Running an ABC/XYZ-Segmentation and a Time-Series Analysis

Objective

After completing this lesson, you will be able to understand and apply the ABC/XYZ - segmentation using the time-series analysis results.

Time-Series Analysis

If you are not sure about the nature of time-series data in your planning area, you can set the system to analyze the values of a selected input key figure based on predefined rules, and save the identified properties in the background. This process is called time-series analysis.

The figure describes the Analyzing of the Time-Series.

The process consists of the following steps:

As an administrator or configuration expert, you set up the rules for the analysis in the Manage Forecast Automation Profiles app.
As a business user, you schedule and run a forecast automation job in the Application Jobs app, for which you select the Time Series Analysis option.
Optionally, you can check the results of the forecast automation job that you just ran. To do so, select or open the corresponding forecast automation profile in the Manage Forecast Automation Profiles app and choose Show Analysis Results.
A new screen opens that shows the following information:
- The planning objects for which the time-series have been analyzed on the selected calculation level.
- A summary of the time-series properties that were identified for each planning object.
- Details about each property that was checked during the analysis.
Note
You can filter and personalize the table to see only the results that are relevant for you.
If you are working with a large number of planning objects and want to check the distribution of the various time series properties among them, you can switch from the table view to a pie chart. When you choose a segment in the pie chart, information is displayed about the property it represents as well as the number and percentage of planning objects identified as having that property.
You can filter the properties displayed in the pie chart, if you wish. For example, you can set the system to show only the distribution of downward trend and additive seasonality.
The results are displayed even if you didn’t choose any attribute or key figure for saving them. This allows you to assess the efficiency of the forecast automation profile and change some parameters in it before executing the next run, thus getting more exact results.
Utilize analysis results.
Time series properties are leveraged in one or more of the following applications:
- In the Manage Forecast Models app, where you can choose the Consider Time Series Properties option after setting the system to utilize multiple forecasts using the Choose Best Forecast method. If you do so, the system checks the time-series properties that were identified by the most recent forecast automation job for the selected planning objects, and uses them to filter out the algorithms that are not expected to calculate an appropriate forecast.
- Additionally, you can set the outlier correction preprocessing algorithm to consider time-series properties. If you do so, the algorithm can detect outliers that don’t vary significantly from the mean or median but do vary from the seasonality or trend pattern in the data.
- In the Manage ABC/XYZ Segmentation Rules app, where the system considers the results automatically during XYZ segmentation.
- In the SAP IBP, add-in for Microsoft Excel, where you can use them to limit and filter your planning views if you have saved the properties in an attribute and other results in key figures.
  Note
  You can modify the values of the attribute that is used to save time series properties but these manual changes will not influence the automated forecasting processes.
- In the Assign Forecast Models app, where you can filter the set of available planning objects in order to identify the forecast models that suit them the most.

Use the Time-Series Analysis

ExerciseStart Exercise

Business Example

As inventory planner, you would like to analyze the time-series to understand the different behaviors of the demand signals within the product portfolio. For this exercise, we already prepared a time-series analysis profile. The main purpose of this exercise is not the customizing itself, but the information provisioning to advise inventory planners how to use SAP Best Practices content within the system.

Prerequisites

Authorizations to access and set up a time-series analysis with a profile through the Manage Forecast Automation Profiles app are prerequisites.
A defined Planning Area where this analysis should run, for example, ZIOUNIFIED ( a copy of SAPIBP1) is mandatory.
Define a key figure which serves as an input key figure, for example, IOSALES, as an initial step.
Define Calculation Levels at which this analysis should run, for example, a combination of customer, location, and product at a weekly level: WKPROLOCCUSTGROUP.

Steps

Open the SAP Fiori launchpad within SAP IBP, and verify the options within the Manage Forecast Automation Profiles app.
1. Log on to the SAP IBP Web UI.
2. Navigate to Demand Planner Group and open the Manage Forecast Automation Profiles app.
3. Note that you can use following options for all available profiles: Create, Edit, Delete, Show Analysis Results, and Run Now.
4. In addition, you can Search for your profiles.
5. You can also click on the icon at the very top on the right side to share the Share Forecast Automation Profile via the Share in Social Collaboration Tool option to send this information to other users.

