Method guideDelivery

Monte Carlo Forecasting

A practical method guide: clarify prerequisites and materials, understand the complete process, and look up outputs, pitfalls, and stop criteria.

ComplexityHigh

Time30-90 min Setup, danach laufend

Participants1-8

FormatWorkshop + async

MaturityEstablished

Plan Monte Carlo Forecasting

Prerequisite

What needs to be finished first

Complete firstComparable subject

Choose one question: when will N items finish or how many by date D; check service boundary, Done, data window, and comparability.

Without a common frame, values are not comparable.

Complete firstEvidence access

Remaining count or target date, consistent historical throughput sample, system/WIP context, many-run simulation, and quantile report

Missing data remains visible as uncertainty.

Preparation

What needs to be ready before start

Materials

Remaining count or target date, consistent historical throughput sample, system/WIP context, many-run simulation, and quantile report

People / roles

Service owner · flow/data expert · delivery team · decision makers

Pre-read

Choose one question: when will N items finish or how many by date D; check service boundary, Done, data window, and comparability.

Time needed

60–120 minutes for data check, simulation, and decision review; rerun regularly.

Setup

Prepare a workspace with Question · Sample · Simulate · Quantiles · Check · Decide; separate example from live data.

Core question

The one question this method answers

What probability distribution for completion duration or delivered count emerges when actual historical throughput values are repeatedly resampled?

Flow

Phases with goal, actions, output and pitfall

Die Simulation zieht aus beobachtetem Durchsatz oder Cycle Time viele mögliche Zukünfte und verdichtet sie zu Perzentilen. Dadurch wird Vorhersage als Wahrscheinlichkeitsbild lesbar, das sich mit neuen Daten nachschärfen lässt und Diskussionen über Zusagen deutlich erdet.

Monte Carlo Forecasting · working structure

Fictional example

Fictional values show the actual method fields.

Monte Carlo Forecasting · working structure
Forecast part	Input/logic	Fictional value	Check	Outcome	Decision	Reforecast
QuestionAdd evidence and rationale.	Date for quantity	When will 30 items finish?	One count question; start today	Open	P85 internal, P95 external	Remaining count changes
SampleAdd evidence and rationale.	Historical weekly throughput	5,7,4,6,8,5,0,7,6,4,9,5 per week	Same team/Done/WIP; keep zero	12 empirical values	No mean division	Policy change
SimulationAdd evidence and rationale.	Empirical resampling	10,000 paths; sample with replacement to ≥30	Each draw=historical week	Duration per path	Sort quantiles	Check data window
QuantilesAdd evidence and rationale.	Sorted durations	P50 6; P85 7; P95 8 weeks	No average date	Distribution, not point	External 8 weeks	Weekly

1. Input/logic
Question
Date for quantity
Sample
Historical weekly throughput
Simulation
Empirical resampling
Quantiles
Sorted durations
2. Fictional value
Question
When will 30 items finish?
Sample
5,7,4,6,8,5,0,7,6,4,9,5 per week
Simulation
10,000 paths; sample with replacement to ≥30
Quantiles
P50 6; P85 7; P95 8 weeks
3. Check
Question
One count question; start today
Sample
Same team/Done/WIP; keep zero
Simulation
Each draw=historical week
Quantiles
No average date
4. Outcome
Question
Open
Sample
12 empirical values
Simulation
Duration per path
Quantiles
Distribution, not point
5. Decision
Question
P85 internal, P95 external
Sample
No mean division
Simulation
Sort quantiles
Quantiles
External 8 weeks
6. Reforecast
Question
Remaining count changes
Sample
Policy change
Simulation
Check data window
Quantiles
Weekly

All values are fictional.

