040_Mg_DP_SCM_APO

© 2012 SAP AG

Best-Practice Document

Manage Demand Planning in SAP SCM (4.1, 5.x, and 7.0)

Dietmar-Hopp-Allee 16 D-69190 Walldorf

DATE

October.2012

SOLUTION MANAGEMENT PHASE SAP SOLUTION

Operations & Continuous Improvement SAP Supply Chain Management (SCM)

TOPIC AREA SOLUTION MANAGER AREA

Application & Integration Management Business Process Operations

Best-Practice Document Manage Demand Planning in SAP SCM (4.1, 5.x, and 7.0)

© 2012 SAP AG page 2/49

Table of Content

1 Management Summary 3

1.1 Goal of Using this Service 3

1.2 Alternative Practices 3

1.3 Staff and Skills Requirements 3

1.4 System Requirements 4

1.5 Duration and Timing 4

1.6 How to Use this Best-Practice Document 4

2 Best-Practice Procedure and Verification 5

2.1 Preliminary Tasks 5

2.2 Monitoring Procedure 9

2.2.1 Business Process Step 1: Extract Data from InfoCube (SAP NetWeaver BW) 10

2.2.2 Business Process Step 2: Generate Flat File (Legacy System) 12

2.2.3 Business Process Step 3: Extract Data from Logistics Information System (ERP) 13

2.2.4 Business Process Step 4: Load Data into APO Data Mart (APO) 13

2.2.5 Business Process Step 5: Set Up Master Data (APO) 14

2.2.6 Business Process Step 6: Load Data into Planning Area (APO) 18

2.2.7 Business Process Step 7: Calculate Proportional Factors (APO) 19

2.2.8 Business Process Step 8: Adjust Historical Data (APO) 20

2.2.9 Business Process Step 9: Create Demand Plan (APO) 23

2.2.10 Business Process Step 10: Determine Forecast Accuracy (APO) 24

2.2.11 Business Process Step 11: Perform Interactive Planning (APO) 24

2.2.12 Business Process Step 12: Plan Promotions and Product Life Cycle (APO) 26

2.2.13 Business Process Step 13: Copy liveCache Data to InfoCube (APO) 28

2.2.14 Business Process Step 14: Transfer DP to SAP ERP (APO) 28

2.2.15 Business Process Step 15: Perform Material Requirements Planning MRP (ERP) 29

2.2.16 Business Process Step 16: Release DP to SNP (APO) 29

2.2.17 Business Process Step 17: Compute Constrained Plan (APO) 30

2.2.18 Business Process Step 18: Release SNP Plan to DP (APO) 30

2.3 Related Tools and Activities 31

3 Further Information 42

3.1 Troubleshooting 42

3.2 Background Information and References 44

3.3 Index 48


© 2012 SAP AG page 3/49

1 Management Summary

To ensure that this best-practice document is the one you need, consider the following goals and requirements.

1.1 Goal of Using this Service This best-practice document enables you to set up a business process management and monitoring concept for the business process Demand Planning, which is part of the SAP Supply Chain Management (SCM) solution, using the SAP Advanced Planning and Optimization (SAP APO). It is valid for SAP SCM 4.0, 4.1, 5.0, 5.1 (2007), and 7.0.

This business process management and monitoring concept aims to:

Define procedures for business-process-oriented monitoring, error handling, and escalation management for Demand Planning.

Define the roles and responsibilities for all persons involved in the customer’s support and monitoring organization with respect to Demand Planning.

These procedures ensure the smooth and reliable flow of the core business process to meet your business requirements.

1.2 Alternative Practices You can get SAP experts to deliver this best practice on site if you order the Solution Management Optimization (SMO) service SAP Business Process Operations.

1.3 Staff and Skills Requirements To implement this best practice, you require the following teams:

Application Management Team

The application management team should create the SCM/APO business process management concept specified in this best-practice document. This team combines experts from several areas of your company:

Business department

Solution support organization (for example, the IT department and the Help Desk)

Implementation project team

Execution Teams

The execution teams are the following groups, which are taken together from your Solution Support Organization:

The business process champion for each business process

Application support

Development support

Program Scheduling Management

Software monitoring team

System monitoring team

More information about roles and responsibilities of these teams can be found in the superordinate best-practice document General Business Process Management, which you can obtain through SAP Solution Manager.


© 2012 SAP AG page 4/49

Necessary or Useful Training Courses (with a focus on important system operation know-how):

ADM355 APO System Administration

PDEF08 LiveCache Administration

TTW060 SAP APO Technical Administration

TEWA60 SAP APO liveCache Monitoring

Necessary or Useful Training Courses (with a focus on DP process and planning know-how):

SCM210 Core Interface APO

SCM220 Demand Planning

1.4 System Requirements This document applies to SAP SCM 4.1, 5.0, 5.1 (2007), and 7.0.

1.5 Duration and Timing

Duration

Creating a business process management concept may take approximately one week per business process.

Implementing the business process management concept may take approximately one additional week.

Timing

The best time to apply this best-practice document is during the planning phase or during the implementation phase of your SAP solution.

1.6 How to Use this Best-Practice Document In advance, read the entire document to get an overview of its structure, content, and details.

Determine your APO DP core business process to be monitored. If you use the Core Interface (CIF) within this process, also get the respective best-practice document dedicated to CIF. Record all relevant steps of your core business process; use the example business process as template. Exclude template process steps that you do not perform.

For every process step, take the monitoring elements from the tables and insert them into your own template. Complete the information according to your specific requirements, for example, frequency and time of monitoring activity. If the process step includes CIF data transfer, add the information from the CIF document, section Operation and Monitoring of the CIF.

Do not forget to include the respective information for interfaces other than CIF and for business process steps performed with your own (Y-, Z-) programs. Determine the related monitoring activities, tools, and responsible teams and fill in the table accordingly.

For activities that are not directly related to a business process step, like those mentioned in section System Administration Related to the APO CIF in the CIF best-practice document, create a separate table.

Proceed in the same way with all your other core business processes and other activities you want to monitor.


© 2012 SAP AG page 5/49

2 Best-Practice Procedure and Verification

2.1 Preliminary Tasks

The SCM System Landscape The substantial components of an SAP SCM system landscape are summarized in the following table and shown schematically in the subsequent figure.

SAP APO system SAP Advanced Planning and Optimization within the SAP SCM system facilitates the strategic, tactical, and operational planning processes.

APO consists of several software components: A relational database system (RDBMS) just as there is in any SAP ERP system, known as the APO DB; an SAP NetWeaver; the APO application programs; a separate, very fast, object-oriented SAP DB database called liveCache; application programs running in liveCache (the liveCache Applications); and a number of programs that execute elaborated optimization algorithms, called the optimizer engines. These components can run on the same or on different servers.

OLTP system The online transactional processing system covers functionality for sales and distribution, material and inventory management, controlling, shop floor control, logistic execution, and so on.

OLAP system An online analytical processing system, such as SAP NetWeaver Business Warehouse (SAP NetWeaver BW), provides cumulated historical data as a basis for future extrapolation purposes in APO Demand Planning.


© 2012 SAP AG page 6/49

The various strategies for using SAP APO, SAP ERP, SAP NetWeaver Business Warehouse (SAP NetWeaver BW), and possibly other data processing systems in combination are called integration scenarios.

Usually, an APO system is connected to one or several systems that provide historical and current data, such as sales figures, for your Demand Planning business process. These systems can be SAP ERP, SAP NetWeaver BW, and/or third-party and legacy systems or a combination of these. As SAP APO comprises a complete SAP NetWeaver BW, the tools for extracting data from other systems and transferring them to APO for Demand Planning purposes are the same as SAP NetWeaver BW uses. The APO BW reads the data using remote function call (RFC) techniques, IDocs, or OS file access.

The planning results that come out of the Demand Planning process are fed into other data processing systems for further planning or execution purposes. The system types that can be used here are SAP APO itself (in particular its modules SNP and / or PP/DS), SAP ERP, or again third-party (legacy) systems. Third-party systems are not considered in this Best Practice document.

This best-practice document is therefore based on a general and common integration scenario for setting up an SAP Supply Chain Management solution using SAP APO. As data sources, SAP NetWeaver BW, SAP ERP, and flat files from any other system type can be connected to the SAP APO system. The results from the Demand Planning process are processed by another module of the APO system itself, or are transferred to an SAP ERP OLTP system via the Core Interface (CIF).


© 2012 SAP AG page 7/49

For information about solution management for CIF, refer to the respective best-practice document. SAP Advanced Planning and Optimization (SAP APO) is the planning component of SAP Supply Chain Management (SAP SCM). SAP APO is used to make strategic, tactical, and operational decisions and supports you in performing the following planning activities:

Demand Planning (DP)

Supply Network Planning (SNP)

Production Planning (PP)

Detailed Scheduling (DS)

Deployment

Transport Load Builder (TLB)

Transport Planning and Vehicle Scheduling (TP/VS)

Global Available-to-Promise (GATP)

SAP APO is primarily a planning tool, although some industry-specific execution functions are available (such as production backflush for repetitive manufacturing). In standard business scenarios, execution functions such as confirmations, goods receipt, purchasing, and so on are performed in the SAP ERP OLTP system, which contains all functionality for (among many others) materials management (MM), sales and distribution (SD), production order processing (PP-SFC), Logistics Execution System (LES), and controlling (CO).

Additional information can be found in the SAP Documentation.

Demand Planning

The core business process you run in your company for planning your demand and supply, though it is also based on the SAP SCM solution, may differ more or less from the canonical process described as follows and illustrated on the next page.

The past and actual data your demand planning is based on is obtained from the source systems such as SAP ERP, SAP NetWeaver BW, or others, and loaded into BW InfoCube(s) located in the APO system, the APO data mart(s). New characteristic value combinations need to be generated, for example, if new products are planned or if a planning-relevant customer purchases a product for the first time. The master data needs to undergo a realignment step, for example, if the sales organization structure or product hierarchy is changed. Normally, some historical data is copied from InfoCubes to key figures stored in liveCache, in the planning area(s). For disaggregation purposes, proportional factors have to be (re)calculated. Historical data has to be adjusted and cleared from exceptional influences (such as promotions, outliers, or unique occurrences like the introduction of a new currency) in order to serve as a reliable basis for the forecast.

In the batch forecast run, the demand plan is created. Afterwards, the accuracy of the forecast can be determined by comparing the differences between the actual values and the ex-post values. In interactive planning, the results of a forecast created in the background are verified, manual corrections and hardly quantifiable processes and trends derived from experience are applied to the demand plan, and computations of key figures are performed using macros. (However, interactive planning should be restricted to the processing of small data volumes. Larger amounts of data should be processed in background wherever possible.) Using dedicated tools, the influence of promotions can be estimated and evaluated, and the different phases of the life cycle of products can be modeled.

Many APO installations save the data kept in liveCache into an InfoCube from time to time and use this InfoCube, for example, for reporting purposes involving the Business Explorer. The final and confirmed demand plan can be either transferred to SAP ERP creating planned independent requirements and serving as basis for the material requirements planning run (MRP). You can take production or distribution restrictions into consideration by releasing the demand plan to the SNP application of APO. You can then transfer the resulting supply network plan back to DP for comparison with the unconstrained demand plan. Any major discrepancies between these two plans can trigger re-forecasting, and ultimately, re-planning. For example, you may determine that the capacity situation demands a change in the promotion strategy.


© 2012 SAP AG page 8/49

The following figure shows all the business process steps whose monitoring is defined by this best –practice document:

Extract Data from InfoCube

SAP NetWeaver BW System

Generate Flat File

Legacy or Third-Party System

SAP APO System

Load Data into APO Data Mart

Update Master Data

Load Data into Planning Area

Calculate Proportional Factors

Adjust Historical Data

Create Demand Plan

Determine Forecast Accuracy

Perform Interactive Planning

Plan Promotions and Product Life Cycle

Copy liveCache Data to InfoCube

Transfer DP to R/3

Release DP to SNP

Demand Planning

Supply Network Planning

Extract Data from LIS

SAP ERP OLTP System

Perform Material Requirements Planning

(MRP)

Compute Constrained Plan

Release SNP Plan to DP

INTERFACES

tRFC / IDoc

OS File System

tRFC / IDoc

CIF


© 2012 SAP AG page 9/49

2.2 Monitoring Procedure

In applying this best-practice procedure, you create a company-specific, process-oriented monitoring concept. This concept consists of monitoring activities to be performed for each business process step and its respective monitoring objects.

When adapting this concept for your company, you must specify the times, responsible teams, and escalation paths (teams) for the monitoring activities associated with each business process step and its monitoring objects.

Under each business process step, you will find the following information:

A detailed functional description of the process step

Monitoring activities for the process step

Error handling, restart ability, and escalation

A monitoring object table, listing each relevant monitoring object, showing the:

o Monitoring object

o Monitoring transaction or tool

o Monitoring frequency

o Monitoring time (intentionally left blank, to be filled in according to your schedule)

o Indicator or error

o Monitoring activity or error handling procedure

o Responsible team

o Escalation procedure

As the frequency of Demand Planning processes vary from daily activities at certain companies to only monthly or quarterly activities at others, the monitoring frequency in these monitoring object tables is partly only a rough estimate and has to be adapted to your particular business process. During the going-live and stabilization phase of your APO implementation project, you should closely monitor all items listed in this document. Once you have some experience with system behavior, error occurrences, and application operations, you can decrease the monitoring frequency, but you should never reduce it to zero (except for functions you do not use). Normally, you need to monitor important planning jobs after each run. You can check whether regular jobs with lower priority (such as, certain clean-up jobs) run as scheduled less frequently; for example, you can check daily jobs weekly.

The following seems obvious but should nevertheless be mentioned: Besides the monitoring of jobs described in the business process steps below, it is essential that you check all jobs that are running in your system at least several times per day for abnormal terminations (status “cancelled”, see Error Handling, Restartability and Escalation), and that you investigate and correct these terminations appropriately. This check can be done easily, for instance, with transaction SM37 by entering the time interval since the latest check and selecting all jobs with status “cancelled”. If you have no automatic notification in place that informs the people responsible for Program Scheduling Management in your Support Organization of abnormally terminated jobs, you need to take measures to ensure that this is done manually in a reliable and timely manner.

