Notice: Amazon Forecast has been discontinued by AWS. No new customers are accepted as of July 2024. Existing workloads should be migrated to alternatives such as Amazon SageMaker Canvas or custom ML models on SageMaker.
What is Amazon Forecast?
Amazon Forecast is a fully managed machine learning service for time series forecasting. The service uses the same technology as Amazon.com for product demand predictions. You upload historical data and receive accurate forecasts without ML expertise.
Core Features
- AutoML: Automatic algorithm selection and hyperparameter optimization
- Multiple Algorithms: DeepAR+, Prophet, ARIMA, ETS, and more
- Related Data: Integration of external factors like prices, weather, or holidays
- Probabilistic Forecasts: Confidence intervals for uncertainty quantification
- What-If Analyses: Scenario simulation for planning purposes
Typical Use Cases
Demand Forecasting: Forecast product demand for optimal inventory levels. Reduce overstock and stock-outs simultaneously.
Resource Planning: Plan personnel capacity, server capacity, or material requirements based on forecasted needs.
Financial Planning: Create revenue and cost forecasts for budgeting and financial planning with quantified uncertainties.
Benefits
- No ML expertise required
- Better accuracy than traditional methods through deep learning
- Fast implementation in days instead of months
- Scales automatically with data volume
Integration with innFactory
As an AWS Reseller, innFactory supports you with Amazon Forecast: data preparation, model training, integration into existing planning systems, and interpretation of forecasts for business decisions.
Typical Use Cases
Frequently Asked Questions
Do I need ML expertise for Forecast?
No, Forecast is an AutoML service. You upload historical data and the service automatically selects the best algorithm and optimizes hyperparameters.
What data formats are supported?
Forecast accepts CSV files with time series data. Additionally, you can add Related Time Series (prices, promotions) and Item Metadata for better forecasts.
How accurate are the forecasts?
Accuracy depends on data quality and quantity. Forecast provides confidence intervals (P10, P50, P90) for each forecast so you can quantify uncertainty.
What is the difference from SageMaker?
Forecast is a specialized service for time series forecasting without code. SageMaker offers full control for custom ML models but requires ML expertise.
Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of AWS (official documentation). This page does not represent an offer by AWS.