MAPE Calculator (Mean Absolute Percentage Error)

About the MAPE Calculator (Mean Absolute Percentage Error)

Mean Absolute Percentage Error (MAPE) is one of the most commonly used metrics for evaluating forecast accuracy. It expresses the average absolute error as a percentage of actual demand, making it intuitive and easy to communicate across the organization.

MAPE is calculated by taking the absolute difference between actual and forecast for each period, dividing by the actual value, averaging across all periods, and multiplying by 100. A MAPE of 10% means the forecast is, on average, within 10% of actual demand.

This calculator accepts pairs of actual and forecast values and computes the overall MAPE, along with per-period percentage errors for diagnosis.

Use the result to compare operating scenarios, pressure-test assumptions, and rerun the model when volumes, rates, or service targets change.

Use the output to compare options, spot the main cost drivers, and rerun the math when lane assumptions or operating constraints change.

Use the output to compare options, spot the main cost drivers, and rerun the math when lane assumptions or operating constraints change.

When This Page Helps

MAPE provides a scale-independent measure of forecast accuracy that is easy for non-technical stakeholders to understand. A single percentage number communicates forecast quality quickly. This calculator eliminates manual error computation, helping demand planners track and report forecast performance efficiently.

How to Use the Inputs

1. Enter actual demand values separated by commas.
2. Enter corresponding forecast values separated by commas.
3. Ensure both lists have the same number of values.
4. Review the MAPE percentage result.
5. Check individual period errors for outlier identification.
6. Use the result to benchmark and improve forecasting methods.

Example Calculation

Tips & Best Practices

• MAPE below 10% is generally considered excellent; 10–20% is good; above 30% indicates poor accuracy.
• MAPE penalizes under-forecasting less than over-forecasting — consider symmetric MAPE (sMAPE) for balance.
• Exclude periods with zero actual demand, as they cause undefined results.
• Use MAPE alongside MAD and Bias for a complete picture of forecast quality.
• Track MAPE by product family, region, or forecast horizon to identify systemic issues.
• MAPE tends to favor intermittent demand items because low actuals inflate percentage errors.

MAPE Interpretation Guidelines

Less than 10% MAPE indicates a high-quality forecast suitable for fine-tuned inventory management. Between 10–25% is acceptable for most supply chain planning. Above 25% suggests the forecasting method or data needs significant improvement. Always compare MAPE against a naïve forecast (e.g., last period's demand) to confirm your model adds value.

Alternatives to MAPE

For intermittent demand, use MASE (Mean Absolute Scaled Error). For symmetric evaluation, use sMAPE. For operations that care about absolute quantities rather than percentages, use MAD. Each metric highlights different aspects of forecast quality.

Using MAPE for Continuous Improvement

Track MAPE over time to measure the impact of forecasting improvements. Set MAPE targets by product group and review monthly. Recognize that no forecast is perfect — even world-class organizations rarely achieve MAPE below 5% across their full product portfolio.

Frequently Asked Questions

• What does MAPE measure?

  MAPE measures the average absolute forecast error expressed as a percentage of actual demand. It answers the question: on average, how far off is our forecast in percentage terms?

• What is a good MAPE value?

  It depends on the industry and demand variability. For fast-moving consumer goods, <10% is excellent. For high-variability or intermittent demand, 20–30% may be acceptable. Always benchmark against your historical MAPE.

• Why is MAPE problematic for low-volume items?

  When actual demand is small, even a one-unit error produces a high percentage. For example, actual = 2 and forecast = 3 yields 50% error. For intermittent demand, use MAD or MASE instead.

• What happens when actual demand is zero?

  MAPE is undefined when actual = 0 because of division by zero. This calculator excludes those periods. Consider using MAD or weighted MAPE for datasets with zero-demand periods.

• Is MAPE symmetric?

  No. MAPE penalizes over-forecasting more heavily than under-forecasting due to the asymmetry of percentage errors. Symmetric MAPE (sMAPE) uses the average of actual and forecast in the denominator for balance.

• How does MAPE relate to forecast bias?

  MAPE measures error magnitude but not direction. A process could have low MAPE but systematic bias (consistently over or under). Use the Bias/Tracking Signal calculator alongside MAPE for a complete view.