PERT and CPM: Techniques for Project Scheduling and Control

Verified Sources

Sep 11, 2026

Program Evaluation and Review Technique (PERT) and Critical Path Method (CPM) are network-based techniques for planning, scheduling, monitoring, and controlling projects.

Both methods represent a project as a set of activities connected by dependencies. Their central objective is to determine when the project can finish and which activities require the closest management attention. PERT emphasizes uncertainty in activity durations, whereas CPM generally uses a single duration estimate and focuses on schedule flexibility, resource decisions, and time-cost trade-offs.

The methods are complementary:

PERT asks: “Given uncertainty, how long is this activity or project likely to take?”
CPM asks: “Which sequence of activities determines the project completion date?”
Combined use asks: “Which uncertain activities on the critical path create the greatest completion risk?”

A project network can be visualized as follows:

In this network, Activity C cannot begin until both A and B are complete. Such relationships are the foundation for both PERT and CPM.

Optimized Project Scheduling: Combining PERT and the Critical Path Method - Overview of PERT three-point estimating and its integration with CPM. ↩

Core distinction

PERT models duration uncertainty with optimistic, most-likely, and pessimistic estimates. CPM normally uses one duration per activity and calculates the critical path, early and late dates, and float.

1. Project Network Concepts

Before applying either technique, convert the project scope into a work breakdown structure and then identify the activities required to produce each deliverable.

Each activity should have:

Element	Meaning
Activity	A task consuming time and possibly resources
Predecessor	An activity that must occur before another activity
Successor	An activity that follows another activity
Duration	The time required to perform the activity
Milestone	A zero-duration event marking a significant point
Path	A connected sequence of activities
Critical path	The longest-duration path through the network
Float or slack	The amount of scheduling flexibility available

Two common diagram conventions are:

Activity-on-node (AON): activities appear in boxes or nodes; arrows show dependencies.
Activity-on-arrow (AOA): activities appear on arrows; nodes represent events. Dummy activities may be required to represent logic correctly.

Modern scheduling practice commonly uses AON because it represents activity relationships directly and supports common dependency types such as finish-to-start, start-to-start, finish-to-finish, and start-to-finish.

A valid network should satisfy these rules:

It has a clearly defined start and finish.
Every activity has a logical predecessor and successor, except beginning and ending activities.
The network contains no circular dependencies.
Every dependency reflects a real project constraint.
Parallel activities are shown in parallel rather than unnecessarily placed in sequence.

Understanding the Basics of CPM Calculations - PMI discussion of activity-on-node networks, CPM terms, forward pass, backward pass, and float. ↩

Build a Project Network

1
Step 1
Break the project into activities that are specific enough to estimate, sequence, assign, and monitor.
2
Step 2
For every activity, determine which activities must be completed before it can start. Avoid creating dependencies merely because tasks are listed in a particular order.
3
Step 3
Use one duration for conventional CPM or three duration estimates for PERT.
4
Step 4
Place activities in dependency order and connect predecessors to successors.
5
Step 5
Check for missing predecessors, circular relationships, unnecessary constraints, and disconnected activities.
6
Step 6
Perform CPM forward and backward passes; when using PERT, first calculate expected durations and uncertainty measures.
7
Step 7
Compare actual progress with the baseline and recalculate the network when durations, logic, or scope change.

2. The Critical Path Method (CPM)

CPM was developed for projects in which activity durations can be estimated with reasonable confidence. It is called “critical path” because the project’s minimum theoretical duration is determined by the longest path through the network—not by the shortest path or by the sum of every activity.

If a project contains several paths, the path with the greatest total duration controls the earliest possible project completion date. Activities on that path generally have zero total float. A delay to one of them can delay the project unless corrective action changes the schedule.

CPM notation

For each activity, calculate:

ES: Early Start
EF: Early Finish
LS: Late Start
LF: Late Finish
TF: Total Float
FF: Free Float

For a duration $d$ :

EF = ES + d

LS = LF - d

For an activity with multiple predecessors:

ES = \max(EF_{\text{predecessors}})

For an activity with multiple successors:

LF = \min(LS_{\text{successors}})

Total float is:

TF = LS - ES = LF - EF

Free float is the amount of time an activity can be delayed without delaying the early start of any immediate successor. For a finish-to-start relationship:

FF = \min(ES_{\text{successors}}) - EF

An activity is usually considered critical when:

TF = 0

However, schedules may contain multiple critical paths, negative float caused by imposed deadlines, or near-critical activities with very little float. Therefore, criticality should be monitored as a risk condition rather than treated as a permanent label.

