Decision Making Decision method

PERT Estimation

Combines three estimates into an expected value and makes variation and risk visible.

Optimistische, wahrscheinlichste und pessimistische Einschätzungen werden mathematisch gewichtet, statt einfach gemittelt zu werden. So entsteht eine Planungszahl, die Unsicherheit sichtbar mitführt und weniger von Extremwerten verzerrt wird. Nützlich ist vor allem die konsistente Verbindung von Schätzung und Risikobild.

Core question

What PERT expected values and variances follow from O/M/P, and how are independent work packages aggregated correctly?

Plan PERT Estimation Method guide

Start comparison

Give feedbackIs this method useful to you?

Participants: 1-8
Time: 45–90 minutes for a small set of independent work packages; network scheduling separate.

The steps form one complete method run.

Scope
PERT frame
Capture work packages, unit, Done, dependencies, and common risks.
Values
O/M/P table
Set O, M, and P per package as plausible scenarios with O ≤ M ≤ P.
Calculate
PERT measures
For each package calculate E=(O+4M+P)/6, SD=(P−O)/6, and variance=SD².
Aggregate
Aggregate
Only under independence add expected values and variances; total SD is square root of summed variance.
Interpret
Planning finding
Use measures with assumptions; do not claim confidence without a validated distribution.

Add new evidence at review and version estimates with the prior state.

A good fit when

Bei komplexen Vorhaben verfälscht ein reiner Mittelwert die Unsicherheit oft zu stark. PERT Estimation verbindet drei Lagebilder zu einem erwarteten Wert und hält das Risiko der Streuung mit im Blick.

Ideal for

Schedule and cost estimation
Project planning
Risk analysis

Not good for

Very small agile stories
Pure prioritization
Work without an expert baseline

Outputs

Planning finding
PERT table with E=(O+4M+P)/6, SD=(P−O)/6, variance=SD², aggregated independent variances, and stated model limits.

PERT Estimation · working structure

Fictional example

Fictional values show the actual method fields.

PERT Estimation · working structure
Package	O / M / P	E = (O+4M+P)/6	SD = (P−O)/6	Variance	Dependency	Model limit
A · TokenisationAdd evidence and rationale.	4 / 7 / 16 days	(4+28+16)/6 = 8.00	(16−4)/6 = 2.00	4.00	Independent of B for example	Check shared certification
B · RolloutAdd evidence and rationale.	2 / 3 / 8 days	(2+12+8)/6 = 3.67	(8−2)/6 = 1.00	1.00	Independent of A for example	Rollout window fixed
AggregateAdd evidence and rationale.	Do not average again	8.00+3.67 = 11.67	√(4+1) = 2.24	5.00	Only under independence	No guaranteed interval

1. O / M / P
A · Tokenisation
4 / 7 / 16 days
B · Rollout
2 / 3 / 8 days
Aggregate
Do not average again
2. E = (O+4M+P)/6
A · Tokenisation
(4+28+16)/6 = 8.00
B · Rollout
(2+12+8)/6 = 3.67
Aggregate
8.00+3.67 = 11.67
3. SD = (P−O)/6
A · Tokenisation
(16−4)/6 = 2.00
B · Rollout
(8−2)/6 = 1.00
Aggregate
√(4+1) = 2.24
4. Variance
A · Tokenisation
4.00
B · Rollout
1.00
Aggregate
5.00
5. Dependency
A · Tokenisation
Independent of B for example
B · Rollout
Independent of A for example
Aggregate
Only under independence
6. Model limit
A · Tokenisation
Check shared certification
B · Rollout
Rollout window fixed
Aggregate
No guaranteed interval

All values are fictional.

Reference · original teaching example

First step

Close work packages, unit, Done, and dependencies; explicitly check independence before aggregating variance.

Open the method guide

Good to do first

Decision Making

Three-Point Estimation

Supplies the O/M/P scenarios per work package as the calculation basis for PERT.

Where to go next?

Delivery

Monte Carlo Forecasting

Uses variance and dependencies for a simulation-based delivery distribution.

Decision Making

Risk Matrix

Formally assesses the open independence assumptions and uncertainties by likelihood and impact.

Deep dive

Purpose

On complex initiatives, a plain average tends to understate just how uncertain the outcome really is. It separates options, evaluation criteria, and open risks. The result is captured as a PERT Estimate, Expected Value, and Risk Notes.