A number of jobs must run periodically in a live SAP ERP installation, for example, the jobs for deleting outdated jobs or spool objects. For details and comments, see SAP Note 16083.

Depending on the types of data processing systems you use in your solution landscape, and which of these contribute data for your Demand Planning process, you will possibly not execute all of the business process steps 1, 2, and 3 described below, but only one or two of this type.

http://service.sap.com/~iron/fm/011000358700000431401997/0016083E


© 2012 SAP AG page 10/49

2.2.1 Business Process Step 1: Extract Data from InfoCube (SAP NetWeaver BW)

Every SAP APO system includes a complete SAP NetWeaver Business Warehouse (SAP NetWeaver BW) system, which should be used exclusively for APO purposes. In this business process step, data is extracted from a separate SAP NetWeaver BW system. All periodic monitoring tasks must still be performed in the APO system, as well as job definition and most of the configuration tasks.

The Administrator Workbench (transaction code RSA1) is the tool for loading data into the InfoCubes from a source system. Using the Administrator Workbench scheduler, you create an InfoPackage specifying source, target, time, background parameters, and selection criteria for the data transfer to the InfoCube. You can see a log for the request that is generated for the transfer job in the Manager that you can access from the Administrator Workbench by selecting the InfoCube and choosing it in the context menu.

Up to SAP SCM 5.0, a single job is set up in the BW of the APO system with different processing steps to

extract the required data from the source system, transfer it to APO, and write it into the InfoCube(s). For the data transfer method, you can select either of two options in the respective InfoSource:

Transfer into Persistent Staging Area (PSA): The data is sent directly from the source system to the Business Information Warehouse and is stored in the Persistent Staging Area (PSA). From there your data can either be updated manually or automatically in the InfoCubes. The transfer type for this is transactional remote function call (tRFC).

In customizing transaction RSCUSTV6, you can set threshold values for the data load, such as packet size and partition size.

Data transfer intermediate documents (IDocs): The data is packed into IDocs by the source system and sent to the SAP NetWeaver BW. In SAP NetWeaver BW, your data is saved persistently and non-transparently in the IDoc store. From there the InfoCubes can be updated manually or automatically.

SAP recommends using tRFC, as it has a performance advantage. If IDocs are used, you can easily find the required monitoring screens (IDoc list, status and data records as in transaction BD87) in transactions RSA1 or RSMO.

As of SAP SCM 5.0 (based on SAP BI 7.0), the InfoPackages are only used for transferring the extracted data to the Persistent Staging Area (PSA); use of PSA is mandatory. The further posting to the data targets (InfoCubes or DataStore Objects) is done by predefined data transfer processes (DTPs). For the entire data load (a combination of InfoPackage and DTP), the use of process chains is mandatory. Maintenance and monitoring of process chains is done in transaction RSPC as also the scheduling of the entire data extraction process.

For details and performance hints, see SAP Notes 130253, 400191, 409641, 124532 and 115407.

For all types of databases, dropping the indexes of the fact table before loading a huge amount of data is a viable option. However, it depends on the ratio of the data volume in the cube and the data volume to be loaded whether it is faster to update the indexes with the data or drop and rebuild them. There is no general rule, so do a test and choose the faster method. With Oracle, indexes are rebuilt in parallel if the database has the correct parameters, so it is often faster with dropped indexes. See also the best-practice documents dealing with the data transfer to an SAP NetWeaver BW system.

Monitoring Activities

Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in the following table in order to safeguard this business step.

The data transfer process, as well as certain actions in RSA1 and RSMO, write detailed information to a central SAP ERP basis component called the application log. The log entries are subdivided into objects and sub-objects and classified by importance. In case of errors or problems, the application log can be used to get more detailed information about what happened and how to resolve the problem. The monitoring transaction for the application log is SLG1, where you should specify the (sub) objects and the time interval you want to be displayed. Relevant objects for the SAP NetWeaver BW are:







© 2012 SAP AG page 11/49

RSAR BW Metadata maintenance

RSAU Update rules

RSD BW data basis

RSDMD Master data maintenance

RSO_REPOSITORY BW Repository

RSSM Scheduler; Monitor; Tree callback

RSTRAN_R Transformation Runtime

RSTRFN Transformation

Jobs for Running and Monitoring "Extract Data from InfoCube" (APO)

To ensure a timely and consistent data transfer from your SAP NetWeaver BW into your APO system, schedule the following job on a regular basis:

SCM 4.1: Data Extraction and Load with report RSBATCH1. Define this job using the scheduler in the Administration Workbench (transaction RSA1) of the SAP NetWeaver BW component of APO. The scheduler generates a background job with name prefix BI_BTCH and processes the selected InfoPackage. With the help of InfoPackage groups, you can collate data requests that logically belong together and schedule them using the scheduler.

As of SCM 5.0: Data Extraction and Load as a combination of single steps of a process chain. The scheduler in the Administration Workbench (transaction RSA1) of the SAP NetWeaver BW component of APO would still generate a background job with name prefix BI_BTCH and process the selected InfoPackage, but this will only result in loading the data to the PSA. The further processing to the data targets has to be triggered manually by scheduling the appropriate data transfer process (DTP) in this case.

In Administrator Workbench monitoring (select Monitoring in transaction RSA1 or call transaction RSMO), you can oversee and control the data loading process into the InfoCubes and you can analyze errors that may have occurred (choose Monitor Assistant). You can also call the monitor from a selected InfoPackage or DTP. To be sure to select the correct time window and data to be displayed, call the selection screen by choosing Monitor New Selections.

In the Process Chain Display (transaction RSPC), you can easily oversee and control the data loading process chains and its single components; by double-clicking single steps you will branch to the respective (for example, InfoPackage or DTP) maintenance dialogue where the BI monitor is available also.

Jobs for the Maintenance of the Application Log

The application log entries are stored in tables with name prefix BAL*. As there are very many applications that use this Basis component and often many table entries are made, it is important to delete obsolete application logs from the database regularly to prevent these tables from overflowing and to help ensure that the applications and log retrieval run smoothly. You can delete them using transaction SLG2 or in the background using the appropriate report:

Delete Obsolete Application Logs with report SBAL_DELETE. A log can only be deleted when it has reached its expiry date or if it has the "Deletion before expiry" attribute. For more information, see SAP Note 195157.



© 2012 SAP AG page 12/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error

Monitoring Activity or Error Handling Procedure

Respon-sibility

Escalation Procedure

APO job BI_BTCH*

These jobs perform the data transfer from the different data sources to APO InfoCubes.

RSMO or SM37

Depend-ing on your process

Status Check if job is running as scheduled.

If the job is not scheduled as provided by application support, schedule it accordingly.

Program scheduling management

Contact application support.

APO: Administration Workbench Monitor

RSMO After every data load

Red or yellow traffic lights

After selecting one request, read the error message and diagnosis under tab strip Status and the processing log under Detail.

Analyze and correct the error situation.



APO: Process Chain Display Log View

RSPC After every data load

Red or yellow steps in the chain

Choose Process Monitor

from the step’s context menu.

Read the error message and diagnosis under tab strip Status and the processing log under Detail.




APO: Evaluate Application Log for BW related objects

SLG1 At least daily

Log class Check if there are very important logs (flagged red) or important logs (flagged yellow).

Read message long text and/ or details, if present. Analyze error situation, evaluate error severity and impact, and take corrective action.



APO report SBAL_DELETE

This report deletes obsolete Application Logs.

SM37 Weekly Status Check if job is running as scheduled.

If the job is not scheduled as provided by application support, schedule it to run weekly.

Program scheduling management


See general issues of Error Handling, Restartability, and Escalation.

2.2.2 Business Process Step 2: Generate Flat File (Legacy System)

This business process step is similar to Business Process Step 1: Extract Data from InfoCube (SAP NetWeaver BW), except that the InfoSource is defined differently. The data from the legacy or third-party system must be provided by methods and programs of that system in flat file form with format *.csv (comma separated values) or *.asc (ASCII). The file should be sorted by key sequence and be uploaded from the application server rather than a presentation server, if possible. For performance issues, see SAP Notes 124532 and 359761. It is also valid for SAP SCM 4.1 and onwards, except that report




© 2012 SAP AG page 13/49

/SAPAPO/RMDP_ICUBE_PERFORM has been replaced by /SAPAPO/TS_PSTRU_TOOL; see SAP Note 393655. For details, see Maintenance of DP Data Storage Structures.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed under Monitoring of Business Warehouse Activities in order to safeguard this business step.

2.2.3 Business Process Step 3: Extract Data from Logistics Information System (ERP)

Like Business Process Step 2: Generate Flat File (Legacy System), this business process step resembles Business Process Step 1: Extract Data from InfoCube (SAP NetWeaver BW). In this case, the data is automatically updated from selected info structures in the Logistic Information System (LIS) of an SAP ERP system to the BW component of APO. As the extractor used to get the data from LIS is generic, you can use any SAP-delivered or custom-defined info structure (table Snnn with 3-digit number nnn) as a data source. Demand Planning can use information from all parts of logistics, such as sales, inventory, purchasing, shop floor control, and quality management.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed under Monitoring of SAP NetWeaver Business Warehouse Activities in order to safeguard this business step.

2.2.4 Business Process Step 4: Load Data into APO Data Mart (APO)

As described under Business Process Step 1: Extract Data from InfoCube (SAP NetWeaver BW), you use the Administration Workbench scheduler of your APO system to set up jobs for the data transfer from the different data sources to APO. Such a job extracts the required data in the source system, transfers it to APO, and updates the specified InfoCube(s) in different processing steps. You can control and monitor these steps from the BW monitor transaction RSMO. In case the data update from the Operational Data Store (ODS) or PSA to the InfoCube is not performed automatically but triggered manually, you need to monitor this step again using RSMO.

For the data load step, it is important that statistics and indexes of the InfoCube(s) the data is loaded into are current. To check this in transaction RSA1, you can right-click on the respective InfoCube and choose Manage. On the next screen, choose the Performance tab and select Check Statistics (available only for certain databases; see below), Check Indexes, and Check Aggregate Indexes, respectively. InfoCube statistics should be created regularly in background using the tool BRCONNECT. Alternatively to transaction RSA1, you can also use transaction RSRV for the analysis of SAP NetWeaver BW objects. For information about data transfer to an SAP NetWeaver BW system, see also the best-practice documents.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in the following table and under Monitoring of SAP NetWeaver Business Warehouse Activities in order to safeguard this business step.

Jobs Necessary to Keep the APO Data Structures in Good Condition (APO)

To ensure that database statistics and indexes for BW-specific data structures are kept up to date, and that liveCache data structures (time series) are kept consistent, schedule the following jobs to run on a regular basis:

Compute Histograms for SAP NetWeaver BW InfoCubes using the BRCONNECT tool as well as the “normal” database statistics.



© 2012 SAP AG page 14/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


SAP tool BRCONNECT

This tool can be used to create DB statistics for all InfoCubes.

Only for systems running with Oracle, SAP DB or IBM DB2/UDB (DB6) database

DB13 Weekly Status Check if tool is running as scheduled.

If the tool is not scheduled on a regular basis, schedule it to run once a week with parameter percent = 100.

Database administration

Contact software monitoring team.

Indices of APO InfoCubes

RSA1 or RSRV

Weekly Red or yellow traffic light

Select Check (Aggregate) Indexes. If the traffic light is yellow or red, select Repair Indexes (now) or Delete Index (batch) and Create Index (batch).




2.2.5 Business Process Step 5: Set Up Master Data (APO)

During operation, new characteristic value combinations occur. For instance, your company introduces new products or existing products are additionally manufactured at another location. There are six ways of creating characteristic value combinations:

Individually and manually, by completely defining characteristic values in transaction /SAPAPO/MC62 (this creates one data record). You also can delete characteristic value combinations with /SAPAPO/MC62 (single values or generic by using the * wildcard).

Automatically, by generating the combinations based on existing data from an InfoCube. With this option, the system generates all the new combinations that it finds in the horizon specified. You can plan automatic generation periodically so that new characteristic value combinations are added each time but old ones are never deleted. This function can also be called from transaction /SAPAPO/MC62 or can be executed in the background using report /SAPAPO/TS_PLOB_MAINTAIN.

You can create them by referring to a different planning object structure (POS) in your system.

New characteristic combinations can be based on an InfoProvider (like an Infocube, but other Infoproviders are also possible).

They can be based on a flat file upload.

Characteristic value combinations can be automatically created using a new BAdI.

You may wish to realign your data because of structural changes in the characteristics. For example, one of your existing products will in the future belong to a new product group, so you want to assign the product to the new product group and realign the planning data to the new characteristic value combination(s). You can create the new characteristic value combinations automatically using an SAP realignment tool. You create a table that contains the changes, and new characteristic value combinations are created from this table. You decide whether you want to copy existing combinations or to replace them. The data stored in the key figures is copied correspondingly.

Remember that you cannot modify standard characteristics 9ALOCNO (location) and 9AMATNR (product).


© 2012 SAP AG page 15/49

Concerning the usage of basic characteristics and / or navigation attributes in your planning object structure, see SAP Note 413526.

For the generation of characteristic combinations and the corresponding time series, it is important to update statistics and indexes of the planning object using report /SAPAPO/TS_PSTRU_TOOL.

From SAP SCM 5.1, you can delete obsolete combinations. Obsolete objects are combinations without APO master data, planning data, or master and planning data.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in the following table and under Maintenance of DP Data Storage Structures in order to safeguard this business step.

Jobs for Running and Monitoring "Set up Master Data" (APO)

To ensure a complete and correct setup of master data needed for your Demand Planning process, schedule the following jobs to run on a regular basis:

Generate new Characteristic Value Combinations

This can be done with report /SAPAPO/TS_PLOB_MAINTAIN. This report generates the same characteristics and characteristic value combinations (that is, the corresponding time series) in the selected planning object structure, as it finds in the source version of the InfoCube specified. The report must run after every data load from the InfoCube that contains the relevant characteristic combinations.

