Skip Headers
Oracle® Business Intelligence Applications Fusion Edition Installation and Configuration Guide
Release 7.9.5

Part Number E12083-01
Go to Documentation Home
Home
Go to Book List
Book List
Go to Table of Contents
Contents
Go to Index
Index
Go to Feedback page
Contact Us

Go to previous page
Previous
Go to next page
Next
View PDF

H Using Oracle Business Analytics Warehouse Exception Reports

Note:

Exception Reports are only supported with Siebel CRM sources.

This chapter covers exception reports, which provide information about the source data used for ETL processes that can lead to erroneous results in the Oracle Business Analytics Warehouse or may cause data loss during the ETL process. These reports point out some of the known problematic areas, but they should not be relied upon to find all potential data inconsistencies in the source data.

This chapter includes the following topics:

H.1 Understanding Oracle Business Analytics Warehouse Exceptions

Exception reports are defined for the following components:

H.1.1 List of Values Exceptions

List of Values include High and Low values that can be used as bucket values in categories for effective analysis. If these values are not contiguous (such as gaps or overlaps in defined values), the ETL process cannot accurately categorize the values.

An example of List of Values exceptions is shown in Table H-1.

Table H-1 List of Values Exceptions (Example)

Type Name Low High

MY_TYPE

< 100

10

100

MY_TYPE

80 – 200

80

200

MY_TYPE

250 – 300

250

300

MY_TYPE

350 – 400

350

400


Notice that, in the example for Type MY_TYPE, there are overlaps and gaps between records. There is an overlap of ranges between the first and second row. There is a gap between second and third row, and between third and fourth rows.

The following LOV types are analyzed for List of Value Exceptions:

  • ACCNT_REVENUE_SIZE

  • ACCNT_EMP_SIZE

  • LEAD_AGE_DAYS

  • OPTY_REVENUE_SIZE

  • OPTY_UNIT_SIZE

  • ACCNT_REVENUE

  • QUOTE_AGE_DAYS

  • ACCNT_REVN_GROWTH

  • APPROVAL_AUTH_SIZE

  • SR_CHART_AGE

  • ASSET_COST_CATEGORY

H.1.2 Cost List Exceptions

Cost Lists for specified products and currency should not have overlapping time periods. If multiple cost lists are defined for a product and currency during a given time period, then the cost for the product may not be computed correctly in the Oracle Business Analytics Warehouse.

An example of Cost List exceptions is shown in Table H-2.

Table H-2 Cost List Exceptions (Example)

Cost List Product Name Currency Start Date (MM-DD-YYYY) End Date (MM-DD-YYYY) Cost

Cost List 1

Product 1

USD

01-01-2000

12-31-2000

10.00

Cost List 2

Product 1

USD

06-01-2000

06-01-2001

12.00

Cost List 3

Product 1

USD

06-01-2001

06-01-2002

13.00


In the example, Cost List 1 and 2 have definitions of cost overlapping over 06-01-2000 to 12-31-2000.

H.1.3 Products Without a Cost List

During the ETL process, the costs of the products are calculated based on the Cost List table. If the cost lists are not defined correctly, the cost of the products cannot be calculated correctly in the Oracle Business Analytics Warehouse. This exception mapping queries the product table and looks for a minimum of one cost list to be defined. The products with no cost list definition are flagged as exceptions.

H.1.4 Exchange Rate Exceptions

The Oracle Business Analytics Warehouse supports transactions in many different currencies. Oracle Business Intelligence converts all currencies in the Oracle Business Analytics Warehouse to a single currency for analysis purposes. The ETL Base Exchange Currency parameter in System Preferences, indicates the currency to which all the financial amounts will be converted. The Exchange rates are derived from the Exchange Rate tables in the Oracle Business Analytics Warehouse. If the currency exchange rates do not change for a period of 30 days, then Oracle Business Intelligence flags it as an exception.

If there are time period gaps in the exchange rate data, the ETL process defaults to the most recent recorded exchange rate. If the actual exchange rate is significantly more or less favorable than what is recorded in the database, the outdated exchange rate distorts the true value of currency amounts in the Oracle Business Analytics Warehouse.

Note:

Exchange rates are derived from records that are of type 'Daily' in the Oracle Business Analytics Warehouse. If any other types have been defined, they are not handled without some customization.

H.1.5 Invalid Hierarchy Exceptions

Accounts, divisions, products, and opportunities can all have hierarchical relationships. These entities are denormalized within the Oracle Business Analytics Warehouse database to a fixed number of levels. Oracle Business Intelligence supports up to ten hierarchies levels in the Oracle Business Analytics Warehouse. If the depth of hierarchies extends beyond this number, results become inconsistent or incomplete.

H.1.6 Circular Hierarchy Exceptions

Circular Hierarchies arise when the parent-child relationship has circular references. See Table H-3 for an example.

Table H-3 Circular Hierarchy Example 1

Child Parent

A1

A2

A2

A1


Oracle Business Intelligence flags exceptions for two levels. Circular references over two hierarchies are not flagged. See Table H-4 for an example.

