Forecast Summary Chart

Summary

The forecast summary chart in Forecast Manager shows a summary of several different forecast-related series in one easy-to-read chart.

It is visible primarily in Forecast Manager under the "Summary" chart tab, and also in the "Forecast Summary..." popup dialog accessible throughout StockIQ in various places.

The goal of this chart is to give you a nice visual picture of your demand and forecast, so that you can quickly asses the state of a part, such as:

• Is my demand increasing or decreasing significantly?
• Do I have substantial seasonality on this item?
• Does my forecast appear to be in line with my demand history?
• Do I have outliers for which I should create an event?

In Forecast Manager, the header of the chart shows you:

• the name of the item or hierarchy node you have selected
• Whether it is "Auto" or "Manual" forecasted
• If it has been detected to be "Highly Seasonal"
• It's overall last 12 months vs next 12 months forecasted growth, comparing the independent demand to the operational forecast.

Chart Series

The chart shows several different series on it.  Broadly, they can be broken up into demand series, forecast series, and then the statistical model series.

Hovering your mouse over any particular point will show you all values for that particular month and week, as well as some non-charted information like the number of hits, historical COGS and Revenue for historical points.

You can also control some of the data shown on the chart using the Forecast Manager Display Options

TIP: You can show/hide series in the chart (and all charts in StockIQ) by clicking on the series in the legend.

TIP: You can hide all series but one by using ALT+Click on the series in the legend

TIP: Use your mouse wheel to zoom in and out on the chart

Demand Series

• Independent Demand - Independent demand are the sales of this product directly, not as part of a kit or a hub/spoke relationship.  You can double-click on any point in the independent demand series to see the Sales Order Detail popup dialog.
• <Additional Demand Series> - You may also add additional demand series to your chart, such as bringing in Point-Of-Sale (POS) data to help visualize customer behaviors.
• Dependent Demand - This is demand placed on the part as a result of it being part of a dependency, such as being a component of a finished good assembly, or being a hub in a hub/spoke relationship.  You can double-click on any point in the dependent demand series to see the Sales Order Detail popup dialog, which will show the Dependent Demand tab.  
  • NOTE: If you are using customer-centric or hybrid forecasting, this series is not shown at the item-site-forecast group level, since dependent demand is only something that can be calculated by the item-site.
• Total Demand - This is the sum of your independent and dependent demand.  
  • NOTE: If you are using customer-centric or hybrid forecasting, this series is not shown at the item-site-forecast group level, since dependent demand is only something that can be calculated by the item-site.
• Lost Sales - If able, StockIQ will calculate lost sales for this item and add them to the demand history as a purple area above the independent demand for that particular period.  You can control whether these lost sales estimates are applied to your demand history for the forecast algorithm using the Lost Sales Settings Tab

Forecast Series

• Operational Forecast - This is the independent, operational forecast to which you are purchasing and operating for this item.  Usually this comes from the statistical model if you are auto forecasting, but it can also be any manual forecast that you enter, that is different from the statistical forecast.
• Dependent Forecast - This is portion of forecast that is due to a dependency of which this item is part, such as being a component of an assembly, or a hub for some spoke warehouses.  This forecast is not directly editable, since it is the result of the forecasted replenishment requirements of the assemblies and spokes.
  • NOTE: If you are using customer-centric or hybrid forecasting, this series is not shown at the item-site-forecast group level, since dependent forecast is only something that can be calculated by the item-site.
• Total Forecast - This is the sum of your independent and dependent forecast.
  • NOTE: If you are using customer-centric or hybrid forecasting, this series is not shown at the item-site-forecast group level, since dependent forecast is only something that can be calculated by the item-site.
• (Optional Forecasts) - In StockIQ, you can configure additional forecast series for display, such as an organizational Budget forecast, an annual snapshot forecast, etc.  These will be displayed on the chart as well if you wish. 

Statistical Model Series

• Trend - The trend shows you the overall movement of this part through history, and projected out into the future.  Normally with the StockIQ forecast algorithm, you will see it rise and fall over time as demand for the product changes.  If you increase volatility, the trend will follow the history of the item more closely, and be more reactive, at the cost of stability / jitter.   If StockIQ detects an item to be highly seasonal, the trend will be a straight line through the years of history that you have, showing the overall trend of increase or decrease on the item.
• Statistical Model - This is StockIQ's statistical model for the item, based on the Algorithm Settings.  It is not directly adjustable, but you can adjust the calculations by changing the algorithm parameters for whatever algorithm you have selected.
• Simulated Trend, Simulated Model - If you run a simulation of changes using the Edit --> Run Simulation option in Forecast Manager, two additional series will appear, showing you the results of the simulation, as compared to your current trend and model.

• Statistical Model Band - This is the light red band (aka "the pink haze" to some ;-) ) appears on the Forecast Manager screen. The red shaded area signifies the expected error range in which the forecast of this SKU may lie in a given time period, based on the expected statistical model error as shown in the 'Forecast Error' tab. The smaller the band area, the more precise the statistical model is able to predict the forecast. You may see a larger band area for SKUs with very little demand history and highly sporadic demand.

  The purpose of the band is to give the users a visualization of what the accuracy of the statistical model looks like, and the range of future possible error on the forecast.

Events

Events are also shown on the chart as bars, which indicate the quantity of the event.  When you mouse over a period in which there are events, you will see a little lightning bolt icon showing the name and quantity of the event.  There is now an option in Forecast Manager Display Options to prevent the Events from showing on the chart.  See Events Overview

On Hand Quantities

If you are viewing a forecast summary chart at the on-hand level, you will also see the "Projected On Hand Quantity" and "Projected Actual Available Quantity" series presented in the chart.  These are from the replenishment order schedule, and represent your projected on hand quantity (assuming all forecasts, demand, supply, and StockIQ suggestions are followed), as well as the projected actual available (which is the same as above minus StockIQ suggestions).

Leading Indicators 

The Forecast Manager chart shows the effect of any applied leading indicator. It compares your demand history against two forecast models:

 When the projected leading indicator shows a big spike or dip, the Statistical Model line should shift in the same
 direction, with the effect fading back toward the "Before LI" line over the fade period.