Adapt Time Series for New Characteristic Combinations

Use report /SAPAPO/TS_LCM_PLOB_DELTA_SYNC. For example, if you create a new characteristic combination without the option Create time series objects (see transaction /SAPAPO/MC62), you get the error “No liveCache anchor found” when selecting this characteristic combination. In this case, execute report /SAPAPO/TS_LCM_PLOB_DELTA_SYNC for the basis planning object structure of the planning area. This creates the corresponding liveCache anchors and liveCache time series for all planning areas that use this basis planning object structure and for all versions of these planning areas for which time series objects already exist. /SAPAPO/TS_REALIGNMENT cannot work without the liveCache anchors. So before executing realignment (or if missing liveCache anchors are reported), run report /SAPAPO/TS_LCM_PLOB_DELTA_SYNC. Mark the option display error log to get a log in case of unsuccessful execution. See also SAP Note 573127. If necessary, you can determine and repair inconsistencies for the planning area by executing report /SAPAPO/TS_LCM_CONS_CHECK and choosing the Repair option (see Maintenance of DP Data Storage Structures).

Realignment of Characteristic Value Combinations

As far as you have the need to consider structural changes in your characteristic combinations, you can use transaction /SAPAPO/RLGCOPY or, if you want to do the realignment as an automated task, use report /SAPAPO/TS_REALIGN_COPY directly for this activity.

Re-Initialize Your Planning Area with report /SAPAPO/TS_PAREA_INITIALIZE.

The creation of time series objects for the planning area is known as initializing the planning area. The system creates a network of characteristics and key figures in liveCache, time series objects are created for a version and a period of time. You can extend the time series into the future and delete the oldest periods from liveCache by re-initializing the time series periodically. In this way, you can maintain a "rolling horizon." The system recognizes where new characteristic value combinations or new time series exist, and creates time series objects only for the new master data. You can define key-figure specific initialization horizons in order to save a lot of memory in your liveCache. For example, for historical key figures, you can initialize only time buckets in the past and for forecast figures only buckets in the future. This can be defined in a pop-up window during the start of the report or in the transaction for key figure administration. Report /SAPAPO/TS_PAREA_INITIALIZE locks the complete version of the planning area. If the lock cannot be set, the report terminates. You have the additional possibility to perform the initialization with parallel processes by defining a parallelization profile in transaction /SAPAPO/SDP_PAR. For




© 2012 SAP AG page 16/49

this, you have to define a block size and a maximum number of parallel processes for the planning area. After starting the initialization job, it automatically creates the desired number of parallel processes. If the report terminates abnormally for any reason other than lock problems (you can see this in the job log), run report /SAPAPO/TS_LCM_CONS_CHECK with option Repair (see Maintenance of DP Data Storage Structures) before restarting/SAPAPO/TS_PAREA_INITIALIZE. For example, if you no longer need a certain planning version, you can remove its data from liveCache by deleting the time series objects with report /SAPAPO/TS_PAREA_DE_INITIALIZE.

Jobs for the Maintenance of the Planning Object Structure

Maintain Statistics and Indexes for Planning Object Structures with report /SAPAPO/TS_PSTRU_TOOL. Run this report in separate background jobs, choosing options Calculate Statistics and Check Indexes for every master planning object structure and review the list of messages being displayed. If problems are reported by the index check, run it with the option Repair Indexes. A separate background job (job name with prefix BI_STAT) is started to execute this request. These administrative jobs must be run after every load of data to a planning object structure. For further details, see SAP Notes 503363 and 510639. Since SAP SCM 5.1, POS-related basic functionality is moved to SCM BASIS.

Note: to cover any unexpected issues, the old functionality is kept as well. It can be still used when user parameter /SAPAPO/SCMB_PSTRU is set. This is a temporary solution, and will be removed in later releases.

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


Report /SAPAPO/TSPLOB_MAINTAIN

SM37 Depending on your process


If the report is not scheduled as provided by application support, schedule it accordingly.



Output of APO report

/SAPAPO/TS_PLOB_MAINTAIN

SP01 After every run

Messages in spool list

Review the messages listed like “Number of incorrect combinations.”

Application support

Contact process champion.

APO report /SAPAPO/TS_LCM_PLOB_DELTA_SYNC

This report creates liveCache anchors and new time series.






Output of APO report /SAPAPO/TS_LCM_PLOB_DELTA_SYNC

This report creates liveCache anchors and new time series.

SM37 After every run

Status If status is “cancelled”, check job log for error messages.






© 2012 SAP AG page 17/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/TS_REALIGN_COPY

This report realigns characteristic combinations.



If the report is not scheduled as provided by Application Support, schedule it accordingly.



Output of APO report /SAPAPO/TS_REALIGN_COPY

Special log file, view and administrate with /SAPAPO/RLGCOPY

After every run

Messages in log file. If report was run in background, an additional spool file exists

Review the list for problems reported like “Error: Realignment job cancelled,” “No connection to liveCache,” or “Aggregate, indexes and statistics must be checked” (and more).

Take corrective action if necessary.

Application support

Contact business process champion.

APO report /SAPAPO/TS_PAREA_INITIALIZE

This report re-initializes time-series objects.






Output of APO report /SAPAPO/TS_PAREA_INITIALIZE



Review the list for problems reported like “Time stream (XXX) for year (YYYY) not generated”.

Take corrective action if necessary.

Application support


For release 4.1:

APO report /SAPAPO/TS_PSTRU_TOOL with option Calculate Statistics

This report checks and generates statistics for a POS.

SM37

After every data load into the POS





For release 4.1:

Output of APO report /SAPAPO/TS_PSTRU_TOOL


Message type (icon)

Look for stop icons (cancellation), and red (errors) or yellow (warnings) lights. Read message long text if present.




© 2012 SAP AG page 18/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/TS_PSTRU_CONS_CHECK for checking and repairing inconsistencies (if repairing is possible)


Status Check if job ended correctly. Program scheduling management



2.2.6 Business Process Step 6: Load Data into Planning Area (APO)

You store permanent data in InfoCubes. In general, this is either historical actual data or forecasts that you want to keep for future comparison. Actual data for planning is kept in liveCache in planning areas. Demand Planning itself is conducted in liveCache. For example, to use historical data as the basis for forecasting, you either access data directly from the InfoCube or transfer data between it and liveCache. Both methods have advantages and disadvantages. Placing a lot of a data in liveCache improves performance, but it also requires more memory, and you need to consider this when sizing your APO system and when designing your planning area.

SAP strongly recommends storing all key figures that are used in background or interactive Demand Planning in liveCache, even if not enough memory is available. It is faster to read data from liveCache (even if it is swapping) than to read from an InfoCube.

As it is not possible to interactively change key figures that are stored in an InfoCube, you need to copy key figures to the liveCache if you want to change them. However, you should not change historical actuals; instead, copy them to a separate key figure “corrected history” that you can maintain manually afterward.

For this business process step, use transaction /SAPAPO/TSCUBE for an online copy (do this only in exceptional cases and for a very small amount of data), or use report /SAPAPO/RTSINPUT_CUBE for background mass data processing. It is possible to set up parallel jobs using individual selections for shorter processing times. To set up the parallelization correctly, you have to make sure that the individual selections only contain selected combinations exactly once; make sure there is no overlap between the selections. To achieve a high efficiency of the parallel jobs, make sure that each selection contains roughly the same number of characteristics, or in other words: Each selection should be of the same size and running time as the other selections used in parallel. For an automated parallelization, you can also use parallelization profiles for loading data into a planning area. You can navigate from transaction /SAPAPO/MC8E directly to the maintenance of parallel processing profiles or use transaction /SAPAPO/SDP_PAR for direct access. Using parallelization profiles is safer and more convenient than the manual creation of selections, because the system automatically ensures there is no overlap in the parallel processing and the workload from the job is distributed equally between the number of processes specified. Two options help reduce the upload data quantity and can help you with the creation of parallel selections: The flag Ignore data sets with 0 forecast quantity directly reduces the data quantity for processing. The flag Ignore values which are not within the selected horizon reduces the data quantity as well. You can navigate directly to the report log by clicking a button.

Other data that you want to use in a specific Demand Planning business process may already be present in liveCache, but is stored in a different planning area and/or version. In this case, you can use a similar function called Copy/Version Management (TA /SAPAPO/TSCOPY, report /SAPAPO/RTSCOPY), which allows you to copy data between planning areas and versions. Again, it is possible to set up parallel jobs using individual selections for shorter processing times. There should be no overlap between selections, and they should be of roughly the same size and running time as the other selections used in parallel. You can also use parallelization profiles for copying data between planning areas and versions. You can navigate from transaction /SAPAPO/MC8E directly to the maintenance of parallel processing profiles or use transaction /SAPAPO/SDP_PAR for direct access. The flag Ignore values which are not within the selected horizon reduces the data quantity; you can navigate directly to the report log by clicking a button.


© 2012 SAP AG page 19/49

Jobs for Running and Monitoring "Load Data into Planning Area" (APO)

To ensure a complete and correct copy of key figures, schedule the following jobs to run on a regular basis:

Load Planning Area Version with report /SAPAPO/RTSINPUT_CUBE. To get a list of success and error messages from this report, mark the Log checkbox in the selection variant you use for this job. The list looks like the application log (transaction SLG1), but is not stored separately from the job. For performance hints, see SAP Note 428102.

Copy Data between Planning Areas and Versions with report /SAPAPO/RTSCOPY. To get a list of success and error messages from this report, mark the Log checkbox in the selection variant you use for this job. The list looks like the application log (transaction SLG1), but is not stored separately from the job.



Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/RTSINPUT_CUBE

This report copies key figures from an InfoCube to a Planning Area






Output of APO report /SAPAPO/RTSINPUT_CUBE


Message type (icon)

Look for stop icons (cancellation) and red (errors) or yellow (warnings) lights. Read message long text if present.

Application support


APO report /SAPAPO/RTSCOPY

This report copies key figures between planning areas and versions.






Output of APO report /SAPAPO/RTSCOPY


Message type (icon)


Application support



2.2.7 Business Process Step 7: Calculate Proportional Factors (APO)

One way to perform disaggregation is to use constant proportional factors. The constant proportional factors are derived from historical data or, occasionally, from past planning data. You can do disaggregation



© 2012 SAP AG page 20/49

differently for different key figures. Generate constant proportional factors if you wish to disaggregate planning data based on the constant proportional factors either always or if no planning data exists. The proportional factors are stored in key figure APODPDANT. Using transaction /SAPAPO/MC8V, the computation of the proportional factors for a selected planning area can be started to run online or in the background. Alternatively, you schedule a background job (in transaction SM36) that runs report /SAPAPO/RMDP_SHARE_MANAGER.

You can use parallel processing for the calculation of proportional factors. You can navigate from transaction /SAPAPO/MC8V directly to the maintenance of parallel processing profiles or use transaction /SAPAPO/SDP_PAR for direct access. After defining block size and number of maximum parallel processes, you have to assign the parallel processing profile to the selection variant. The next time you run this selection as background job, it will automatically run with parallel processing.

From SAP SCM 5.0, the proportional factor key figure (APODPDANT) should have the following settings (see SAP Note 1140247):

- fixable = true

- distinguish between zero and nothing = true

- zero fixable = true

After upgrade, the time series consistency check for planning areas will display “Key figure description for key figure APODPDANT is inconsistent.” It has to be fixed by running the check in repair mode.



Jobs for Running and Monitoring "Calculate Proportional Factors (APO)"

To allow a correct disaggregation, schedule the following job to run on a regular basis:

Calculate Proportional Factors with report /SAPAPO/RMDP_SHARE_MANAGER. If you wish to view the results, select the Log for Proportion calculation checkbox. Caution: Logging is intended for use in the test phase only. As all characteristic value combinations and the calculated factors are listed (and for all periods, if Calculate detailed proportions for all periods has been selected), this log can become very large.

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/RMDP_SHARE_MANAGER

This report calculates the proportional factors.







2.2.8 Business Process Step 8: Adjust Historical Data (APO)

The original actuals can be adjusted for different circumstances: phase-in/out profile, workday adjustment, past promotions, outlier corrections, and manual adjustments. Various automatic procedures that are activated in the forecast profile can be used to adjust the actuals. Workday adjustment ensures that higher values are forecasted for periods that have many work days. Historical data must be adjusted to take this into account. To generate more exact forecasts, remove the impact of one-time promotions or delivery problems

http://service.sap.com/~IRON/FM/011000358700000431401997/1140247E


© 2012 SAP AG page 21/49

from the actuals. It is usual to make these adjustments not in the original key figure but in the corrected history key figure.

Usually, these corrections are done by macros run in background for mass data processing. You can also adjust the actuals manually. By making manual modifications, you can fine-tune the demand plan. Management overrides are one example of this. To document a modification, write a note in the Demand Planning table and attach it to the cell and the level in which you made the modification. You can also check for any errors in the demand forecast and run some macros to correct it or select a suitable forecast model. For manual corrections, see Business Process Step 11: Perform Interactive Planning (APO). For more information and details on background macro processing, see SAP Notes 412429 and 546079.

Some hints on the design and usage of macros for background processing:

Create a separate data view for each macro background job. The data view should contain only key figures used by the macro itself, and only macros executed as activities by the mass-processing job. Use a different data view whenever different key figures are needed by two macros, or whenever the time periods are different.

Remove all unused macros from planning books associated with mass processing jobs, as they are loaded in an unorganized manner that can be performance-intensive. Unused macros are not helpful to your business process.

Only those macros that are called by an activity are executed in background. These activities are: Start, End, Level change. Default macros are not executed automatically, as they are in dialog. When defining a job, you must therefore first establish which start and default macros have to be executed and in which sequence, so that your own self-defined macro can work correctly. Macros for background should not involve actions like setting colors, hiding fields, display icons, or sending pop-up windows that are frequently included in default macros for online use.

The planning book used for background macros should have only the functionality it uses, that is, application Demand Planning and no other functionality.

For performance reasons, run macros for a delta period rather than the entire horizon wherever possible. Under most circumstances, historical data is not changed, other than adding new periods, one time bucket profile at a time. Rather than performing the full calculation over a long historical period, run a macro to do this once, and then design a new macro that only performs the calculation for the “delta” period, or the period that has just entered the historical horizon.

If you want to speed up the processing, you can use parallel processing profiles to run macros, for example, for the adjustment of historical data. You can navigate from transaction /SAPAPO/MC8E directly to the maintenance of parallel processing profiles or start transaction /SAPAPO/SDP_PAR for direct access. After defining the block size and the number of maximum parallel processes, you have to assign the parallel processing profile to the selection variant. The next time you run this selection as a background job, it will automatically run with parallel processing.