How it works

Optimistic, most likely, and pessimistic estimates are weighted mathematically instead of simply averaged. This produces a planning figure that visibly carries uncertainty along with it and is distorted less by extreme values. What is especially useful is the consistent link between estimate and risk picture.

In detail

PERT Estimation uses the same three values as Three-Point Estimation but gives the most likely value more weight. That produces an expected value that is less distorted by extremes than a simple average. The method can also consider variance or standard deviation to communicate schedule or cost risk. It remains an expert estimate and should not be mistaken for a precise prediction.

Facilitation

Explain the formula and the meaning of the three values up front. Calculate transparently and record assumptions, especially when pessimistic values are driven by dependencies or external risks. Use the numbers as decision input, not as a delivery promise.

The method guide walks execution through 5 phases, timing, 5 pitfalls, and clear stop criteria.

Open method guide

Your perspective

Does this method help with your work?

Your feedback helps shape the next improvements to methodatlas.

Sources and topics

Topics

Estimation Uncertainty Risk

Sources

PERT Estimationnasa.govNASA Cost Estimating Handbook (OCFO); institutional reference on cost estimating methods with an appendix on risk and uncertainty.

Artifact templates

PERT Estimation: WorksheetSet subject, data, roles, flow, and review.

markdown

pert-estimation-working-template.md

Set subject, data, roles, flow, and review.

PERT Estimation: Worksheet

Question

To be confirmed

Desired outcome

To be confirmed

Scope

Complete run Time: 45–90 minutes for a small set of independent work packages; network scheduling separate.

Complete this worksheet on paper or in your own document during the work.

Preparation for this scope

Method setup

Materials: Defined work packages, O/M/P scenarios, historical data, PERT worksheet, dependencies, and assumption log
Roles: Estimator · work owner · risk/data expert · schedule owner
Advance information: Close work packages, unit, Done, and dependencies; explicitly check independence before aggregating variance.
Overall time needed: 45–90 minutes for a small set of independent work packages; network scheduling separate.
Setup: Prepare a workspace with Scope · Values · Calculate · Aggregate · Interpret; separate example from live data.

Prepare for the work steps

Scope: PERT frame

Close work packages, unit, Done, and dependencies; explicitly check independence before aggregating variance.
Defined work packages, O/M/P scenarios, historical data, PERT worksheet, dependencies, and assumption log

Values: O/M/P table

PERT frame

Calculate: PERT measures

O/M/P table

Aggregate: Aggregate

PERT measures

Interpret: Planning finding

Aggregate

PERT Estimation · working structure

Package	O / M / P	E = (O+4M+P)/6	SD = (P−O)/6	Variance	Dependency	Model limit
A · Tokenisation
B · Rollout
Aggregate

A · Tokenisation: Add evidence and rationale.
B · Rollout: Add evidence and rationale.
Aggregate: Add evidence and rationale.

The blank worksheet contains prompts only.

Reference · original teaching example

Scope: PERT frame

Expected artifact: PERT frame

Capture work packages, unit, Done, dependencies, and common risks.

Entry:

...

Are packages separable and common causes visible?

Values: O/M/P table

Expected artifact: O/M/P table

Set O, M, and P per package as plausible scenarios with O ≤ M ≤ P.

Entry:

...

Do values describe one scope and explainable conditions?

Calculate: PERT measures

Expected artifact: PERT measures

For each package calculate E=(O+4M+P)/6, SD=(P−O)/6, and variance=SD².

Entry:

...

Are weighting, unit, and calculation visible?

Aggregate: Aggregate

Expected artifact: Aggregate

Only under independence add expected values and variances; total SD is square root of summed variance.

Entry:

...

Were variances rather than standard deviations added?

Interpret: Planning finding

Expected artifact: Planning finding

Use measures with assumptions; do not claim confidence without a validated distribution.

Entry:

...

Are model limits, risks, and review triggers recorded?

Open questions and next steps

...

Method guide: https://methodatlas.meierhoff-systems.de/en/methods/pert-estimation/run-sheet

Alternatives and neighbors

Fitting alternatives with decision guidance, followed by other related methods.

All methods

Agile

Ideal Days

Instead of PERT Estimation, choose Ideal Days when you want to express effort in pure working days and keep team calibration manageable.

Agile

Story Points

Instead of PERT Estimation, choose Story Points when you want to estimate relative complexity instead of calendar time and calibrate your estimation as a team.

Decision Making

Wideband Delphi

Moves options, criteria, and risks toward a concrete result through "clarify the estimate object and unit", "collect independent estimates", and "document another round and the outcome".