12LS 7 · LF 12E · 2dTF 1ES 9 · EF 11LS 10 · LF 12F · 3dTF 0ES 12 · EF 15LS 12 · LF 15
Activity-on-node diagram for the example. Highlighted boxes and arrows are the critical path A → B → D → F (3 + 4 + 5 + 3 = 15 days). TF = total float.

Step 4 — Read the result

There are two paths. A → B → D → F takes 15 days and has zero float: it is the critical path. A → C → E → F takes 14 days, so the activities that are only on that path (C and E) have one day of total float.

Look closely at C. It has one day of total float but zero free float: if C slips by a day, E starts a day later (E's early start moves), but the project still finishes on day 15 because E absorbs it. E, the last activity on that path, has one day of both. This is the distinction exam questions probe.

Now suppose C slips by 2 days. A → C → E → F becomes 16 days, longer than the original critical path, so the project finishes on day 16 and the critical path has moved to A → C → E → F. A slip larger than an activity's total float delays the project by the difference (2 − 1 = 1 day).

Shortcut for exam questions

For a small network, list every path from start to finish and add the durations. The longest total is the critical path; each other path's float is the critical path length minus that path's length. Here: 15 and 14, so the non-critical path has 1 day of float. Use the full forward/backward pass only when the question asks for a specific activity's ES, LF or free float.

Day 0 or day 1?

Some textbooks start the first activity on day 1 and use EF = ES + duration − 1. The dates shift by one, but the critical path and every float value come out the same. Pick one convention and stick with it within a question.

Common exam traps

  • "The critical path is the shortest path."

    It is the longest path through the network, which is why it sets the shortest possible project duration. Both phrases appear in answer options; only one describes the path.

  • "Critical path activities always have zero float."

    Usually, but not always. If an imposed finish date is earlier than the calculated finish, the critical path shows negative float, which tells you the plan does not meet the deadline as it stands. PMI defines the critical path as the longest path, not as the zero-float path.

  • Compressing a non-critical activity to save time.

    Crashing or fast tracking only shortens the project if it is applied to the critical path. After each change, recheck: a near-critical path (here, the 14-day path) can become the new critical path, which caps how much compression helps.

  • Assuming CPM accounts for who is available.

    CPM uses logic and durations only. Apply resource leveling afterwards; that can lengthen the schedule and change which path is critical.

  • A critical activity is late, so you immediately crash it.

    Scenario questions reward analysis first: confirm the impact on the end date and objectives, consider options (compression, re-sequencing, using float elsewhere), then follow change control if the baseline must change. Jumping to an action without assessing is usually the distractor.

  • Mixing up total float and free float.

    Total float protects the project finish date; free float protects the next activity’s early start. Free float can never exceed total float.

On the newer exams, CPM is mostly tested through situations rather than arithmetic: what a delay does, where to compress, why a second critical path raises risk. The step-by-step method is set out in the PMBOK Guide 6th edition's Develop Schedule process; the principle-based 7th edition covers scheduling more briefly within the Planning performance domain.

CPM compared with related techniques

TechniqueWhat it doesHow it differs from CPM
Precedence diagramming (PDM)Draws the activity-on-node network with FS, SS, FF and SF dependencies.PDM builds the network; CPM calculates dates and float on it.
Resource levelingMoves activities so no resource is over-allocated.Runs after CPM and can extend the end date and change the critical path.
Critical chainSchedules with resource constraints and adds project and feeding buffers.Starts from a CPM network but protects the resource-constrained chain with buffers instead of float on each activity.
PERT / three-point estimatingEstimates each duration as (O + 4M + P) / 6.Produces the durations; CPM then uses them to find the path.
Agile release planningSequences backlog items by value within fixed iterations.No network or float; schedule flexibility comes from scope, so CPM is used only for predictive parts of a hybrid project.

Practice and related pages

Free practice questions on this concept (full explanations, no signup):

Study next:

Key Points

  • •Forward pass gives ES and EF; at a merge, take the largest predecessor EF
  • •Backward pass gives LS and LF; at a burst, take the smallest successor LS
  • •Total float = LS - ES (or LF - EF); free float = earliest successor ES - EF
  • •Critical path = the longest path, which sets the shortest possible project duration
  • •Logic and durations only: resource limits are handled afterwards by resource leveling
  • •Compression only helps on the critical path; watch near-critical paths

Exam Tip

For path questions, list every path and add the durations: the longest is critical and every other path's float is the difference. Use a full forward and backward pass only when asked for a specific ES, LF or free float. In situational questions, analyze the impact of a delay before choosing crashing or fast tracking.

Frequently Asked Questions

Related Topics

High-yield topics our learners drill most before exam day.

Resource Leveling

Resource leveling is a resource optimization technique that adjusts activity start and finish dates so that demand for a resource never exceeds its availability, accepting a longer schedule and a changed critical path if necessary.

Resource Smoothing

Resource smoothing is a resource optimization technique that adjusts activities within their available float so that resource requirements do not exceed predefined limits, without changing the project end date.

Burndown Chart

A burndown chart plots the work remaining (usually in story points) against time in a sprint or release, alongside an ideal line to zero, so the team can see each day whether it is on track to finish.

Burn-Up Chart

A burn-up chart plots completed work rising over time against a separate total-scope line, so progress and scope changes are visible independently; the work is finished when the two lines meet.

Deliverables

A deliverable is any unique and verifiable product, result, or capability to perform a service that must be produced to complete a process, phase, or project, such as a software release, an audit report, or a trained support team.

Power Interest Grid (Power/Influence Grid)

The power interest grid is a 2x2 stakeholder mapping tool that plots each stakeholder by power (ability to affect the project) and interest (concern about its outcome) to decide how to engage them: manage closely, keep satisfied, keep informed, or monitor.

Weighted Shortest Job First (WSJF)

Weighted Shortest Job First (WSJF) is a prioritization model, popularized by SAFe, that ranks work by cost of delay divided by job duration (usually estimated as job size), so the items that lose the most value per unit of time spent waiting are done first.

Risk Register

The risk register is a project document that records the details of individual project risks, including their identification, analysis results, response plans, and current status.

Stakeholder Mapping

Stakeholder mapping is the visual representation of stakeholder relationships, influence, interest, or other attributes using grids, matrices, or diagrams to support analysis and engagement planning.

Part of

Schedule Management

Study full domain →

Practice Critical Path Method (CPM)

Scenario-based questions with a full explanation on every answer — including why the distractors are wrong.