If you need to run two macros on the same key figures with the same level of aggregation, put both into one planning book and execute them as two actions within the same background activity (defined with transaction /SAPAPO/MC8T, see below). You thus save the time needed for loading the data for a second time.



Jobs for Running and Monitoring "Adjust Historical Data" (APO)

To ensure correct and timely Demand Planning, schedule the following jobs to run on a regular basis:

Run Demand Planning in the Background with report /SAPAPO/TS_BATCH_RUN. This mass processing job is created with its necessary settings and scheduled using transactions /SAPAPO/MC8x. You can create and change such jobs using the respective transactions /SAPAPO/MC8D and /SAPAPO/MC8E. Define the activities needed for these jobs in advance via transaction /SAPAPO/MC8T (usually called from the customizing tree). An activity can be a forecast computation, the execution of one or several macros, a release of DP data to SNP orders, or the transfer to SAP ERP. /SAPAPO/MC8I enables you to check the job definition prior to execution and find errors like




© 2012 SAP AG page 22/49

“no macro defined,” “data view does not exist,” or “release profile not found.” With /SAPAPO/MC8G, the jobs are scheduled. To get an overview of the jobs, display the job logs, or jump to spool lists, you can use transaction SM37. To check the results of a DP mass-processing job, use the DP job log. To request a job log, select Generate log when you create a job. To view the job log, call transaction /SAPAPO/MC8K. The job log shows whether the job completed successfully (green traffic light), with warnings (yellow traffic light), or with errors (red traffic light); a message for every characteristic value combination that was processed in the job; forecast error messages if the job included a forecast; and other details. You can also delete job logs with this transaction. You can copy jobs with /SAPAPO/MC8J and delete them with /SAPAPO/MC8F. Parallel processing: Due to the large amount of data to be processed and the limited time available for it, it is often necessary to process background runs (of macros or whatever) in parallel. You can do this by defining several jobs in /SAPAPO/MC8x running the same activity but having a different selection of characteristics. Each selection ID should define a set of characteristic value combinations of approximately equal size, and no individual characteristic combination should belong to more than one selection ID, as this can cause lock issues and it is superfluous to process a combination more than once. These jobs can then be scheduled to run at the same time. Remember that hardware resources like number of CPUs, CPU speed, main memory, and other restrictions limit the total number of jobs that can run on a system at the same time. You can also use parallel processing profiles to run macros or any other planning activity like performing a statistical forecast. You can navigate from transaction /SAPAPO/MC8E directly to the maintenance of parallel processing profiles or use transaction /SAPAPO/SDP_PAR for direct access. After defining the block size and the number of maximum parallel processes, you have to assign the parallel processing profile to the selection variant. The next time you run this selection as a background job, it will automatically run with parallel processing.

Delete DP Job Logs with report /SAPAPO/TS_BATCH_LOGFILE (or transaction /SAPAPO/MC8K). Old logs from planning activities must be deleted regularly to maintain high performance for the related transactions and prevent database tables from overflowing. Make sure that you delete the logs in packages. Select the option Delete all logs with minimum age, reducing the parameter days. See also SAP Note 512184.

Demand planning activities can be executed periodically with very different frequencies. Key figures can be computed with batch macros daily or weekly, whereas a release of the demand plan to production planning or ERP is usually performed at larger time intervals, for example, once a month or once a year. How often you monitor the jobs and delete the DP job logs depends on how often you run these functions.

For dependencies and the concurrent execution of jobs, see Operational Management – Parallel and Concurrent Execution of Jobs. For the handling of alerts generated by a forecast run or a batch macro, see also Exception-Based Monitoring of APO Demand Planning.

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/TS_BATCH_RUN

This report performs specified DP mass processing activities.






DP Job Log /SAPAPO/MC8K

After each run of /SAPAPO/TS_BATCH_RUN

Red or yellow traffic lights shown

According to the warning or error reported (see message long text)

Application support




© 2012 SAP AG page 23/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/TS_BATCH_LOGFILE

This report deletes old DP job logs.






See general issues of Error Handling, Restartability and Escalation.

2.2.9 Business Process Step 9: Create Demand Plan (APO)

The historic data from the InfoCube is used for forecasting. In this process step, you create or modify demand forecasts by applying the most suitable forecast model such as univariate forecasting, multiple linear regression (MLR), and/or composite forecasting. Typically, you do a planning run for multiple products in the background with mass processing, and then adjust the figures interactively. See Business Process Step 11: Perform Interactive Planning (APO).

The planner interactively defines various master forecast profiles to forecast typical historical time series showing the constant, trend, or seasonal patterns. If different key figures are to be forecasted, you must have master profiles for each key figure to be forecasted. You can assign a master forecast profile with a univariate forecast profile and/or an MLR profile and/or a composite profile.

The forecast results depend greatly on which forecast model you use and how you set it up. For a detailed overview on the strategies delivered by SAP, see the SAP Documentation, section Forecast Strategies. You can store multiple Demand Planning scenarios to evaluate different forecast models and settings in Demand Planning before deciding what data to use onward. A demand plan is stored in a planning version within a planning area. It is a good idea to have (one or more) additional key figures containing different forecast scenarios based on the same primary (historical) data rather than creating a separate planning version for each scenario: This reduces the space required in liveCache because the primary data is stored only once instead of n times for every version. Moreover, the different scenarios can be compared easily in a single planning book.

If your business process allows it, you can run your forecast at an aggregated level, which is selected in transaction /SAPAPO/MC8E when defining your background job. The fewer characteristics you select, the coarser and faster your forecast run will be. The forecast at the more detailed levels will result from disaggregation according to the calculation type for this key figure (see transaction /SAPAPO/MSDP_ADMIN). It may even be an advantage and not just a loss of information if you consider the trends and seasonal variation of your historical data not on all the characteristic combinations, but only on a selected subset. Statistical variation at a very detailed level can conceal correlations and trends that are apparent at a more aggregated, coarser level.

If you use MLR, see the SAP Documentation, section Measures of Fit. It is recommended to remove uncorrelated parameters from the profile once they are measured to be non-correlated, since the runtime of the MLR depends on the number of variables used. See also the SAP Documentation, sections Forecast Strategies and Automatic model selection.

There are six forecast errors you can select when defining a univariate forecast profile in transaction /SAPAPO/MC96B, for example mean absolute deviation (MAD) or mean square error (MSE). For performance reasons, it is advisable to configure as few errors as possible.

See also SAP Notes 388216 and 372939.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in Running Demand Planning in Background in order to safeguard this business process step.




© 2012 SAP AG page 24/49

2.2.10 Business Process Step 10: Determine Forecast Accuracy (APO)

APO offers the following ways to monitor forecast accuracy:

Statistical error analysis

Univariate forecast error measurements

Multiple linear regression model measures of fit

Planned/actual comparison

Viewing purpose-designed KPIs with a BW front end (Business Explorer [BEx])

Executing this business process step, you will most probably run one or several macros in background mode to calculate the forecast errors and evaluate the results in transaction /SAPAPO/SDP94 (see Business Process Step 11: Perform Interactive Planning (APO)) or the Alert Monitor. SAP NetWeaver BW reporting tools such as Business Explorer (BEx) are also commonly used.

Once you have gained some experience with production data, it is advisable to restrict the number of forecast accuracy checks. This prevents you from doing things like calculating huge numbers of errors or goodness of fit data for large numbers of different models that do not fit the data.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in Running Demand Planning in Background and Exception based Monitoring of APO Demand Planning in order to safeguard this business step.

2.2.11 Business Process Step 11: Perform Interactive Planning (APO)

Within transaction /SAPAPO/SDP94, you can do almost everything that SAP APO Demand Planning offers. In particular, you can evaluate planning results, compare different scenarios, enter management overwrites, correct key figures manually, run macros, aggregate and disaggregate data, maintain proportional factors manually, fix values and thereby prevent them from being changed by macros or batch jobs, create promotions, run forecasts, and create or view alerts. Demand Planning supports online simulation in multiple planning scenarios, consistent planning throughout your enterprise (top-down, middle-out, or bottom up), drilling up and down, aggregation and disaggregation, and slice-and-dice techniques. Top-down planning is at a high level, such as for a whole region. Middle-out planning is at a medium level, such as for each division within the region. Bottom-up planning is at a detailed level, such as for each product in each division within the region. Demand Planning can also display the planning data in a graphic. Since SAP SCM 5.1, a new chart engine has been integrated, which allows for more flexible business graphics to be created. There are two user parameters that allow the switch between the old and the new graphical engine. For the graphic in interactive planning (/SAPAPO/SDP94), the user parameter /SAPAPO/SDP_CHART_V is used, whereas for the graphic displayed on the forecasting views, user parameter /SAPAPO/FCST_CHART should be used.

By default, the new chart engine will be used; however if the value CHART is set for the user parameter, then the old chart engine will be used.

For performance reasons, any interactive processing of data should be restricted to very small data volumes by specifying the selection criteria in as much detail as possible. This is independent of when this planning step takes place, for instance before or after a planning run. Tasks that can easily be automated, for example because they always take place at a certain point in time and consist of a fixed sequence of calculations (as it applies to macros in most cases), should be performed as background jobs rather than interactively.

In some cases, it is a good idea to define fixed aggregates, according to SAP Note 503363, item 5. In these cases, choose one or two aggregates that will significantly decrease the data volume used by your interactive planning and background jobs. Be aware of the consequences: Creating an aggregate will help the performance of reading data from the liveCache, but generally hurts performance of writing data to the liveCache. You can use transaction /SAPAPO/MC62 to determine which sets of characteristics reduce the number of characteristic combinations significantly. See also SAP Note 413526 on the usage of navigation attributes.




© 2012 SAP AG page 25/49

Some hints on the design and usage of planning books and macros for interactive planning:

Create a separate planning book for each user that contains only the views, macros, key figures, and any other views that are used by that individual user. Users who perform more than one task type should also consider creating more than one planning book to accommodate each task type.

The planning book used for Supply and Demand Planning should contain only the functionality that an individual user actively uses while performing a specific task. For example, a user who performs promotion adjustments should have a planning book specific to this task that has functionality for promotions, any forecasts used, interactive planning, and nothing else.

Minimize the number of default macros used for Supply and Demand Planning; see if any of these can be executed less frequently as start, end, or drill-down macros. Each online user should verify that the macros in her/his planning book are all used regularly as part of the standard business practice, and remove any that are not.

Implement composite time-bucket profiles with coarser time buckets in the long-term past and future, and fine-time buckets only near the present time. For example, showing a year as 9 weeks and 10 months dramatically improves the performance without decreasing the time period shown. Also, any users who do not actively view and use historical data should consider the history not shown flag, as this allows macros to use the data without loading the data onto the screen.

Each online user should create several selection variants to restrict the number of characteristic combinations considered to those that he or she needs for a specific task. For example, a planner who looks at corrected forecasts for location ABC all day on Monday can create four selection variants, each containing different products, for which the forecast is to be corrected. Since SAP SCM 5.1, transaction /SAPAPO/MC77 for maintaining selection variants has been enhanced to allow the assignment of selection variants to a group of users.



The most important high-level tool for monitoring the planning situation and exception-driven application management is the Alert Monitor. The use of the Alert Monitor is supported by an e-mail interface for notifying the responsible persons in case of new alerts. Therefore, planners and/or IT staff do not need to call the monitor transaction frequently just to confirm that nothing important has happened since the last refresh. However, the overall situation shown in the Alert Monitor should be supervised daily or adapted to the frequency of the Demand Planning process in your company.

Jobs for Monitoring “Perform Interactive Planning (APO)”

To ensure a timely and efficient notification about exceptions in Demand Planning, schedule the following jobs to run on a regular basis:

Send Alert Monitor mails with report /SAPAPO/AMON_MAIL_BROADCAST. Depending on user profiles, mails are sent with an overview of existing alerts. The responsible persons then should call the Alert Monitor, investigate the reasons for the alerts, and take corrective actions to keep the production plan close to the needs of your company.

Delete Alert Monitor alerts with report /SAPAPO/AMON_REORG. It deletes old alerts and is of particular importance if you use alerts stored in the database. On database alerts and dynamic alerts in Demand Planning, see Alert Monitor, below.

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


Alert Monitor /SAPAPO/AMON1

At least daily

Check for forecast alerts and Supply and Demand Planning alerts and correct the planning for the reported object appropriately.

Application support



© 2012 SAP AG page 26/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/AMON_MAIL_BROADCAST

This report ensures sending of mails about existing alerts.

SM37 Daily Status Check if job is running as scheduled.

If the report is not scheduled on a regular basis, schedule it to run every hour to at least daily, depending on your requirements.



Alert Monitor mails

SO01 (or respective e-mail system)

Depending on your requirements, at least daily

Check if the mail lists alerts that are important for you. Go to Alert Monitor and process the alerts appropriately.

Application Support


APO report /SAPAPO/AMON_REORG

This report deletes Alert Monitor alerts.


If the job is not scheduled on a regular basis, schedule it to run once a day.



Additionally, see general issues of Error Handling, Restartability, and Escalation.

2.2.12 Business Process Step 12: Plan Promotions and Product Life Cycle (APO)

Promotion Planning

With promotion planning, you can extract past actions (such as special offers) from the actuals so that they are not included in the forecast. Promotions can have a major impact on consumer behavior. In Demand Planning, you can plan promotions or other special events separately from the rest of your forecast. Use promotion planning to record either one-time events, such as the millennium, or repeated events, such as quarterly advertising campaigns. Other examples of promotions are trade fairs, coupons, freestanding inserts, competitors' activities, market intelligence, upward/downward economic trends, hurricanes, and tornadoes. Promotional uplifts can be defined in units or percentages by promotion patterns. A promotion pattern that occurred in the past can be automatically detected using historical sales or estimated by the planner. A promotion pattern can be archived in a promotion catalog; it can therefore be reused if a promotion of the same type is repeated. A copy function in the promotion catalog also supports "like" modeling of similar products, regions, and so on. Several techniques are available for estimating the impact of a historical promotion such as multiple linear regressions, with or without a trend or seasonality. See also SAP Note 540282.