PERT vs. CPM: Comparing Construction Scheduling Methods - Comparison of probabilistic PERT estimates and CPM critical-path analysis. ↩
Forward Pass and Backward Pass Calculations - Formulas for early and late dates and total float. ↩

Calculate CPM Dates with Forward and Backward Passes

1
Step 1
For activities with no predecessors, set ES to 0 or to the project-calendar start point.
2
Step 2
Calculate EF using EF = ES + duration. For an activity with multiple predecessors, use the largest predecessor EF as its ES.
3
Step 3
The largest EF at the project finish represents the earliest possible completion time under the stated logic and durations.
4
Step 4
For the final activity or activities, set LF equal to the project duration.
5
Step 5
Calculate LS using LS = LF − duration. For an activity with multiple successors, use the smallest successor LS as its LF.
6
Step 6
For each activity, calculate TF = LS − ES or TF = LF − EF.
7
Step 7
Activities with zero total float form one or more critical paths.
8
Step 8
Verify that every activity satisfies EF = ES + duration, LS = LF − duration, and TF = LS − ES. PMI describes forward and backward passes as fundamental CPM calculations for early and late dates and float.

Footnotes

Forward Pass and Backward Pass Calculations - Formulas for early and late dates and total float. ↩

CPM Worked Example

Assume the following activity network:

Activity	Immediate predecessor(s)	Duration
A	None	4
B	None	3
C	A	5
D	B	6
E	C and D	2

The network is:

Forward pass

Activity	ES	Duration	EF
A	0	4	4
B	0	3	3
C	4	5	9
D	3	6	9
E	$\max(9,9)=9$	2	11

The earliest possible completion is 11 time units.

Backward pass

Activity	LS	Duration	LF
E	9	2	11
C	4	5	9
D	3	6	9
A	0	4	4
B	0	3	3

Float and critical paths

Activity	ES	EF	LS	LF
A	0	4	0	4
B	0	3	0	3
C	4	9	4	9
D	3	9	3	9
E	9	11	9	11

There are two critical paths:

A → C → E: $4+5+2=11$
B → D → E: $3+6+2=11$

This example demonstrates that a project can have more than one critical path. Any delay in either branch can delay the project unless the schedule is revised.

CPM Path Durations in the Worked Example

Both paths determine the 11-unit minimum project duration.

Do not confuse the critical path with the busiest path

The critical path is the longest logical path by schedule duration. It is not necessarily the path with the most activities, the highest cost, or the greatest number of assigned people.

3. The Program Evaluation and Review Technique (PERT)

PERT is designed for projects in which activity durations are uncertain, such as research, development, innovation, major engineering, or first-of-a-kind work. Instead of using one duration estimate, PERT uses three:

Optimistic time ( $O$ ): The shortest reasonable duration under favorable conditions.
Most-likely time ( $M$ ): The duration expected under normal conditions.
Pessimistic time ( $P$ ): The longest reasonable duration under unfavorable but plausible conditions.

PERT applies a weighted average that gives four times as much weight to the most-likely estimate as to each extreme:

t_e=\frac{O+4M+P}{6}

The estimated activity variance is:

\sigma^2=\left(\frac{P-O}{6}\right)^2

The standard deviation is:

\sigma=\frac{P-O}{6}

The weighted-average formula and variance estimate are conventional PERT approximations based on a beta-shaped distribution of possible activity durations.

Example

Suppose an activity has:

$O=4$ days
$M=7$ days
$P=16$ days

Then:

t_e=\frac{4+4(7)+16}{6} =\frac{48}{6} =8\text{ days}

The variance is:

\sigma^2=\left(\frac{16-4}{6}\right)^2 =2^2 =4\text{ days}^2

The standard deviation is:

\sigma=2\text{ days}

The expected duration is therefore 8 days, while the uncertainty around the estimate is represented by a standard deviation of 2 days.

PERT Distribution - Technical description of the beta-PERT distribution and its use in project completion-time analysis. ↩

Perform a PERT Analysis

1
Step 1
Create a complete activity list and establish the dependency network.
2
Step 2
For every activity, obtain optimistic, most-likely, and pessimistic durations. Estimates should reflect known assumptions, risks, calendars, and resource conditions.
3
Step 3
Use tₑ = (O + 4M + P) / 6 for each activity.
4
Step 4
Use variance = ((P − O) / 6)² and standard deviation = (P − O) / 6.
5
Step 5
Use each expected duration in a CPM-style forward and backward pass to identify the expected critical path.
6
Step 6
For a selected path, add expected durations and—under the usual independence approximation—add variances.
7
Step 7
Compare a target completion date with the expected project duration using a standardized normal approximation or, for complex schedules, simulation.
8
Step 8
Prioritize activities with high variance that are also on, or close to, the critical path.
9
Step 9
Replace assumptions with actual performance data as the project progresses.

