Automatically include local weather information
With Weather Index, Amazon Forecast can increase your forecasting accuracy by automatically ingesting local weather information in your demand forecasts with one click and at no extra cost. Weather conditions influence consumer demand patterns, product merchandizing decisions, staffing requirements, and energy consumption needs. When you use the Weather Index, Forecast trains a model with historical weather information for the locations of your operations and uses the latest 14-day weather forecasts on items that are influenced by day-to-day variations to create more accurate demand forecasts.
Works with any historical time series data to create accurate forecasts
Amazon Forecast can use virtually any historical time series data (e.g., price, promotions, economic performance metrics) to create accurate forecasts for your business. For example, in a retail scenario, Amazon Forecast uses machine learning to process your time series data (such as price, promotions, and store traffic) and combines that with associated data (such as product features, floor placement, and store locations) to determine the complex relationships between them. By combining time series data with additional variables, Amazon Forecast can be 50% more accurate than non-machine learning forecasting tools.
Automated machine learning
No machine learning expertise is required to build an accurate time series-forecasting model that can incorporate time series data from multiple variables at once. Amazon Forecast includes AutoML capabilities that take care of the machine learning for you. Once you provide your data into Amazon S3, Amazon Forecast can automatically load and inspect the data, select the right algorithms, train a model, provide accuracy metrics, and generate forecasts.
Based on the same technology used at Amazon.com
Amazon Forecast includes algorithms that are based on over twenty years of forecasting experience and developed expertise used by Amazon.com. Using AutoML, Amazon Forecast will automatically select the best algorithm based on your data sets.
Easily evaluate the accuracy of your forecasting models
Amazon Forecast provides comprehensive accuracy metrics to help you understand the performance of your forecasting model and compare it to previous forecasting models you’ve created that may have looked at a different set of variables or used a different period of time for the historical data. Amazon Forecast allows you to create multiple backtest windows and visualize the metrics, helping you evaluate model accuracy over different start dates.
Amazon Forecasts and their associated accuracy metrics are visualized in easy-to-understand graphs and tables in the service console. Once forecasts are generated, you can navigate to the relevant forecast by picking it from a list of available forecasts. For example, a specific product within your full catalog of products. Visualization allows you to quickly understand the details of each forecast and determine if adjustments are necessary.
Integrate with your existing tools
Amazon Forecast can be easily imported into common business and supply chain applications, such as SAP and Oracle Supply Chain. This makes it easy to integrate more accurate forecasting into your existing business processes with little to no change.
Generate probabilistic forecasts
Unlike most other forecasting solutions that generate point forecasts, Amazon Forecast generates probabilistic forecasts at three different quantiles by default: 10%, 50% and 90%. In addition, you can choose any quantile between 1% and 99%, including the 'mean' forecast. This allows you to choose a forecast that suits your business needs depending on whether the cost of capital (over forecasting) or missing customer demand (under forecasting) is of importance.