Lifecycle Planning

A product's life cycle consists of different phases: launch, growth, maturity, and discontinuation. In this business process step, you model the launch, growth, and discontinuation phases. For any characteristic values combination, you can use a phase-in profile, a phase-out profile, a “like” profile, or any combination of these. A phase-in profile and a phase-out profile change the demand history or demand forecast of a product by given percentages for a specified period or several periods. See also SAP Note 642593 for details.

These planning tasks are usually executed interactively using transaction /SAPAPO/MP34, which can also be called from Supply and Demand Planning /SAPAPO/SDP94 using the Promotion button. Promotion evaluations can be started via transaction /SAPAPO/MP39. With this transaction, you can also see the attributes of the selected promotions, whereas the Business Explorer shows you key figures only.




© 2012 SAP AG page 27/49



Jobs for Running and Monitoring "Plan Promotions and Product Life Cycle (APO)"

To keep the promotion key figure(s) in liveCache consistent with the promotion data, schedule the following job to run on a regular basis:

Reconcile Inconsistencies between the Promotion Key Figure and Activated Promotions with transactions /SAPAPO/MP46 (Activate and Deactivate Promotions) and /SAPAPO/MP47 (Update for Promotion Consistency), which provide the functionality for mass operations on promotions. The group of transactions in this area can be centrally accessed by running transaction /SAPAPO/MP42 (Promotion Management). Here, you can reset the promotion key figures, deactivate promotions, and activate all promotions again. Use /SAPAPO/MP47 to update the promotion key figures in liveCache. If you want to run one of these transactions as scheduled background job, you can run report /SAPAPO/PROMOTION_ACTIVATION, which provides the functionality of transaction /SAPAPO/MP46, or report /SAPAPO/PROMOTION_UPDATE, which provides the functionality of transaction /SAPAPO/MP47.

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/PROMOTION_UPDATE.

This report reconciles inconsistencies between promotions and planning areas.






Output of APO report /SAPAPO/PROMOTION_UPDATE

SP01 After each run


Look for warning and error messages like “Error saving the promotion data” or “No promotion level is assigned.”

Application support




© 2012 SAP AG page 28/49

2.2.13 Business Process Step 13: Copy liveCache Data to InfoCube (APO)

There are two typical reasons why a business process can contain a business process step that performs a copy of liveCache data to an InfoCube:

1. You want to perform an analysis of the data with an external tool or, for example, with the Business Explorer, or you want to use the data for any kind of further data processing or reporting purposes.

2. You are copying liveCache data as a backup. By copying data, for example, directly after performing a background planning run, you make sure that the planning results can be recovered in case something goes wrong (for example, a power failure or a liveCache crash) before the next regular system backup takes place.

If you are copying liveCache data for reason No. 1 and you use the data for further data processing or reporting purposes, this copy step has to be performed as part of your business process. The processing time can be reduced, for example, by setting up multiple, small copy jobs that can be run in parallel.

If you are copying liveCache data for reason No. 2, there are only a few scenarios where copying the liveCache data to an InfoCube is the best method to serve this purpose.

Copying all data in a planning area to an InfoCube using the methods described below is very memory- and resource-intensive and takes a long time. For this reason, we recommend that you run no other processes concurrently with this. Do not run such a process step more frequently than required for your business process purposes.

There are two methods to export data from planning areas to InfoCubes:

Directly in move mode (“Extract Data from a Planning Area”)

Indirectly in delta mode using ODS objects (“Update InfoCubes from Planning Areas”)

For details, see the SAP Documentation and the best-practice documents dealing with the data transfer to an SAP NetWeaver BW system.

Extracting data directly has the disadvantage that the complete data set is copied to the InfoCube at each update, so the same data is written again and again. With the second method using ODS objects, it is easier to update InfoCubes, as this procedure only updates the new and changed data. However, the delta mode is relatively slow. Depending on your absolute data volume and the relative amount of changes, decide for yourself which method to use.

For performance improvement, see SAP Notes 482494, 384023, and 386735. For other information, see SAP Note 426806.


Apart from safeguarding the general aavailability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed under Monitoring of SAP NetWeaver Business Warehouse Activities in order to safeguard this business step.

2.2.14 Business Process Step 14: Transfer DP to SAP ERP (APO)

You can transfer the results of your Demand Planning activities (that is, your demand plan) to your SAP ERP system for further planning steps, in particular for material requirements planning (MRP). This transfer has to be performed as a mass-processing job in background. The information is taken from the time series of the appropriate final forecast key figure and transferred to SAP ERP via the Core Interface (CIF). In SAP ERP demand management, planned independent requirements are created in the same buckets as those of the data view on which the transfer job is based. The CIF queue entries are written by the transfer job and processed according to the CIF settings.

The transfer job uses report /SAPAPO/TS_BATCH_RUN with a planning activity using an appropriate transfer profile. You can make these activity and profile settings in SCM customizing and define the background jobs needed in the transactions /SAPAPO/MC8x. For details, see Run Demand Planning in the Background and SAP Note 546079.







© 2012 SAP AG page 29/49

As the CIF is an essential component of SAP APO and its business processes, monitoring and administration of this interface is of critical importance for the performance and reliability of any business process that exchanges data between the SCM system and the ERP systems connected to it.

Therefore, it is extremely important that you follow the business process management procedure described in the best-practice document dedicated to the CIF interface.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in Running Demand Planning in Background as well as those listed in the best-practice document mentioned above concerning CIF in order to safeguard this business step.

2.2.15 Business Process Step 15: Perform Material Requirements Planning MRP (ERP)

The planned independent requirements created in ERP by business process step 14 can be used to determine the quantities and requirement dates of raw materials and components needed for the manufacturing of the products planned in Demand Planning. The monitoring procedure for this business process step is described in the best-practice document Manage Production Planning in SCM / APO.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in the best-practice document mentioned above concerning Production Planning in order to safeguard this business step.

2.2.16 Business Process Step 16: Release DP to SNP (APO)

This business process step makes the demand plan available to the supply chain planner if the supply chain planner performs SNP using liveCache orders. The supply chain planner can then use the demand plan as a basis for making sourcing, deployment, and transportation decisions. The simulation process in Demand Planning can be done in conjunction with the simulation of planning scenarios to drive Sales and Operations Planning. For example, you may release two Demand Planning scenarios to SNP for simulation purposes: one based on the assumption of a sales increase of 5% in the next six months, and one based on a sales increase of 10% in this time frame. From a technical point of view, you are copying data from liveCache time-series objects to a forecast ATP category (for instance “FA” = independent requirements) in the liveCache orders.

This release function can be performed online with transaction /SAPAPO/MC90 or report /SAPAPO/RTSOUTPUT_FCST (only for a small amount of data, see SAP Notes 403050 and 546079), or in the background with report /SAPAPO/TS_BATCH_RUN with a planning activity using an appropriate release profile. These activity and profile settings are made in APO customizing and the background jobs needed are defined in the transactions /SAPAPO/MC8x. For details, see Run Demand Planning in the Background. Parallel processing of this function is possible: see also Operational Management – Parallel and Concurrent Execution of Jobs.

There is a Create new orders flag in the release profile, which is called Add data in report /SAPAPO/RTSOUTPUT_FCST. If you set this indicator, new order quantities are created in SNP when you release the demand plan to SNP. The system ignores existing orders in the same SNP version for the same product, location, and periods, so if you release to SNP 20 times, 20 separate orders are created in SNP. If you do not set this indicator, existing order quantities in the same SNP version for the same product, location, and periods are overwritten. Set this indicator only under exceptional circumstances.

Remarks: In contrast to DP, where data is stored either in time series or an InfoCube, in interactive SNP and in distribution resource planning (DRP), data is stored in liveCache orders. However, if you are using interactive Sales and Operations Planning (SOP), for example, to execute supply and demand propagation, the data you wish to use must be stored in liveCache time series as in DP. This also applies for the use of other SNP functionality. For details, see the SAP Documentation, section Data Storage in SNP and DP. In




© 2012 SAP AG page 30/49

SOP you can release the results back to SNP with transaction /SAPAPO/MC90 as well, but with the button Release: Extended, or you can do this in background with report /SAPAPO/RTSOUTPUT: Data Transfer Time Series to Order Network. You may also have mixed planning areas with some order key figures and some time series key figures. To copy selected time series key figures from your DP planning area to your SNP planning area, use report /SAPAPO/RTSCOPY, which is described under Business Process Step 6: Load Data into Planning Area (APO).


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in Running Demand Planning in Background in order to safeguard this business step.

2.2.17 Business Process Step 17: Compute Constrained Plan (APO)

Supply Network Planning with liveCache order objects and/or time series is used to create a rough-cut plan taking limited resources into consideration, such as production or distribution capacities. The supply chain planner can use this to plan resources based on a full and reliable picture of the demand situation. The demand planner can use it to monitor where adjustments to the demand plan were required due to production capacity limitations, distribution limitations, or other constraints.

As the computation of an SNP plan can comprise almost the full functionality of the APO SNP module, such as the use of background planning runs, interactive planning, macros, optimizers, heuristics, and others, no details of this business process step are discussed in this best-practice document. They are part of another document dedicated to SNP. For an overview of available documents, see the best-practice documents in the SAP Service Marketplace.

2.2.18 Business Process Step 18: Release SNP Plan to DP (APO)

In Supply Network Planning (SNP), releasing the plan entails the following concepts:

Releasing the SNP-confirmed forecast (that is, the results from an optimization run) to DP using transaction /SAPAPO/SNPFCST (report /SAPAPO/RELEASE_SNP_FORECAST)

Releasing the SNP plan to DP to compare the constrained plan with the unconstrained plan, or to use the data from SNP in Sales and Operations Planning using transaction /SAPAPO/LCOUT (report /SAPAPO/RTSINPUT, data transfer Order Network to Time Series)

The only relevant case for the business process in this document is the second one: In this “traditional” release of the supply network plan, you release the final supply network plan back to the demand planner to compare the unconstrained demand plan to the constrained supply network plan. Major discrepancies between these two plans could trigger re-forecasting, and ultimately, re-planning. For example, you may determine that the capacity situation demands a change in the promotion strategy. Technically, you must transfer the data from the relevant order categories to the appropriate time series key figures.



Jobs for Running and Monitoring "Release SNP-Plan to DP (APO)"

To ensure the availability of SNP data for the DP process, schedule the following job to run on a regular basis:

Release Order Objects to Time Series with report /SAPAPO/RTSINPUT. To obtain a result log, select the appropriate checkbox in the selection screen of the report. Caution: As all characteristic value combinations that are being updated are listed, this list can become very large.


© 2012 SAP AG page 31/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/RTSINPUT

This report releases SNP orders to DP key figures.






Output of APO report /SAPAPO/RTSINPUT

SP01 After each run

Message type (icon)


Application support



2.3 Related Tools and Activities

2.3.1.1 Alert Monitor The Alert Monitor allows a management-by-exception strategy, so it is of special concern in this and other SCM business process scenarios. The Alert Monitor is a standalone component of SAP SCM that enables you to have a unified approach to handling problem situations. It notifies you if a problem occurs during an ATP check or SNP run, or when production plans, demand plans, or vehicle schedules are being generated in one of the SAP SCM applications.

The Alert Monitor is a tool with which planners can monitor the state of a plan. The monitoring results can be used to readjust the plan whenever necessary. The purpose of the Alert Monitor is to inform planners if a condition of a plan has been violated. The Alert Monitor belongs to the suite of supply chain monitoring components in SAP SCM, together with the Supply Chain Cockpit and the Plan Monitor. It can be used by any supply chain manager or planner who practices exception-based management in the following areas:

Demand Planning (DP)

Supply Network Planning (SNP)

Production Planning/Detailed Scheduling (PP/DS)

Available-to-Promise (ATP)

TLB/Deployment

Vehicle Scheduling (VS)

The Alert Monitor can be called as a standalone application via transaction code /SAPAPO/AMON1 as well as from various planning applications like /SAPAPO/SDP94, /SAPAPO/PPT1, and others. Alerts can also be displayed in the Supply Chain Cockpit /SAPAPO/SCC01. When displaying alerts in /SAPAPO/SDP94, for example, the display of alerts is restricted to those alerts which are assigned to the selected planning book.

The goal of Demand Planning is to create accurate forecasts and demand plans, and the role of the Alert Monitor is to notify you of exceptions that occur during the process. An exception is any situation that needs further adjustment. A sudden switch in trends, a new fad, or an unexpected change in the circumstances of a supply chain partner could all lead to exception situations. Most exceptions, if not handled immediately, can have consequences up or down the supply line.

The Alert Monitor is an online tool designed to help you catch real-time exceptions before they turn into problems and, thus, to help you to forecast your business needs more effectively. Not only can you determine which types of exceptions you want to be notified of, you can also prioritize alerts, thus preventing an information overload. In an exception-based Demand Planning scenario, you as the planner must make the ultimate decision regarding the priority of an alert. If you regularly compare your forecast figures with your actual figures, you will soon be able to determine whether a 15% deviation is serious or not. This knowledge would enable you to set priority variants for alerts so that, for example, you get a warning if the deviation is


© 2012 SAP AG page 32/49

10% but an error alert if it is 15%. The real values in your business process can vary very much from the values presented in this example.

Alert priorities are identified by icons displayed in the profile or in the monitoring slots of the Supply Chain Cockpit control panel. The three possible priority levels in the Alert Monitor are:

1. Error 2. Warning 3. Information

The priorities should be defined in a way that (under normal conditions on a well-run system) information alerts take the biggest share and errors are generated only in rare cases. However, even an information alert should reflect an exception to the planning situation and not a “success” message saying “everything ok.”

Whereas in the DP application you can view only one planning book at a time, you can view several in the Alert Monitor, making it the ideal tool for tracking the quality of your alerts over a specific period of time.

In the Demand Planning process, the following exception situations may arise:

Changes in bias, new trends, unexpected demands

Orders exceed forecast

Orders fall short of forecast, which may lead to excess inventory

For Demand Planning, forecast alerts and Supply and Demand Planning alerts are important. Alert types of both of these areas can be used in the DP and SNP applications.

The Alert Monitor comes equipped with two forecast alert types that correspond to the exception situations that can arise in forecasting and Demand Planning, namely univariate forecasting alerts (based on the six forecast errors like mean absolute deviation and mean square error) and multiple linear regression alerts (based on the five MLR measures of fit).