Select the predefined profile ID: Time Series Analysis - ZIOUNIFIED ## in planning area ZIOUNIFIED, and verify the settings. If there is no profile assigned to your number or the settings differ, please create a new profile and/or apply the settings mentioned below

Forecast Automation Profile Setting


Planning Area	ZIOUNIFIED
Input for Analysis
Calculation Levels	Customer ID - Location ID - Product ID
Periodocity	Week
Additional Calculation Levels for Analysis	not needed
Historical Periods	52
Minimum Required Data Points	10
Input for Analysis	IO Sales
Target UoM	EA (Each)
All other settings	leave as set by default
Outputs for Calculation Level	LOCID - PRDID and Customer Group

Select the ID: Time Series Analysis - ZIOUNIFIED ## in planning are ZIOUNIFIED.
When you select this profile you will see three tabs: Calculation Levels, Analysis, and Administrative Information.
Note the options on the top – this profile can be edited or deleted.
In case you need to Show Analysis Results, you can also use this option.
Under Calculation Levels, identify the Target Calculation Level settings, and the Additional Calculation Levels for Analysis settings (if any).
Select Edit.
Select IO SALES under Input for Analysis.
Select EA (Each) under Target UoM.
Observe how the Target Calculation Level changes.
Select Week under Periodicity.
Under Analysis, you can maintain the time settings as Historical Periods and Minimum required data points.
Select 52 weeks in the past as historical periods.
There should be a minimum value as input to execute this analysis. Therefore, under Minimum required data points, maintain 10 inputs.
Under Time Series Analysis, inventory planners can maintain values to influence the analysis of time-series characteristics like seasonality, trend and intermittency, and volatility. In order to continue with the exercise the default values shown can remain unchanged.
Under Outputs for Calculation Levels, observe that the weekly level including the combination LOCID - PRDID, and Customer Group were selected.
Under Change Point Detection, observe the predefined settings within this profile.
Choose Save.
If a warning message appears, please confirm.
Return to initial screen of the app.

Run your ID: Time Series Analysis - ZIOUNIFIED ##.
1. Return to the main screen of the Manage Forecast Automation Profiles app and select the ID: Time Series Analysis - ZIOUNIFIED ##.
2. Choose Run Now.
3. Note that you are redirected automatically to the Application Job app where you can see the Forecast Automation Template and the Forecast Automation Job Name.
4. Choose Step 2 and verify that you activate this job immediately.
5. Since we are not working in cycles, we have selected Single Run.
  Note
  You can schedule Application Jobs with certain frequency using other options available.
6. Choose Step 3.
7. Select the ZIOUNIFIED as Planning Area.
8. Select your profile.
9. A filter is not necessary at this point.
10. Note that the following Analysis Settings are checked: Time Series Analysis and Change Point Detection.
11. We want to activate this job as soon as possible based on the Execution Time under Current Period.
12. Add a comment as Running a time-series analysis, and select the Reason Code: RC_INVENTORY.
13. Finally, don't share this information with anyone at this point.
14. Choose Schedule.
15. Go to the Application Jobs app to control the status of the job.
16. Click on Go to refresh the status of the job.
17. When the status of the job is marked as Finished click on the job. Choose Log Status: Information to analyze the technical steps executed by the application job.
  Note
  This analysis requires a lot of system resources to identify change points, trend, seasonality, and continuity.
18. Open the Manage Forecast Automation Profiles app.
19. Select the ID: Time Series Analysis - ZIOUNIFIED ##.
20. Click on Show Analysis Results to see the list of results.
21. Click on Show Distribution to see the distribution chart.

Result

Results

At the end of this exercise, you have used the Manage Forecast Automation Profiles app to analyze the Time-Series. We have not run the ABC/XYZ Segmentation yet. In the upcoming exercise, we will analyze the product portfolio using the ABC/XYZ Segmentation.

ABC/XYZ Segmentation

Purpose

ABC/XYZ segmentation is the method of grouping planning objects based on the values of a selected key figure. The two types of segmentation serve different purposes as follows:

ABC segmentation is the prioritization of planning objects based on their relative importance. For example, you may want to categorize the combinations of product and customer based on the revenue key figure.
XYZ segmentation is the classification of planning objects based on their demand volatility. For example, you may want to categorize the combinations of product and location based on the confirmed quantity key figure or a forecast error measure such as MAPE.

The following figure shows the characteristics that the various combinations of ABC and XYZ segments imply:

The figure describes the ABC/XYZ Segmentation.

For example, a planning object that has the values BY assigned to it was sold in medium volume during the segmentation period and is moderately volatile.

Segmentation helps you define more specific alerts and reports, and generate more accurate results for demand planning and inventory planning. Among others, it allows you to do the following:

Tailor your forecast strategy and inventory optimization to the characteristics of segments.
Assign more appropriate forecast models to planning objects.
Analyze forecast accuracy by product segments.
Identify planning objects with a relatively high or low level of forecastability.
Identify inventory items that require closer attention.
Define alerts for specific segments only.
View analytic charts with regards to specific segments.