4. PERT Completion-Time Probability

For a particular path, the expected duration is the sum of the expected durations of its activities:

\mu_{\text{path}}=\sum t_{e,i}

If activity durations are treated as approximately independent, the path variance is:

\sigma^2_{\text{path}}=\sum \sigma_i^2

and the path standard deviation is:

\sigma_{\text{path}}=\sqrt{\sum \sigma_i^2}

For a target completion time $T$ , calculate the standardized score:

Z=\frac{T-\mu_{\text{path}}}{\sigma_{\text{path}}}

The approximate probability of completing by $T$ is:

P(\text{completion}\leq T)\approx \Phi(Z)

where $\Phi(Z)$ is the cumulative standard normal distribution.

Example

Assume the expected duration of a critical path is 40 days and its standard deviation is 4 days. For a target of 46 days:

Z=\frac{46-40}{4}=1.5

A standard normal table gives approximately:

\Phi(1.5)\approx 0.933

Thus, under the assumptions of the approximation, the probability of completing by day 46 is about 93.3%.

This result should not be interpreted as a guarantee. PERT probability calculations can be weakened by correlated activities, inaccurate estimates, resource constraints, changing scope, multiple competing paths, and the tendency of the critical path to change as uncertainty is realized. For complex projects, Monte Carlo simulation is often more appropriate than relying on a single normal approximation.

Stochastic Project Duration Analysis Using PERT - Academic treatment of PERT means, variances, beta distributions, and stochastic project-duration analysis. ↩

5. Comparing PERT and CPM

Dimension	PERT	CPM
Primary purpose	Model uncertainty and estimate probable completion	Determine schedule duration and controlling activities
Activity durations	Three estimates: $O$ , $M$ , and $P$	Usually one duration estimate
Nature	Probabilistic	Deterministic, or scenario-based
Main output	Expected duration, variance, and completion probability	Early dates, late dates, float, and critical path
Best suited to	Novel, uncertain, research, and development projects	Repetitive, well-understood, construction, and production projects
Main question	“How likely is completion by a target date?”	“Which activities control completion?”
Risk treatment	Explicitly models duration uncertainty	Usually requires separate risk analysis
Cost-time analysis	Less central in the basic method	Commonly supports crashing and time-cost trade-offs
Main limitation	Sensitive to estimate quality and statistical assumptions	Can create false precision when durations are highly uncertain

PERT and CPM are not mutually exclusive. A project team can use PERT estimates to produce expected durations and then apply CPM calculations to determine the expected critical path. This hybrid approach is common when the project has both well-understood and uncertain activities.

Optimized Project Scheduling: Combining PERT and the Critical Path Method - Overview of PERT three-point estimating and its integration with CPM. ↩

Evolution and Practical Development of PERT and CPM

Network scheduling emerges

1950s

Large engineering and defense programs increasingly require formal methods for representing dependencies and coordinating complex work."

CPM is developed

1957

DuPont and Remington Rand develop the Critical Path Method for planning maintenance and construction-related work with relatively predictable activity durations."

Critical Path Method - Background and practical explanation of CPM scheduling and critical-path calculations. ↩

PERT is introduced

Late 1950s

PERT is developed for the U.S. Navy’s Polaris program to manage uncertainty in a large research and development effort."

PERT - Explanation of three-point estimation and PERT’s project-management context. ↩

Methods become general project tools

1960s onward

PERT and CPM are adopted across construction, manufacturing, information systems, engineering, and public-sector programs."

Integrated scheduling and risk analysis

Contemporary practice

Scheduling software combines network logic, calendars, resource constraints, baseline control, probabilistic analysis, and scenario modeling."

6. Time-Cost Trade-Offs and Crashing

One important CPM application is schedule crashing. Crashing is used when a project must finish earlier than the normal schedule, but it should be applied selectively because not every activity can be shortened and shortening a noncritical activity may not reduce project duration.

For an activity with normal and crash estimates:

Normal time: $T_n$
Crash time: $T_c$
Normal cost: $C_n$
Crash cost: $C_c$

The cost slope is:

\text{Cost slope} = \frac{C_c-C_n}{T_n-T_c}

A lower cost slope generally indicates a less expensive candidate for shortening. However, the activity must be on the current critical path, and crashing it may create another critical path.

Crashing procedure

Calculate the current critical path.
Identify crashable activities on that path.
Calculate each activity’s cost slope.
Select the least-cost activity that can reduce the project duration.
Crash it by an allowable increment.
Recalculate the network.
Check whether another critical path has emerged.
Continue until the target date is reached or no further economical reduction is possible.

Example

Activity	Normal time	Crash time	Normal cost	Crash cost
A	8	6	4,000	5,000
B	10	7	6,000	7,500

For Activity A:

\text{Cost slope}_A=\frac{5{,}000-4{,}000}{8-6}=500

For Activity B:

\text{Cost slope}_B=\frac{7{,}500-6{,}000}{10-7}=500

Both activities have the same incremental cost of 500 currency units per time unit. The choice should therefore consider resource availability, risk, quality, and whether crashing one activity would create a second critical path.