You need to maintain alert profiles to maintain a user-specific selection of alerts for your area of responsibility. There are forecast alert profiles and Supply and Demand Planning alert profiles.

The Supply and Demand Planning alert profile contains a number of alerts that are specific to the tactical planning and sourcing decisions relevant for purchasing, manufacturing, and distribution. There are two different types of Supply and Demand Planning alerts:

Dynamic alerts mirror the current planning situation but are not saved in the database. Alerts are macro-dependent and thus can reflect the actual data in liveCache. Because large numbers slow down performance, this alert type is not suitable for dealing with a large volume of planning objects. For small data volumes up to approximately 1000 alerts, dynamic alert generation is usually faster than reading alerts from the database. For large data volumes, database alerts are usually faster.

Database alerts show the planning situation as it was during the planning run or the last executed macro. When dealing with large volumes, it is best to perform a background planning run using the appropriate database macro. The results of the planning run show the situation as it was at the time of the run, so with database alerts you see a snapshot of the plan during run time.

You also have the option of creating customer-specific dynamic or database alert types to be used with Supply and Demand Planning macros.

Database alerts that are no longer in use (for example, because they have been acknowledged or because they are outdated) must be deleted from the database. Among forecast and Supply and Demand Planning database alerts, this applies to ATP, VS, and TLB/Deployment alerts. However, PP/DS alerts are dynamic only and are not stored in the database.

Generally, if an alert is generated, three activities must take place:

1. Analyze alert situation.

2. Inform responsible planner.

3. Make adjustment to plan and/or profile.

To support the Alert Monitor's function as a tool for exception-based Demand Planning, you can send messages via e-mail to other planners to inform them of the alert situation. You can also have messages sent automatically to your own inbox to inform you of alerts in your area.


© 2012 SAP AG page 33/49

You can maintain a list of favorite Alert Monitor profiles so you can switch back and forth easily between various profiles. For example, you may have an alert profile for your own area of responsibility, but you may also want to look at alerts in other areas.

Keep alerts to a minimum. Too many alerts slow performance and may cause you to overlook the really important ones.

You can find more information concerning this tool in the SAP Documentation under "Supply Chain Monitoring – Alert Monitor." See also SAP Note 375965 and chapter Exception-Based Monitoring of Demand Planning.

2.3.1.2 Monitoring of SAP NetWeaver Business Warehouse Activities This section points out the concept of monitoring activities related to data transfers and the maintenance of data-keeping structures in the SAP NetWeaver BW component of an SAP SCM system. The data transfer can take place inbound or outbound the SAP NetWeaver BW (APO) system to and from other SAP or third-party components, as well as within one SCM system between different data storage units like InfoCubes and liveCache. When adapting this concept for your company, you must specify the times, responsible teams, and escalation paths (responsible teams) for the following monitoring activities and objects.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in the following table in order to safeguard any business step that is based on SAP NetWeaver BW techniques. See also the best-practice documents dealing with the data transfer to an SAP NetWeaver BW system.

The data transfer process from a source system to SAP NetWeaver BW InfoCube(s), as well as certain actions in RSA1 and RSMO, writes detailed information to a central SAP ERP Basis component called the application log. It can be browsed using transaction SLG1. The log entries are subdivided into objects and sub-objects and classified by importance. In case of errors or problems, the application log can be used to get more detailed information about what happened and how to resolve the problem. The monitor transaction for the application log is SLG1, where you should specify the (sub) objects and the time interval you want to be displayed. Relevant objects for the SAP NetWeaver Business Warehouse are:

RSAR BW Metadata maintenance

RSAU Update rules

RSD BW data basis

RSDMD Master data maintenance

RSO_REPOSITORY BW Repository

RSSM Scheduler; Monitor; Tree callback

Jobs for Running and Monitoring Data Transfers to or Within SAP Business Warehouse

To ensure a timely and consistent data transfer from any data source into an SAP NetWeaver BW structure, schedule the following job on a regular basis:

SAP SCM 4.1: Data Extraction and Load with report RSBATCH1. Define the job using the scheduler in the Administrator Workbench (transaction RSA1) of the BW component of APO. The scheduler generates a background job with name prefix BI_BTCH and processes the selected InfoPackage. With the help of InfoPackage groups, you can collate data requests that logically belong together and schedule them using the scheduler.

As of SAP SCM 5.0: Data Extraction and Load as a combination of single steps of a process chain. The scheduler in the Administrator Workbench (transaction RSA1) of the BW component of APO would still generate a background job with the name prefix BI_BTCH and process the selected InfoPackage, but this will only result in loading the data to the PSA. The further processing to the data targets has to be triggered manually by scheduling the appropriate data transfer process (DTP) in this case.

In Monitoring of the Administrator Workbench (select Monitoring in transaction RSA1 or call transaction RSMO), you have the option to oversee and control the data loading process into the InfoCubes. To analyze any errors that may have occurred, choose Monitor Assistant. You can also call the monitor from a



© 2012 SAP AG page 34/49

selected InfoPackage or via transaction code RSMO. To make sure that you have selected the correct time window and data to be displayed, call the selection screen: choose Monitor New Selections.

Jobs for the Maintenance of the Application Log

The application log entries are stored in tables with name prefix BAL*. As there are many applications that use this Basis component and often many table entries are made, it is important to delete obsolete application logs from the database regularly to prevent these tables from overflowing or impacting the performance of the applications and the log retrieval. The deletion can be performed using transaction SLG2 or in the background with the appropriate report:

Delete Obsolete Application Logs with report SBAL_DELETE. A log can only be deleted when it has reached its expiry date or if it has the attribute "Deletion before expiry." For more information, see SAP Note 195157.

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO job BI_BTCH*

These jobs perform the data transfer from the different data sources to APO InfoCubes.

RSMO or SM37

Depending on your process





APO: Administrator Workbench Monitor

RSMO After every data load

Red or yellow traffic lights

After selecting one request, read the error message and diagnosis under tab strip Status and the processing log under Detail.




APO: Process Chain Display Log View

RSPC After every data load

Red or yellow steps in the chain

Choose Process Monitor from the step’s context menu.

Read the error message and diagnosis under tab strip Status and the processing log under Detail.




APO: Evaluate Application Log for BW related objects

SLG1 At least daily

Log class Check if there are very important logs (flagged red) or important logs (flagged yellow).

Read message long text and/ or details, if present. Analyze error situation, evaluate error severity and impact, and take corrective action.





© 2012 SAP AG page 35/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report SBAL_DELETE

This report deletes obsolete application logs.


If the job is not scheduled as provided by application support, schedule it to run weekly.



Error Handling, Restartability, and Escalation

Error Handling Procedures

Error handling for background jobs is explained in detail in the SAP ERP documentation CD, component BC-CCM, under Background Processing.

If a scheduled job fails, a necessary job is not scheduled, or a scheduled job has status finished, you may need to take action. Consider the status of the job and proceed as follows:

In the case of status scheduled, the job steps have already been defined, but the start condition has not yet been defined. Contact the program scheduling management and clarify when the job will be fully defined.

In the case of status released, the job has been fully defined with a start condition and will wait for that condition to be fulfilled.

In the case of status ready, the start condition of a released job has been fulfilled. A job scheduler has put the job in a queue to wait for an available background work process.

In the case of status active, the job is currently running and can no longer be modified or deleted. Check if the job is within the given time frame. Check for particular dependencies to other jobs. If the job exceeded the given time frame, contact the software monitoring team.

In the case of status finished, all steps that make up this job have completed successfully. Program Scheduling Management must check whether the job ran in the given time frame, and the software monitoring team and/or application support must check the respective job results (such as spool output lists, message logs, and updates).

In the case of status canceled, the job has terminated abnormally. This can happen in two ways. If an administrator intentionally canceled the job, clarify why he or she did so and whether (and if so, when) the job must be re-run. Alternatively, if a program in a job step produced an error such as issuing an "E" or "A" error message, contact the software monitoring team and investigate why the error occurred. If the program is an SAP standard program, search for appropriate SAP Notes on the SAP Notes home page. If you cannot solve the problem, create a support notification on the SAP Service Marketplace.

If there are problems with the Core Interface or with data missing in either SAP ERP or SAP APO, see the Troubleshooting Guide Integration R/3 – SCM in the Literature Center of the SAP R/3 plug-in homepage.

Process Step Restartability

If a background job is canceled and you are deciding whether to restart it, consider possible following jobs or dependencies on other jobs. Restarting the job may also delay the start of following jobs.

Escalation Procedures

In general, we recommend that you search for SAP Notes related to any unknown problems or errors through the SAP Notes home page.

If you have questions or problems that cannot be solved, forward the issue to the next support level. If the corresponding escalation path is not well defined, contact application support.

http://service.sap.com/notes

http://service.sap.com/~form/sapnet?_FRAME=OBJECT&_HIER_KEY=501100035870000009376&_HIER_KEY=601100035870000050275&_HIER_KEY=701100035871000068587&#HOME



© 2012 SAP AG page 36/49

If none of the defined support levels can provide a solution for a particular problem, we recommend that you create a support notification on the SAP Service Marketplace.

2.3.1.3 Maintenance of DP Data Storage Structures When adapting this concept for your company, you must specify the times, responsible teams, and escalation paths (responsible teams) for the following monitoring activities and objects.


The data for your DP process is stored in different logical and physical structures such as InfoCubes, the master planning object structure, and liveCache time series. All these need monitoring and maintenance to keep them in a technically optimal condition and allow fast retrieval of the data stored in them.

Jobs Necessary to Keep the APO Data Structures in Good Condition (APO)

To ensure that database statistics and indexes for BW-specific data structures are kept up to date, and that liveCache data structures (time series) are kept consistent, schedule the following jobs to run on a regular basis:

Compute Histograms for BW InfoCubes with the BRCONNECT tool as well as the “normal” database statistics. For details, see SAP Notes 428212 and 535986.

Check Inconsistencies in Time Series Network with report /SAPAPO/TS_LCM_CONS_CHECK_ALL. This report checks, in all existing time series (in all planning areas and versions) and reports, whether or not there are inconsistencies. If errors are found, you can repair inconsistencies in a specified planning area and version using report /SAPAPO/TS_LCM_CONS_CHECK afterward. Mark the Log checkbox if you want to see a list of characteristic combinations that have inconsistencies. For more information, see SAP Notes 425825 and 520876.

Delete Time Series and Time Bucket Profiles having no liveCache Anchors with report /SAPAPO/TS_LCM_REORG. This report determines all liveCache time series and time buckets profiles from a selected planning version that do not have any liveCache anchors, independent from planning areas. These objects do not refer to the application data so they are not inconsistent, but they are unnecessary and consume liveCache memory. It may be sufficient to run the report approximately once a month. If you have selected the Repair option, any such unnecessary objects are deleted. Caution: When you run report /SAPAPO/TS_LCM_REORG, no other processes for a planning area with key figures based on liveCache time series should be allowed to run with the same planning version. Otherwise, liveCache objects that are still required by the other parallel process may be recognized as unnecessary and even deleted. For this reason, we recommend that you always run the report with the option Lock planning version. In this case, all planning areas that use the planning version to be checked are locked for the relevant version. If you cannot activate the lock, report execution is interrupted and a list of system users is displayed, showing users who have already set a change lock for the planning version.

Maintain Statistics and Indexes for Planning Object Structures with report /SAPAPO/TS_PSTRU_TOOL. Run this report in separate background jobs choosing options Calculate Statistics and Check Indexes for every master planning object structure and review the list of messages being displayed. If problems are reported by the index check, run it with option Repair Indexes. Then a separate background job (job name with prefix BI_STAT) is started to execute this request. These administrative jobs must be run after every load of data to a planning object structure. For further details, see SAP Note 503363.







© 2012 SAP AG page 37/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


Indexes of APO InfoCubes

RSA1 or RSRV

Weekly Red or yellow traffic light

Select Check (Aggregate) Indexes. If the traffic light is yellow or red, select Repair Indexes (now) or Delete Index (batch) and Create Index (batch).



APO report /SAPAPO/TS_LCM_CONS_CHECK_ALL

This report detects inconsistencies in time series.


If the job is not scheduled on a regular basis, schedule it to run once a week.



Output of APO report /SAPAPO/TS_LCM_CONS_CHECK_ALL

SP01 After each run

Red traffic light

Check for time series with inconsistencies reported.

Repair inconsistencies by running report /SAPAPO/TS_LCM_CONS_CHECK for the corrupt time series, setting the repair option.



APO report /SAPAPO/TS_LCM_CONS_CHECK

This report can repair inconsistencies in time series.

SM37 After each run

Status Check if job has finished without errors.



Output of APO report /SAPAPO/TS_LCM_CONS_CHECK

SP01 After each run

Traffic light Check whether the inconsistencies in the selected time series have been repaired.



APO report /SAPAPO/TS_LCM_REORG with option Repair.

This report deletes superfluous time series and time bucket profiles.

SM37 Monthly Status Check if job is running as scheduled.

If the job is not scheduled on a regular basis, schedule it to run once a month.



Output of APO report /SAPAPO/TS_LCM_REORG

SP01 After each run

Traffic light Check whether superfluous data is reported and has been deleted without errors.




© 2012 SAP AG page 38/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/TS_PSTRU_TOOL with option Calculate Statistics

This report checks and generates statistics for a POS.

SM37 After every data load into the POS





APO report /SAPAPO/TS_PSTRU_TOOL with option Check Indexes

This report checks the indices of a POS.

SM37 After every data load into the POS





Output of APO report /SAPAPO/TS_PSTRU_TOOL


Message type (icon)




Job BI_STAT* started by /SAPAPO/TS_PSTRU_TOOL with option Repair Indexes


Status Check if job ended correctly. Program Scheduling Management



2.3.1.4 Running Demand Planning in Background This section outlines the monitoring concept for DP activities in the background for processing mass data. Mass processing allows you to create demand forecasts or run macros for large numbers of products while optimizing system resources. Also, a release of DP data to SNP or a transfer to ERP is usually performed in the background. The different tasks that can be performed in this way need to be customized and set up as “activities” prior to job definition.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in the following table in order to safeguard any business process step that performs DP activities in the background.