Prerequisites

The segmentation process consists of the following steps:

The figure describes the Segmentation process.

Create a segmentation profile:
You can use the Manage ABC/XYZ Segmentation Rules app to create segmentation profiles that contain the configuration data and rules for the segmentation calculations.
Schedule and run segmentation:
Once the segmentation rules are set, you can use the Application Jobs app to schedule or start segmentation runs in the system.
Note
It is not possible to start a segmentation job directly in the SAP IBP, add-in for Microsoft Excel.
For each segmentation run, you can optionally use a planning filter. For example, if you want segments to be calculated by country, you can schedule different jobs using filters for different countries using the same or different segmentation profiles. Note that you can only choose from those filters that are compatible with the segmentation profile.
Visibility filters are not considered for segmentation jobs.
If you delete a segmentation profile that was used in a background job, the job will not run correctly any longer. Even if you recreate the profile with the same name, the job will not work. You need to create a new job using the new profile.
Fine-tune the results:
You can view the segmentation results in the SAP IBP, add-in for Microsoft Excel. You can also overwrite them in two ways, depending on whether you want to change a single value or several values at the same time.
- To change a single segmentation value, do the following:
  1. Select Manage in the Master Data Workbook group of the ribbon.
  2. In the window that opens, choose the master data type that you have selected for storing the segmentation results in.
  3. Choose a planning object and select the Edit checkbox in the same window.
  4. Change the segmentation values for the planning object that you chose.
- To change several segmentation values at the same time, do the following:
  1. Open the worksheet that shows the latest segmentation results.
  2. To open the worksheet, choose Create... under Master Data Workbook in the Master Data group of the ribbon.
  3. Choose the master data type that you have selected for storing the segmentation results in.
  4. Select the Edit checkbox in the same window.
  5. Change the segmentation values for any planning objects listed in the master data workbook that opens.
Leverage the results:
The results of the segmentation run are stored as values (for example, A, B, C) of master data attributes. You can use these values to generate reports or define alerts for the segments you select. For example, you can check forecast accuracy for a specific product segment in a chart in the Dashboard app.
You can also use the segmentation results to filter the set of planning objects in the SAP IBP, add-in for Microsoft Excel and in the Assign Forecast Models app.

Set Up an ABC/XYZ Segmentation

ExerciseStart Exercise

Business Example

As inventory planner, you would like to analyze the product portfolio to understand the different behaviors of the demand signals. For this exercise, we have already prepared an ABC/XYZ profile.

Prerequisites

Authorizations to access and set up an ABC/XYZ segmentation with a profile through the Manage ABC/XYZ Segmentation Rules app, are prerequisites.
Define a Planning Area where this segmentation should run, for example, ZIOUNIFIED.
Define a Key figure which serves as input key figure, for example, IO Demand Forecast (IOFORECAST). For this exercise, we will use IOFORECAST.
Defined Periodicity for this analysis, for example, a combination of customer, location, and product at monthly/weekly level, including the time scope, calculation horizon, and the offset for calculation horizon.

Steps

Open the SAP Integrated Business Planning (IBP) for Supply Chain Web UI and verify the options within the Manage ABC/XYZ Segmentation Rules app.
1. Log on to manage the SAP IBP Web UI.
2. Navigate to the General Planner Group and open the Manage ABC/XYZ Segmentation Rules app.
3. Note that you can use following options: New, Edit, Copy, and Delete for every profile.
4. In addition, you can Search for your profiles.
5. You can also share information via Share Segmentation Profile.
6. In addition, you can filter by Planning Area.

Select the predefined profile ABC_XYZ_Segmentation_Profile, and verify the settings.

