Forecast Usage widget
The Forecast usage widget is a powerful tool for viewing a number of parameters and data points to view item usage over time. The Forecast usage widget can be viewed in two ways, depending on user preference.
When the widget is expanded, it displays the material number on the left side and a subtitle of the forecast mode and usage/projected demand periods. You can view the usage as a bar and line graph or in a spreadsheet format when more specific details are required.
To toggle between the two views, click the graph ()
and grid view (
) icons in the upper right corner of the
widget.
The icon provides access to the forecast and clipping
settings in the Calculation attributes window.
Grid View
Under the grid view, recorded monthly usage levels that are collected over a period of years is shown. This view always shows the same records in a stacked manner-in a compressed year-over-year format in rows. Usage records are arranged per storage locations (SLoc). To view storage locations, click Show store locations to view the underlying stores, and the lower-level store location records. You can toggle this option to hide or view store locations on and off.
Display Filter icon
You can also filter the display usage period by using the filter icon on top of the table. You can select the number or usage periods in groups of 12-month periods. The filter used to display periods impacts the displayed selection of underlying store locations as well, when the top level is expanded. If the display is increased or decreased, all usage periods that are displayed at all levels match the chosen setting. If the display filter selected is larger than the usage periods available, then the extra periods are unavailable on all location levels.
Projected Demand
Projected demand periods are displayed shaded according to the legend visible on the upper right side of the widget. You can edit the top-level location projected demand by using the edit button next to the top-level location, see below for more information on editing usage figures.
Edit quantity for usage figures
The grid view also enables users to either manually enter new usage figures or to alter usage figures by specific factors for each storage location by month.
Users can edit usage figures in one of two ways:
- Select single or multiple figures by clicking them, and then clicking Edit.
- This presents the Edit quantity window that allows users to edit the selected usage figures by specific factors, that is, to add, subtract or multiply by a given value, or to set the usage figure to a specific value.
- You can also edit the usage figures directly within the Forecast Usage widget by clicking the edit button next to the expanded level records, or on the top-level record for Projected Demand. You can enter a new figure either by incrementing the existing value (by using the up and down arrows), or by manually entering a new figure. When the edit button is selected, only the usage periods available for edit can be changed, any usage periods outside the value range are not available for editing.
Graph View
The line and bar graph format of the Forecast Usage widget visually displays an array of data points, including:
- Clipping Limit
- Forecast
- Transfer Usage
- Average
- +1 Std Dev Hovering the mouse over any point within the graph produces a tooltip with specific information regarding that particular point in time on the graph. This allows for analysis of the data points at a granular level.
Legend
When you hover the cursor over an item in the legend it highlights that item exclusively in the graph, fading out all others.
When a legend item is clicked that item becomes hidden and is effectively removed from the graph. Clicking the same legend item a second time restores it to the graph.
Using this method, you can customize the graph to show only the specific data points required.
Calculation attributes
Clicking the icon in the upper right of the widget opens
the Calculation attributes window.
It displays the following settings that can be changed:
- Forecast Modes and Usage Periods
- Projected demand periods
- Forecast periods
- Clipping code
- Forecast Method
- Forecast Factor and Factor due date.
Forecast Modes and Periods
The Forecast Modes and Periods allows users to choose which datasets are used in the calculation of forecast information. This might be actual historical usage data, projections regarding usage or a combination of two modes. You can also set the range of historical records that are viewed (1 – 60 months) and how far into the future forecasts are calculated and projected (1 – 24 months). If the forecast mode is set to New Trend, the usage periods, projected demand periods, and forecast periods cannot be edited. If the forecast modes "Use Projected Demand only" or "Use Actual and Projected usage" are selected, then you are able to edit the projected demand usage periods. For more information on Forecast modes, see the Projected Demand and Forecast Modes topic Projected demand and Forecast Modes topic.
Clipping code
Clipping is used to minimize the impact of irregular usage of an item. When the usage in any particular month exceeds the clipping limit, the calculations replace the expected month's usage with the average monthly usage. When this occurs the month is said to be Clipped.
You can modify the clipping behavior by changing the clipping method. Available options are:
- None: No clipping limit is applied when this is selected. You specify the actual monthly values that are always used, and the clipping limit cannot be changed.
- Manual: You can specify the clipping limit. Months exceeding this limit will be clipped.
- Automatic: This is the default. The system calculates the clipping limit as 2.5 times the standard deviation + the average monthly usage. Months exceeding this limit is clipped . You cannot change the clipping limit by using this option.
Forecast Method
There are various Forecast methods that are used in IBM Maximo Inventory Optimization. This window provides a drop-down list from which users can select a Forecast method to be applied to the data. If you select a Forecast Method that doesn't require a parameter, such as Best, then you are unable to edit the forecast method parameter value. However, if you select a method that does require a parameter, such as Double Exponential, you are able to edit the parameter value.