Jobs for Running and Monitoring "Demand Planning in Background" (APO)

To ensure correct and timely Demand Planning, schedule the following jobs to run on a regular basis:

Run Demand Planning in the Background with report /SAPAPO/TS_BATCH_RUN. This mass processing job is created with its necessary settings and scheduled using transactions /SAPAPO/MC8x. To create a job, call /SAPAPO/MC8D. To change a job, call /SAPAPO/MC8E. To define in advance the activities that are needed for these jobs, call /SAPAPO/MC8T (you can do this from the customizing tree). An activity can be a forecast computation, execution of one or several


© 2012 SAP AG page 39/49

macros, a release of DP data to SNP orders, or the transfer to ERP. To check the job definition prior to execution and find errors like “no macro defined,” “data view does not exist,” or “release profile not found,” call /SAPAPO/MC8I. To schedule the job, call /SAPAPO/MC8G. To see a job overview, display job logs, and jump to spool lists, call standard transaction SM37. To check the results of a DP mass-processing job, use the DP job log. To request a job log, select Generate log when you create a job. To view the job log, call /SAPAPO/MC8K. The job log shows whether the job completed successfully (green traffic light), with warnings (yellow light), or with errors (red light), together with a message for every characteristic value combination that was processed in the job, forecast error messages if the job included a forecast, and other details. You can also delete job logs with this transaction. To copy jobs, call /SAPAPO/MC8J. To delete jobs, you can call /SAPAPO/MC8F. Parallel processing: Given the large amount of data to be processed and the limited time available, you may need to process background runs (macros or any other background jobs) in parallel. You can do this by defining several jobs in /SAPAPO/MC8x running the same activity but having a different selection of characteristics. Each selection ID should define a set of characteristic value combinations of approximately equal size, and no individual characteristic combination should belong to more than one selection ID, as this can cause lock issues and you do not need to process a combination more than once. These jobs can then be scheduled to run at the same time. Remember that hardware resources, like the number of CPUs and their speed, the amount of main memory, and so on, limit the total number of jobs that can run on a system at the same time. You can also use parallel processing to run macros or to perform any other planning activity, such as to perform a statistical forecast. You can navigate from transaction /SAPAPO/MC8E directly to the maintenance of parallel processing profiles to be used for a background planning job or use /SAPAPO/SDP_PAR for direct access. After defining the block size and the number of maximum parallel processes, you have to assign the parallel processing profile to the selection variant. The next time you run this selection as a background job, it will automatically run with parallel processing.

Delete DP Job Logs with report /SAPAPO/TS_BATCH_LOGFILE. Old logs from planning activities must be deleted regularly to maintain high performance for the related transactions and prevent database tables from overflowing. Make sure that you delete the logs in packages. Select the option Delete all logs with minimum age, reducing the parameter days. See also SAP Note 512184.

Demand Planning activities can be executed periodically with very different frequencies. For example, key figures can be computed with background macros daily or weekly, whereas a release of the demand plan to Production Planning or ERP is usually performed at longer intervals, for example once a quarter or once a year. Accordingly, the frequency of monitoring the jobs and deleting the DP job logs depends on how often you run these functions.

For dependencies and concurrent execution of jobs, see Operational Management – Parallel and Concurrent Execution of Jobs. See also Exception-Based Monitoring of Demand Planning for the handling of alerts generated by a forecast run or a batch macro.

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/TS_BATCH_RUN

This report performs specified DP mass-processing activities.








© 2012 SAP AG page 40/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


DP Job Log /SAPAPO/MC8K

After each run of /SAPAPO/TS_BATCH_RUN

Red or yellow traffic lights shown

According to the warning or error reported (see message long text)

Application support


APO report /SAPAPO/TS_BATCH_LOGFILE

This report deletes old DP job logs.







2.3.1.5 Exception-Based Monitoring of Demand Planning This section outlines the concept of monitoring the overall planning situation in Demand Planning. The most important high-level tool for monitoring the planning situation and an exception-driven application management is the Alert Monitor. The usage of the Alert Monitor is supported by an e-mail interface that allows notifying the responsible persons in case of new alerts. Therefore, it is not necessary for planners and/or IT staff to call the monitor transaction frequently just to confirm that nothing important has happened since the last refresh. However, the overall situation shown in the Alert Monitor should be supervised daily or adapted to the frequency of the Demand Planning process in your company.


Apart from safeguarding the general availability and consistency of the system components SAP APO, SAP NetWeaver BW, and SAP ERP, SAP recommends that you monitor the objects listed in the following table in order to safeguard your DP business process.

Jobs for an Exception-Based Monitoring of Demand Planning

To ensure a timely and efficient notification about exceptions in DP, schedule the following jobs to run on a regular basis:

Send Alert Monitor mails with report /SAPAPO/AMON_MAIL_BROADCAST. Depending on user profiles, mails are sent with an overview of existing alerts. The responsible persons then should call the Alert Monitor, investigate the reasons for any alerts, and take corrective actions to keep the demand plan close to the needs of your company.

Delete Alert Monitor alerts with report /SAPAPO/AMON_REORG. This report is of particular importance if you use alerts stored in the database. On database alerts and dynamic alerts in Demand Planning, see Alert Monitor.

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


Alert Monitor /SAPAPO/AMON1

At least daily

Alert list Check for forecast alerts and Supply and Demand Planning alerts and correct the planning for the reported object appropriately.

Application support



© 2012 SAP AG page 41/49

Monitoring Object

Monitor TA/Tool

Monitor Freq.

Monitor Time

Indicator or Error


Respon-sibility


APO report /SAPAPO/AMON_MAIL_BROADCAST

This report ensures sending of mails about existing alerts.

SM37 Daily Status Check if job is running as scheduled.

If the report is not scheduled on a regular basis, schedule it to run every hour to at least daily, depending on your requirements.



Alert Monitor mails

SO01 (or respective e-mail system)

Depending on your requirements, at least daily.

Check if the mail lists alerts that are important for you. Go to Alert Monitor and process the alerts appropriately.

Application support


APO report /SAPAPO/AMON_REORG

This report deletes Alert Monitor alerts.


If the job is not scheduled on a regular basis, schedule it to run once a day.





© 2012 SAP AG page 42/49

3 Further Information

3.1 Troubleshooting

3.1.1.1 Dependencies Remember that there are dependencies (date and time, logical sequence) to business processes and process steps that are not mentioned in this document. These usually comprise, for example:

General SAP ERP system administration (this also applies to the basis of the SAP SCM system), for example:

o Reorganization of jobs, spool entries, and so on

o DB offline backup – During an offline database backup, no online or background activity is possible. Therefore, times for such backups must be scheduled carefully.

o Archiving of DB transaction logs

o Updating table statistics for the DB cost-based optimizer – You should not run this activity at times when application programs are likely to create, delete, or update many table entries.

General SCM-specific system administration:

o Checking the internal (liveCache-APO DB) and external (SCM-ERP) data consistency

o Backup for liveCache

o Reorganization of Component Object Model (COM) objects and optimizer application logs with report /SAPAPO/OM_REORG_DAILY)

Transfer of master data from ERP to SCM:

o Initial transfer of master data records

o Delta transfer of new master data records

o Transfer of changes made to existing master data records – You should not transfer large packages of master data to SCM when CIF is needed for the transfer of transactional data, because this can overload CIF and cause an undesirable delay in communication.

Because of these dependencies, online and background application-system activity cannot always occur whenever desired, but may need to wait, for example, for the completion of administration activity.

Therefore, Program Scheduling Management and the software monitoring group should plan and schedule system maintenance activities to run at appropriate times, for example, overnight or over a weekend. Then, all the work necessary for the company’s core business process can be performed within the time frame(s) determined by the business process champions. Also, certain activities – such as background jobs – should be started only after the respective preceding activity has finished.

3.1.1.2 Troubleshooting If executing this best practice does not produce the desired results, proceed as follows:

If there are or seem to be problems in the data transfer via the Core Interface, see the Troubleshooting Guide for Integration APO (CIF), which you can find in the SAP Service Marketplace in the Literature Center of the R/3 plug-in home page.

Search for related SAP Notes.

Open an SAP customer message describing your problem.

3.1.1.3 Operational Management – Parallel and Concurrent Execution of Jobs The combination of InfoCube lock issues and system resource usage makes concurrent execution of some DP job types problematic.

Recommendation: Observe the following rules for scheduling or executing DP transactions and jobs:

1) Calculate new statistics for all InfoCubes and planning object structures defined on the system regularly (at least once a week) using the BRCONNECT tool.

http://service.sap.com/~form/sapnet?_FRAME=OBJECT&_HIER_KEY=501100035870000009376&_HIER_KEY=601100035870000050275&_HIER_KEY=701100035871000068587&#HOME

http://service.sap.com/R3-Plug-In



© 2012 SAP AG page 43/49

2) Do not run batch jobs in DP that lock the same data (or in other words, the same characteristic combinations) concurrently. Specifically, do not run two jobs in parallel if they use selection variants that contain overlapping characteristic combinations, or else one job must wait until the other has released the lock on the data, resulting in performance problems. You can activate the key-figure-specific locking logic in customizing, if you maintain this in the Supply and Demand Planning administration (transaction /SAPAPO/MSDP_ADMIN): Switch to Planning Areas, then double-click on your planning area on the last tab, Locking Logic. Switch on Detailed Lock and press Enter. Then activate Key-Figure-Specific Lock and, if desired, Do not lock read-only key figures. After saving the new settings, you can perform parallel jobs with overlapping data selections (for example, by selecting all data in one variant) if the key figures used during batch job processing are different (no overlaps) in both background jobs.

3) Mass processing background jobs performing forecasts, macros, release from DP to SNP, and transfer to ERP may be run in parallel, or concurrently with one another, provided that they obey rule 2. Ideally, these jobs should be of approximately the same size so that each job takes the same amount of time to run. This is most easily accomplished by designing selection variants with the same number of characteristic combinations.

4) Loading data from an InfoCube into the liveCache may be executed in parallel by defining several different variants that obey rule 2 if you want to run them at the same time. Ideally, these jobs should be of approximately the same size, so that each job takes the same amount of time to run. This is most easily accomplished by designing selection variants with the same number of characteristic combinations and careful observation that there are no overlaps between the individual selections.

5) As of SAP SCM 5.0, you can activate the new liveCache lock logic to reduce the maintenance efforts for selections and to solve possible locking problems in situations where users and/or programs try to access the same product data at the same time. You can switch on the new optional LiveCache Locks if you maintain this in the Supply and Demand Planning administration (transaction /SAPAPO/MSDP_ADMIN). You can follow the detailed steps outlined in rule 2), the steps for maintaining this new functionality are the same. After the activation of the new liveCache lock logic, you can use wildcards or even define product ranges in the selections used for interactive planning or background jobs. This means you don’t have to frequently maintain detailed “single item” selections anymore. Flexible ranges can cover new products or remove phased-out products automatically. The new liveCache lock logic is more specific and reduces the danger of locking problems during parallel data access.

6) Two users should not run selection variants in DP interactive planning that lock the same datasets (or in other words, the same characteristic combinations); otherwise one user must wait until the other has released the lock on the data which has an negative impact his/her work: The user receives an information that he or she cannot modify the data now and has to try again some time later. You can activate the key-figure-specific locking logic as described in rule 2). If your data view design allows the design of data views with completely different key figures (no overlaps), then the locking problem does not occur. This can be a solution for, for example, promotion planners and baseline planners that have to access the same characteristic combinations in parallel.

7) Execute initial data load jobs, calculation of proportional factors, planning versions or key figures copy jobs, and all mass processing jobs at times of low system load. Typically these activities are performed during the night because of the large data volume processed. We recommend running these activities as background jobs.

8) Plan parallel execution to avoid CPU bottlenecks on the application server. Optimal performance is achieved when one CPU is available for each process. How critical this is depends upon how much of the response time is CPU time. For optimal performance, only run as many jobs as there are CPUs in the system, and less if there are memory limitations. More jobs will run more slowly, depending on the CPU resources needed by the job type. Some job types, such as forecasts, are heavily dependent on the application CPU resources, while others, such as the release from DP to SNP, require less.


© 2012 SAP AG page 44/49

9) Plan parallel execution to avoid memory bottlenecks. To avoid this, perform a trial run of the process to be parallelized and collect the statistical records for the run with full details.

Find out how much total memory your system has and call this M_total.

From the statistical records, find out how much memory was used by the individual process and

call this M_proc.

The maximum number of parallel processes that should run at once is:

N = (M_total) / (M_proc).

10) Do not perform initial data load concurrently with any of the following jobs:

a) Copy key figures

b) Calculate proportional factors

c) Release from DP to SNP

d) Any macro or forecast

11) Do not execute the release from DP to SNP concurrently with any of the following job types:

a) Data import into DP in the same planning area

b) SNP heuristic planning run

c) Daily liveCache reorganization

d) Any DP transaction or mass processing job that locks (uses) the same data as the release

12) Some extract jobs that read data from liveCache (such as copy data from liveCache to an InfoCube for backup) may require a different setting for the number of periods per block to avoid memory problems. If this is the case, execute the following sequence:

a) Execute job /SAPAPO/TS_PAREA_EXTR_SETTING for planning area of interest and set the number of periods necessary for the data extraction job.

b) Run the data extraction job.

c) Execute report /SAPAPO/TS_PAREA_EXTR_SETTING for the appropriate planning area again and reset the number of periods necessary for the data extraction job to 0.

13) If background jobs that generate macros are run at high volume, delete old job logs regularly (for example, once a week, if possible) using transaction /SAPAPO/MC8K.

14) If alerts are generated as part of the business process, regularly delete alerts (for example, once a week, if possible) using transaction /SAPAPO/AMON1 or report /SAPAPO/AMON_REORG.

Enterprise Services Since SCM 5.1, Web services are available that make it possible to execute most Demand Planning functionality via these services. For a detailed list of the services that are included in the Demand Planning enterprise service bundle please visit this link: DP ES Bundle.