Select the ABC_XYZ_Segmentation_Profile under the ZIOUNIFIED (Unified Planning Area).
Copy this profile with the name and the description ABC_XYZ_Segmentation_Profile_##.
Open the copied profile.
When you open the profile, you will see three tabs: General, ABC Segmentation, and XYZ Segmentation. Close the disclaimers (if any).
Note the options on the top: This profile can be edited, copied, shared, and deleted. Note: Sharing is only possible in case the system settings are set accordingly.
Under General, identify the Planning Area and the Change history settings, for example, created by, created on, changed by, and changed on.
Under ABC Segmentation, you can activate the segmentation: ON (Checked)/OFF.
Observe that for this profile, the ABC Code (ABCID) attribute is the field from the Product Table used for saving results.
The IO Demand Forecast key figure is the Segmentation Measure in this profile. If any other Segmentation measure rule was selected, change it to IO Demand Forecast..
Periodicity, Time Scope, Calculation Horizon, offset for Calculation Horizon, and Target UoM are time settings that the inventory planner should consider before running this analysis.
Periodicity: Week
Time Scope: Past
Calculation horizon: 156 weeks (three years in the past)
Offset for calculation horizon: 0 weeks
SAP IBP offers several Segmentation Methods. For this profile, we have predefined (1) By Pareto Principle (Sorted and Cumulated %).
Target UoM: Ea (Each)
Keep the default values for A (70%), B (20%), and for C (10%).
Note
Thresholds for A, B, and C segments should be determined by each company based on business intelligence for the specific industry. Every company should determine all time settings including: periodicity, time scope, calculation horizon, offset for calculation horizon, and target unit of measure as well, based on business intelligence for the specific industry.

Under XYZ Segmentation, inventory planners can maintain settings to classify the product portfolio by frequency of sales similar to ABC/XYZ Segmentation Settings:

XYZ Segmentation
XYZ Segmentation	ON (Checked)
Target Attribute for Segmentation Results	XYZ Code (XYZID)
Key Attributes of the Target Master Data Type	PRDID
Segmentation Measure	IO Demand Forecast
Base Planning Level	LOCID - PRDID - CUSTGROUP
Calculation Level	PRDID
Periodicity	Week
Time Scope	Past
Calculation Horizon	156 Weeks
Offset for Calculation Horizon	0 Weeks
Calculation Strategy	Calculate Variation
Calculation Method	Coefficient of Variation (CV)
Segmentation Method	(1) By Thresholds
Consider Results of Time Series Analysis	Checked
Target UoM	EA (Each)
X	0.3
Y	0.75
Z	No maximum value (by default)

Note

Thresholds for X, Y, and Z segments should be determined by each company based on business intelligence for the specific industry. Every company should determine all time settings including: periodicity, time scope, calculation horizon, offset for calculation horizon and target unit of measure as well, based on business intelligence for the specific industry.

Under Target UoM, you can use the unit of measure that is relevant for your company, for example, for this profile we use Each (EA).
Choose Save. Close the disclaimers/info messages (if any).

Analyze situation through the Excel UI..
1. Go to Master Data Workbook in the Excel UI in the SAP IBP ribbon using the connection ZIOUNIFIED.
2. Create a master data view including the Product table to verify the results of the segmentation profile.
3. In the Create Master Data Workbook window, select Worksheet Type Master DataMaster Data Type: Product. Choose Base Version and OK.
4. Go to ABC Code column and XYZ Code column.
5. Analyze the results. the columns might be empty (partially).
Run the predefined Job Name: ABC/XYZ Segmentation.
1. In WebUI, navigate to General Planner Group and go to the Application Job app.
2. Choose Create and select ABC/XYZ Segmentation.
3. Add your ## in the job name field to be able to find the your job after it is finalized (ABC/XYZ Segmentation ##).
4. Note that you activate this job immediately. Because we are not working in cycles, we have selected Single Run.
  Note
  You can schedule application jobs with certain frequency using other options available.
5. Choose Step 3.
6. Select the ZIOUNIFIED planning area and the _BASELINE version.
7. Select the ABC_XYZ_Segmentation_Profile_##.
8. Mark the current period as Execution Time.
9. Add a comment as ABC-XYZ Segmentation, and select the Reason Code: RC_INVENTORY.
10. It is not necessary that you share this information with anyone at this moment.
11. Choose Schedule.
12. Go to Application Jobs to control the status of the job.
13. Select Go to refresh the status of the application job.
14. When the job already has finished, you will see the Status as Finished.
15. Click on the job. Choose Log Status: Information to analyze the technical steps executed by the application job.
Analyze results through the Excel UI and make manual adjustments if possible.
1. Go to Master Data Workbook in the Excel UI in the SAP IBP ribbon using the connection ZIOUNIFIED.
2. Create a master data view including the Product table to verify the results of the segmentation profile.
3. In the Create Master Data Workbook window, select Worksheet Type Master DataMaster Data Type: Product. Choose Base Version and OK.
4. Go to ABC Code column and XYZ Code column.
5. Analyze the results. Please note, that only some products might have classification data, as not all have a demand forecast.

Result

Results

Now you have used the SAP Best Practices for SAP Integrated Business Planning (IBP) for Supply Chain to analyze the product portfolio through the Time-series Analysis and ABC/XYZ Segmentation.

Go to Course