Use PERT and CPM together

Use PERT to improve duration estimates when uncertainty is high, then use CPM to identify the path that controls the expected completion date. This combines risk awareness with schedule control.

7. Practical Interpretation of Float

Float is not automatically “free time.” It represents schedule flexibility under a particular network model, calendar, and set of assumptions.

Total float

Total float is the amount an activity can slip without delaying the project completion date, assuming successor relationships and other activities remain unchanged:

TF=LS-ES=LF-EF

Free float

Free float is the amount an activity can slip without delaying the early start of any immediate successor:

FF=\min(ES_{\text{successors}})-EF

Negative float

Negative float occurs when the calculated schedule cannot meet an imposed deadline. It signals that the schedule requires corrective action, such as:

Reducing scope
Adding resources
Changing dependencies
Fast-tracking activities
Crashing activities
Revising the required date
Accepting additional risk

Near-critical activities

An activity with small positive float may become critical after a minor delay. Effective schedule control therefore monitors:

Zero-float activities
Low-float activities
High-variance activities
Activities with difficult-to-replace resources
Activities that feed several downstream branches

Frequently Asked Questions

PERT and CPM Key Terms

1 / 9

Question · Term

Critical path

Click to reveal

Answer · Definition

The longest-duration path through the project network; it determines the earliest possible project completion under the modeled assumptions.

8. Integrated PERT–CPM Workflow

A disciplined project team can apply both methods through the following integrated workflow:

Recommended decision rules

Use CPM alone when durations are stable and the primary need is schedule logic and float analysis.
Use PERT when durations are uncertain and management needs an expected duration and uncertainty estimate.
Use PERT plus CPM when uncertainty must be linked to the activities controlling completion.
Use Monte Carlo simulation when there are many uncertain activities, correlated risks, resource constraints, or several plausible critical paths.
Use crashing only after confirming that the selected activity is critical and that the incremental cost is justified.
Recalculate after major changes; a baseline critical path is not a permanent project fact.

Important limitation

A mathematically calculated critical path is only as reliable as the activity list, dependency logic, duration estimates, calendars, and assumptions behind it. Incorrect network logic can produce a precise but misleading schedule.

9. Learning Checklist

You should now be able to:

Define PERT and CPM.
Explain the difference between probabilistic and deterministic scheduling.
Identify activities, dependencies, paths, and milestones.
Build an activity-on-node network.
Calculate PERT expected time, variance, and standard deviation.
Perform CPM forward and backward passes.
Calculate ES, EF, LS, LF, total float, and free float.
Identify one or more critical paths.
Estimate completion probability using a $Z$ -score.
Explain why PERT probabilities depend on assumptions.
Distinguish crashing from fast-tracking.
Use float and variance to prioritize schedule risk.
Explain how PERT and CPM can be integrated in one project-control system.

Knowledge Check

Question 1 of 5

Q1Single choice

What is the primary difference between PERT and CPM?

PERT uses three time estimates to model uncertainty, while CPM usually uses one duration estimate

PERT is used only for cost estimation, while CPM is used only for quality control

PERT eliminates dependencies, while CPM creates dependencies

PERT applies only to completed projects, while CPM applies only to proposed projects

References

Explore Related Topics

Performance Enhancement of Solar Panels Using Thermal Management Techniques: Solar PV Cooling with Phase Change Materials (PCM)

Balancing Productivity and Well-Being Under Tight Deadlines

Risk Management Activities and Risk Prioritization: A Comprehensive Guide

Browse all research articles

PERT and CPM: Techniques for Project Scheduling and Control

Footnotes

Core distinction

1. Project Network Concepts

Footnotes

Build a Project Network

2. The Critical Path Method (CPM)

CPM notation

Footnotes

Calculate CPM Dates with Forward and Backward Passes

Footnotes

CPM Worked Example

Forward pass

Backward pass

Float and critical paths

CPM Path Durations in the Worked Example

Do not confuse the critical path with the busiest path

3. The Program Evaluation and Review Technique (PERT)

Example

Footnotes

Perform a PERT Analysis

4. PERT Completion-Time Probability

Example

Footnotes

5. Comparing PERT and CPM

Footnotes

Evolution and Practical Development of PERT and CPM

Network scheduling emerges

CPM is developed

Footnotes

PERT is introduced

Footnotes

Methods become general project tools

Integrated scheduling and risk analysis

6. Time-Cost Trade-Offs and Crashing

Crashing procedure

Example

Use PERT and CPM together

7. Practical Interpretation of Float

Total float

Free float

Negative float

Near-critical activities

Frequently Asked Questions

PERT and CPM Key Terms

Critical path

8. Integrated PERT–CPM Workflow

Recommended decision rules

Important limitation

9. Learning Checklist

Knowledge Check

References

Explore Related Topics