3.2 Background Information and References

3.2.1.1 Literature For more detailed information about how to administer an SAP NetWeaver system, see:

Frank Föse, Sigrid Hagemann, Liane Will, SAP NetWeaver AS ABAP – System Administration, 2008

For information about the administration of SAP APO systems, see:

Liane Will, SAP APO System Administration, 2002

For information about how to monitor and tune general system performance, see:

Thomas Schneider, SAP Performance Optimization Guide, 2008

For background information on administrative tasks with emphasis on system planning and setup, see:

Hartwig Brand, SAP R/3 Implementation with ASAP, 1999

For more information about these and other books, visit http://service.sap.com/books.

https://www.sdn.sap.com/irj/sdn/wiki?path=/display/ESpackages/Demand+Planning

http://service.sap.com/books


© 2012 SAP AG page 45/49

3.2.1.2 Other Best Practice Documents

In the SAP Service Marketplace, quick link /enterprisesupport → SAP Enterprise Support Academy → Best Practices, you can find several Best Practice documents for solution management, like this one:

Monitoring and Administration for SCM / APO helps you analyze the workload and performance on liveCache and the APO database.

Manage APO Core Interface in SCM deals with the Business Process Management of the APO CIF and is an essential enhancement to this document. All the jobs and monitoring activities listed in the CIF document have to be considered in every business process step listed above that sends or receives data through CIF.

Manage Production Planning in SCM / APO deals with Production Planning and Detailed Scheduling in APO and production execution in SAP ERP, including the MRP run mentioned in Business Process Step 15: Perform Material Requirements Planning MRP (ERP).

Under the alias solutionmanagerbp SAP.com Solution SAP Business Intelligence, you can find the best-practice documents SAP NetWeaver BW Master Data Upload and SAP NetWeaver BW Data Target Upload. These documents can help you in setting up and monitoring the data transfer to an SAP NetWeaver BW system that is connected to a SAP SCM system.

3.2.1.3 SAP Documentation SAP SCM 7.0 documentation is available on CD or in the SAP Help Portal in German or English.

SAP SCM 5.1 (2007) documentation is available on CD or in the SAP Help Portal in German or English.

SAP SCM 5.0 documentation is available on CD or in the SAP Help Portal in German or English.

SAP SCM 4.1 documentation is available on CD or in the SAP Help Portal in German or English.

Print files (PDF format) of several chapters in both languages are available in the Media Center of the SAP Service Marketplace for SCM.

3.2.1.4 SAP Notes See also http://service.sap.com/notes and http://service.sap.com/scm mySAP SCM Technology.

See the best-practice document dedicated to CIF for SAP Notes with information on qRFC and CIF.

The following SAP Notes contain information about the SAP NetWeaver Business Warehouse and the database underlying an SAP APO system:

129252: Oracle DB Statistics for BW Tables

130253: Notes on upload of transaction data into the BW

830576: Parameter recommendations for Oracle 10g

323090: Performance problems due to degenerated indexes

325839: Considerable increase of tablespace PSAPODSD

378509: Oracle DB parameter setting for APO

400191: Further processing of data from the PSA

409641: Examples of packet size dependency on ROIDOCPRMS

428212: Update of statistics of InfoCubes with BRCONNECT

558563: How does a client copy work with Demand Planning?

The following SAP Notes contain information about the APO Demand Planning module:

897887: Performance improvement life cycle planning

892940: MC62 and generation of CVCs: performance improvement

890623: Performance improvement with Lifecycle planning

880012: Makros: Performance-Verbesserungen

http://service.sap.com/enterprisesupport

http://service.sap.com/solutionmanagerbp

http://help.sap.com/

http://help.sap.com/saphelp_scm70/helpdata/de/12/3dfd81126e1c448c870f335f11d016/frameset.htm

http://help.sap.com/saphelp_scm70/helpdata/en/12/3dfd81126e1c448c870f335f11d016/frameset.htm


http://help.sap.com/saphelp_scm2007/helpdata/de/e1/8e51341a06084de10000009b38f83b/frameset.htm

http://help.sap.com/saphelp_scm2007/helpdata/en/e1/8e51341a06084de10000009b38f83b/frameset.htm







http://service.sap.com/scm



















© 2012 SAP AG page 46/49

848109: /SAPAPO/TS_PAREA_DE_INITIALIZE: Performance

832393: Release Restrictions for SCM 5.0

394076: Consulting: USER-EXITS and BAdIs in the forecast

350065: Consultation: User parameters in forecast

333243: Analysis of faulty definition and execution of macros

350381: Promotion: Report for updating active promotions

359761: Demand Planning: loading performance data

372939: APO consulting: Compare forecast in interactive &….

373756: Data extraction from a planning area

386735: DP: Extract data to an IC with delta update

388260: APO Consulting forecast: Automatic model selection

412429: Definition of jobs with macros

420927: Data extraction of selected key figures

425825: Consistency checks, /sapapo/om17, /sapapo/cif_deltareport

426806: Memory/performance problems during data extraction

428102: Performance: Loading planning area version

454644: Consultancy: Lock logic in promotion planning

482494: Loading data from liveCache: Performance optimization

495027: Changing delivered APO InfoObjects (9A*)

503363: Use & management of fixed aggregates in Demand Planning

506393: Conversion exits when creating characteristics combinations

512184: Background processing: Periodically delete job log

515120: Performance of extraction of InfoCube-based key figures

515523: Unexpected numerical results in Demand Planning

520876: Inconsistencies in time series objects

539797: Collective consulting note on macros

539848: Collective consulting note on background processing in DP

540241: Collective consulting note on CBF

540282: Collective consulting note on promotion planning

540571: Collective consulting note on data extraction in DP

540926: Collective consulting note on planning object structures

541189: Collective consulting note on selections in Demand Planning

541252: Collective consulting note for planning book maintenance

541618: Collective consulting note on BW and InfoObjects in DP

541633: Collective consulting note on interactive planning

541703: Collective consulting note on technical subjects in DP

546079: FAQ: Background jobs in Demand Planning

549184: FAQ: What is important for extraction

558563: How does a client copy work with Demand Planning?

558995: Advice on consistency report for forecast profiles

568506: Collective consulting note on consumption of forecasts

568671: Collective consulting note on versions

568669: Collective consulting note on release DP – SNP

570397: Consulting: Workaround - Copying Planning Object Structures














































© 2012 SAP AG page 47/49

571629: How does the note management work?

573127: Creating several char. combinations: /SAPAPO/MC62

576015: Collective Consulting Note for Demand Planning

See also the following SAP Notes:

016083: Standard jobs, reorganization jobs

195157: Application log: Deletion of logs

375965: APO Consulting: Alerts in forecast

425825: Consistency checks, /sapapo/om17, /sapapo/cif_deltareport

495166: Tips and Tricks for Handling Alert Monitor

500063: Overview of performance notes for the Alert Monitor

519014: Handling Planning Version Management

521639: Generation of DB Alerts in Background

564186: CLPSDP: Restrictions in APO Collaborative Planning

572003: SCM operating concept

830673: AMON: Some consulting issues when handling alert monitor

844456: AMON: Writing Alerts in background

884334: Alert monitor: Performance when you determine alerts

3.2.1.5 Feedback and Questions Send any feedback by formulating an SAP customer message to component SV-SMG-SER. You can do this at http://service.sap.com/message.

















http://service.sap.com/message


© 2012 SAP AG page 48/49

3.3 Index

/SAPAPO/AMON_MAIL_BROADCAST ....... 25, 40

/SAPAPO/AMON_REORG ..................... 25, 40, 44

/SAPAPO/AMON1 ............................ 25, 31, 40, 44

/SAPAPO/FCST_CHART ................................... 24

/SAPAPO/LCOUT ............................................... 30

/SAPAPO/MC62 ..................................... 14, 24, 47

/SAPAPO/MC77 ................................................. 25

/SAPAPO/MC8E ........................................... 18, 21

/SAPAPO/MC8T ................................................. 21

/SAPAPO/MC8V ................................................. 20

/SAPAPO/MC8x ......................... 21, 28, 29, 38, 44

/SAPAPO/MC90 ................................................. 30

/SAPAPO/MC96B ............................................... 23

/SAPAPO/MP34 ................................................. 26

/SAPAPO/MP39 ................................................. 26

/SAPAPO/MP42 ................................................. 27

/SAPAPO/MP46 ................................................. 27

/SAPAPO/MP47 ................................................. 27

/SAPAPO/MSDP_ADMIN ............................. 23, 43

/SAPAPO/OM_REORG_DAILY ......................... 42

/SAPAPO/PROMOTION_ACTIVATION ............. 27

/SAPAPO/PROMOTION_UPDATE .................... 27

/SAPAPO/RELEASE_SNP_FORECAST ........... 30

/SAPAPO/RLGCOPY ................................... 15, 17

/SAPAPO/RMDP_ICUBE_PERFORM ............... 13

/SAPAPO/RMDP_SHARE_MANAGER ............. 20

/SAPAPO/RTSCOPY ................................... 19, 30

/SAPAPO/RTSINPUT ......................................... 30

/SAPAPO/RTSINPUT_CUBE ....................... 18, 19

/SAPAPO/RTSOUTPUT ..................................... 30

/SAPAPO/RTSOUTPUT_FCST ......................... 29

/SAPAPO/SCC01 ............................................... 31

/SAPAPO/SDP_CHART_V ................................. 24

/SAPAPO/SDP_PAR .............................. 15, 18, 21

/SAPAPO/SDP94 ..........................................24, 26

/SAPAPO/SNPFCST ........................................... 30

/SAPAPO/TS_BATCH_LOGFILE .......... 22, 23, 39

/SAPAPO/TS_BATCH_RUN ............ 21, 28, 29, 38

/SAPAPO/TS_LCM_CONS_CHECK ............15, 36

/SAPAPO/TS_LCM_CONS_CHECK_ALL .......... 36

/SAPAPO/TS_LCM_PLOB_DELTA_SYNC ........ 15

/SAPAPO/TS_LCM_REORG .............................. 36

/SAPAPO/TS_PAREA_DE_INITIALIZE ........16, 46

/SAPAPO/TS_PAREA_EXTR_SETTING ............ 44

/SAPAPO/TS_PAREA_INITIALIZE ...............15, 17

/SAPAPO/TS_PLOB_MAINTAIN ........................ 14

/SAPAPO/TS_PSTRU_CONS_CHECK ............. 18

/SAPAPO/TS_PSTRU_TOOL .... 13, 15, 16, 17, 36

/SAPAPO/TS_REALIGN_COPY ...................15, 17

/SAPAPO/TS_REALIGNMENT ........................... 15

/SAPAPO/TSCOPY ............................................. 18

/SAPAPO/TSCUBE ............................................. 18

APODPDANT ...................................................... 20

BD87 ................................................................... 10

BRCONNECT ......................................... 13, 36, 45

DB13 ................................................................... 14

MRP .................................................................... 28

RSA1 ...................................................... 10, 14, 33

RSBATCH1 ...................................................11, 33

RSCUSTV6 ......................................................... 10

RSMO ..................................................... 10, 13, 33

RSPC .................................................................. 11

RSRV .................................................................. 13

SBAL_DELETE .............................................11, 34

SLG1 ................................................ 10, 12, 19, 34

SLG2 ................................................................... 11

SO01 .............................................................26, 41


© 2012 SAP AG page 49/49

© Copyright 2012 SAP AG. All rights reserved. No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP AG. The information contained herein may be changed without prior notice. Some software products marketed by SAP AG and its distributors contain proprietary software components of other software vendors. Microsoft, Windows, Excel, Outlook, and PowerPoint are registered trademarks of Microsoft Corporation. IBM, DB2, DB2 Universal Database, System i, System i5, System p, System p5, System x, System z, System z10, System z9, z10, z9, iSeries, pSeries, xSeries, zSeries, eServer, z/VM, z/OS, i5/OS, S/390, OS/390, OS/400, AS/400, S/390 Parallel Enterprise Server, PowerVM, Power Architecture, POWER6+, POWER6, POWER5+, POWER5, POWER, OpenPower, PowerPC, BatchPipes, BladeCenter, System Storage, GPFS, HACMP, RETAIN, DB2 Connect, RACF, Redbooks, OS/2, Parallel Sysplex, MVS/ESA, AIX, Intelligent Miner, WebSphere, Netfinity, Tivoli and Informix are trademarks or registered trademarks of IBM Corporation. Linux is the registered trademark of Linus Torvalds in the U.S. and other countries. Adobe, the Adobe logo, Acrobat, PostScript, and Reader are either trademarks or registered trademarks of Adobe Systems Incorporated in the United States and/or other countries. Oracle is a registered trademark of Oracle Corporation. UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group. Citrix, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame, and MultiWin are trademarks or registered trademarks of Citrix Systems, Inc. HTML, XML, XHTML and W3C are trademarks or registered trademarks of W3C®, World Wide Web Consortium, Massachusetts Institute of Technology. Java is a registered trademark of Sun Microsystems, Inc. JavaScript is a registered trademark of Sun Microsystems, Inc., used under license for technology invented and implemented by Netscape. SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign, Clear Enterprise, SAP BusinessObjects Explorer and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and other countries. Business Objects and the Business Objects logo, BusinessObjects, Crystal Reports, Crystal Decisions, Web Intelligence, Xcelsius, and other Business Objects products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP France in the United States and in other countries. All other product and service names mentioned are the trademarks of their respective companies. Data contained in this document serves informational purposes only. National product specifications may vary. The information in this document is proprietary to SAP. No part of this document may be reproduced, copied, or transmitted in any form or for any purpose without the express prior written permission of SAP AG. This document is a preliminary version and not subject to your license agreement or any other agreement with SAP. This document contains only intended strategies, developments, and functionalities of the SAP® product and is not intended to be binding upon SAP to any particular course of business, product strategy, and/or development. Please note that this document is subject to change and may be changed by SAP at any time without notice. SAP assumes no responsibility for errors or omissions in this document. SAP does not warrant the accuracy or completeness of the information, text, graphics, links, or other items contained within this material. This document is provided without a warranty of any kind, either express or implied, including but not limited to the implied warranties of merchantability, fitness for a particular purpose, or non-infringement. SAP shall have no liability for damages of any kind including without limitation direct, special, indirect, or consequential damages that may result from the use of these materials. This limitation shall not apply in cases of intent or gross negligence. The statutory liability for personal injury and defective products is not affected. SAP has no control over the information that you may access through the use of hot links contained in these materials and does not endorse your use of third-party Web pages nor provide any warranty whatsoever relating to third-party Web pages.

Date post:	27-Nov-2015
Category:	Documents
Upload:	jim-record-sr
View:	45 times
Download:	3 times