Reference · original teaching example

1
Question
Forecast question
Goal
Forecast question: Set forecast type, remaining count/time window, start, and required risk level.
Inputs for this step
- Choose one question: when will N items finish or how many by date D; check service boundary, Done, data window, and comparability.
- Remaining count or target date, consistent historical throughput sample, system/WIP context, many-run simulation, and quantile report
Actions
- Set forecast type, remaining count/time window, start, and required risk level.
Output
Forecast question
Set forecast type, remaining count/time window, start, and required risk level.
Quality check
Is exactly one quantity or date question expressed in count units?
Pitfall
Scope and target date remain unclear during simulation. Fix count, start, Done, and decision quantile first.
Timing: 10–15 minutes
2
Sample
Throughput data
Goal
Throughput data: Check historical throughput from the same system and cadence; retain zeros and normal variation.
Inputs for this step
- Forecast question
Actions
- Check historical throughput from the same system and cadence; retain zeros and normal variation.
Output
Throughput data
Check historical throughput from the same system and cadence; retain zeros and normal variation.
Quality check
Are period, team, WIP policy, and Done comparable enough?
Pitfall
Outliers and zero weeks are removed to improve the forecast. Correct only data errors; retain real variation and mark system changes.
Timing: 15–25 minutes
3
Simulate
Simulation runs
Goal
Simulation runs: Each run samples weekly values with replacement, accumulates until remaining count is reached, and stores duration.
Inputs for this step
- Throughput data
Actions
- Each run samples weekly values with replacement, accumulates until remaining count is reached, and stores duration.
Output
Simulation runs
Each run samples weekly values with replacement, accumulates until remaining count is reached, and stores duration.
Quality check
Does each draw correspond to one observed throughput period?
Pitfall
Thirty divided by average throughput is called Monte Carlo. Resample empirical values and generate many complete future paths.
Timing: 10–20 minutes
4
Quantiles
Probabilities
Goal
Probabilities: Sort durations and report several quantiles as risk levels.
Inputs for this step
- Simulation runs
Actions
- Sort durations and report several quantiles as risk levels.
Output
Probabilities
Sort durations and report several quantiles as risk levels.
Quality check
Are P50/P85/P95 explained as by-with-likelihood rather than guarantees?
Pitfall
Only a mean date appears as a precise commitment. Show quantiles with likelihood, run count, and start context.
Timing: 10–15 minutes
5
Check
Robustness view
Goal
Robustness view: Test sensitivity to data window, scope, and WIP change; backtest prior forecasts where possible.
Inputs for this step
- Probabilities
Actions
- Test sensitivity to data window, scope, and WIP change; backtest prior forecasts where possible.
Output
Robustness view
Test sensitivity to data window, scope, and WIP change; backtest prior forecasts where possible.
Quality check
Does the decision survive plausible data windows?
Pitfall
An arbitrary window is treated as objective. Compare relevant windows and explain the difference.
Timing: 10–20 minutes
6
Decide
Forecast decision
Goal
Forecast decision: Choose accepted risk, record assumptions, and reforecast after scope/system change.
Inputs for this step
- Robustness view
Actions
- Choose accepted risk, record assumptions, and reforecast after scope/system change.
Output
Forecast decision
Choose accepted risk, record assumptions, and reforecast after scope/system change.
Quality check
Is a concrete action tied to the quantile?
Pitfall
Forecast stays fixed while count or system changes. Update remaining count, throughput, and policy regularly.
Timing: 5–10 minutes

Artifact

What comes out at the end

Sketch: what the artifact looks like

Monte Carlo Forecasting · probabilistische LieferungHistorische Flow-Daten simulieren und Liefertermine oder Scope als Wahrscheinlichkeitsverteilung mit Perzentilen kommunizieren

Form

Reproducible Monte Carlo forecast with input sample, sampling logic, run count, P50/P85/P95, assumptions, sensitivity, and reforecast trigger.

Versioning / ownership

Record date, team, data state, and review date; preserve prior versions.

Tool alternatives

Throughput history
Monte Carlo simulation
Quantile and assumption report
Backtest/reforecast log

monte-carlo-forecasting-working-template.md

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

Monte Carlo Forecasting: Worksheet

Question

To be confirmed

Desired outcome

To be confirmed

Scope

Practice run Time: 60–120 minutes for data check, simulation, and decision review; rerun regularly.

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

Preparation for this scope

Method setup

Materials: Remaining count or target date, consistent historical throughput sample, system/WIP context, many-run simulation, and quantile report
Roles: Service owner · flow/data expert · delivery team · decision makers
Advance information: Choose one question: when will N items finish or how many by date D; check service boundary, Done, data window, and comparability.
Overall time needed: 60–120 minutes for data check, simulation, and decision review; rerun regularly.
Setup: Prepare a workspace with Question · Sample · Simulate · Quantiles · Check · Decide; separate example from live data.

Prepare for the work steps

Question: Forecast question

Choose one question: when will N items finish or how many by date D; check service boundary, Done, data window, and comparability.
Remaining count or target date, consistent historical throughput sample, system/WIP context, many-run simulation, and quantile report

Sample: Throughput data

Forecast question

Simulate: Simulation runs

Throughput data

Quantiles: Probabilities

Simulation runs

Check: Robustness view

Probabilities

Decide: Forecast decision

Robustness view

Monte Carlo Forecasting · working structure

Forecast part	Input/logic	Fictional value	Check	Outcome	Decision	Reforecast
Question
Sample
Simulation
Quantiles

Question: Add evidence and rationale.
Sample: Add evidence and rationale.
Simulation: Add evidence and rationale.
Quantiles: Add evidence and rationale.

The blank worksheet contains prompts only.

Reference · original teaching example

Question: Forecast question

Expected artifact: Forecast question

Set forecast type, remaining count/time window, start, and required risk level.

Entry:

...

Is exactly one quantity or date question expressed in count units?

Sample: Throughput data

Expected artifact: Throughput data

Check historical throughput from the same system and cadence; retain zeros and normal variation.

Entry:

...

Are period, team, WIP policy, and Done comparable enough?

Simulate: Simulation runs

Expected artifact: Simulation runs

Each run samples weekly values with replacement, accumulates until remaining count is reached, and stores duration.

Entry:

...

Does each draw correspond to one observed throughput period?

Quantiles: Probabilities

Expected artifact: Probabilities

Sort durations and report several quantiles as risk levels.

Entry:

...

Are P50/P85/P95 explained as by-with-likelihood rather than guarantees?

Check: Robustness view

Expected artifact: Robustness view

Test sensitivity to data window, scope, and WIP change; backtest prior forecasts where possible.

Entry:

...

Does the decision survive plausible data windows?

Decide: Forecast decision

Expected artifact: Forecast decision

Choose accepted risk, record assumptions, and reforecast after scope/system change.

Entry:

...

Is a concrete action tied to the quantile?

Open questions and next steps

...

Method guide: https://methodatlas.meierhoff-systems.de/en/methods/monte-carlo-forecasting/run-sheet

Example output

Concrete filled scenario, fictional example

Fictional example for orientation. Counts and timings describe this case and are not universal requirements.

monte-carlo-forecasting-beispiel.md

Concrete filled scenario, fictional example

Fictional teaching example: Monte Carlo Forecasting · working structure

Forecast part	Input/logic	Fictional value	Check	Outcome	Decision	Reforecast
Question	Date for quantity	When will 30 items finish?	One count question; start today	Open	P85 internal, P95 external	Remaining count changes
Sample	Historical weekly throughput	5,7,4,6,8,5,0,7,6,4,9,5 per week	Same team/Done/WIP; keep zero	12 empirical values	No mean division	Policy change
Simulation	Empirical resampling	10,000 paths; sample with replacement to ≥30	Each draw=historical week	Duration per path	Sort quantiles	Check data window
Quantiles	Sorted durations	P50 6; P85 7; P95 8 weeks	No average date	Distribution, not point	External 8 weeks	Weekly

Pitfalls

Recognize symptoms and steer against them

Quick overview, the full text is in the matching phase in the flow

Forecast questionScope and target date remain unclear during simulation.View in phase
Throughput dataOutliers and zero weeks are removed to improve the forecast.View in phase
Simulation runsThirty divided by average throughput is called Monte Carlo.View in phase
ProbabilitiesOnly a mean date appears as a precise commitment.View in phase
Robustness viewAn arbitrary window is treated as objective.View in phase
Forecast decisionForecast stays fixed while count or system changes.View in phase

Stop criteria

When to pause, resolve prerequisites, or choose another approach

Stop Question: Is exactly one quantity or date question expressed in count units?

Stop Sample: Are period, team, WIP policy, and Done comparable enough?

Stop Simulate: Does each draw correspond to one observed throughput period?

Stop Quantiles: Are P50/P85/P95 explained as by-with-likelihood rather than guarantees?

Stop Check: Does the decision survive plausible data windows?

Stop Decide: Is a concrete action tied to the quantile?

Where to go next?

Delivery

RAID Log

Enters the accepted risk and forecast assumptions into a running log.

Delivery

Communication Plan

Communicates the chosen delivery date as a percentile to stakeholders.

View in playbookImprove Estimation and ForecastingMove from intuition to shared sizing, uncertainty, and delivery forecast.

Finished the method guide?

Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.

Go to profile More methods

Monte Carlo Forecasting · working structure

1. Input/logic

2. Fictional value

3. Check

4. Outcome

5. Decision

6. Reforecast

Forecast question

Throughput data

Simulation runs

Probabilities

Robustness view

Forecast decision

monte-carlo-forecasting-working-template.md

Monte Carlo Forecasting: Worksheet

Question

Desired outcome

Scope

Preparation for this scope

Method setup

Prepare for the work steps

Question: Forecast question

Sample: Throughput data

Simulate: Simulation runs

Quantiles: Probabilities

Check: Robustness view

Decide: Forecast decision

Monte Carlo Forecasting · working structure

Question: Forecast question

Sample: Throughput data

Simulate: Simulation runs

Quantiles: Probabilities

Check: Robustness view

Decide: Forecast decision

Open questions and next steps

monte-carlo-forecasting-beispiel.md

Fictional teaching example: Monte Carlo Forecasting · working structure

Forecast question

Throughput data

Simulation runs

Probabilities

Robustness view

Forecast decision

Where to go next?

Finished the method guide?