Table H-4 Circular Hierarchy Example 2

Child Parent

A1

A2

A2

A3

A3

A1


These produce infinite levels of hierarchies. The same records will be captured under the Invalid Hierarchy exceptions as their hierarchy depths will increase beyond 10 levels.

H.2 Executing Oracle Business Analytics Warehouse Exception Reports

Before loading the Oracle Business Analytics Warehouse for the first time and for the subsequent refreshes, you should plan to spend time cleansing your transactional database data using the exception reports. The process is iterative, and requires coordination with other team members who have responsibility for data in the transactional database, such as the Siebel database administrator. After the initial cleansing, you should generate the exception reports on a scheduled basis to maintain the integrity of your data.

Note:

Rerunning the exception reports overwrites all data contained in this view.

To cleanse the Oracle Business Analytics Warehouse, repeat these actions until the exception report is empty:

H.2.1 Cleansing Data

Use a combination of the Exception Reports and the Diagnostic views to assess changes that need to be made external to the Oracle Business Analytics Warehouse, and changes to the Oracle Business Analytics Warehouse directly.

The ETL Exception Reports list contains one record for each exception flagged in the Oracle Business Analytics Warehouse. The ETL Exception Explanation form, located below the ETL Exception Reports list, describes the selected exception, its effect on the Oracle Business Analytics Warehouse building process, and offers suggestions for repairing the data.

To fix an exception

  1. Select an exception record.

  2. Read and understand the text in the ETL Exception Explanation form.

  3. Click the report link.

    The object's data appears. (For example, if the object is an account, then the Account form appears. If the object is a cost list, then the Cost List list appears.)

  4. Repair the problem, using the text in the ETL Exception Explanation form as a guide.

  5. Return to the ETL Exception Reports list and place a check mark in the Fixed column to indicate to others that this exception has now been fixed.

H.2.2 Using the List of Values View

Use the List of Values view, shown in the figure below, to visually compare how the list of values data extracted from the Oracle Business Analytics Warehouse coordinates with the values loaded into the Oracle Business Analytics Warehouse. The ETL process removes duplicates and overlaps and fills data gaps. Values are extended to span the List of Values (LOV) minimum and maximum values. Duplicates, Range Gaps, and Overlaps are flagged by the exception reports.

Figure H-1 List of Values View

This image is an example of the populated screen.

The top List of Values list shows values from the Oracle Business Analytics Warehouse and the bottom List of Values (Data Warehouse) list shows the data that is to be used in ETL process. You can edit the Oracle Business Analytics Warehouse data directly in this view, but the Oracle Business Analytics Warehouse list is read-only.

Note:

The List of Values is extracted into the Oracle Business Analytics Warehouse where the language is the same as the ETL Default Language set in the DAC Source System Parameters, or whose translate flag is set to 'N,' or those that are active. For more information about setting DAC Source System Parameters, see Section 4.11, "How to Install and Configure the DAC Server").

H.2.3 Using the Exchange Rates View

Use the Exchange Rates view to diagnose currency translation issues in the Oracle Business Analytics Warehouse. The ETL process removes duplicates, fills gaps, and removes overlaps. The ETL process computes exchange rates based on commutative and associative properties, such as product and reverse rates.

The top Exchange Rates list shows currencies, the middle Exchange Rates list shows the Oracle Business Analytics Warehouse values for active currencies and their exchange rates, and the bottom Exchange Rates (Data Warehouse) list shows the values loaded into the Oracle Business Analytics Warehouse for the selected currency in the upper Exchange Rates list to the ETL Base Exchange Currency. The Exchange Rates (Data Warehouse) list is read-only.

  • The Active Currencies predefined query restricts the list to the active currencies in the Oracle Business Analytics Warehouse.

  • The exception reports flag any exchange rates to the ETL Base Exchange Currency that have not been defined within a specified period (30 days) in the DAC Source System Parameters.

H.2.4 Using the Cost List View

Use the Cost List view to display the cost lists from the Oracle Business Analytics Warehouse from the point of view of the product, and a read-only view of the values to be loaded into the Oracle Business Analytics Warehouse. The ETL process removes duplicates, overlaps, and fills gaps.

The Cost List list (top) shows products, and the Cost List Line Items list (middle) shows the cost lists associated with the selected product. The Cost Lists (Data Warehouse) list (bottom) shows the data as it is transformed for the Oracle Business Analytics Warehouse.

  • The exception reports flag products that do not appear in the Cost List list or have Cost List time gaps and overlaps.

  • The Oracle Business Analytics Warehouse contains only one Cost List for a product and a currency at a time.

H.2.5 Using the ETL History View

After all of the exceptions are corrected, the building of the data warehouse can be initiated. This view lists the history of the ETL processes and their statuses. When each ETL batch starts, the name of the process along with the timestamp is set, the status is set to STARTED. When the batch completes, its status is updated to COMPLETED.

H.2.6 Additional Exceptions

Additional exceptions include:

  • The Analysis start and end date in the DAC Source System Parameters must span the entire period of time during which the transactions have occurred. For example, you may want to choose an early and late date range to cover the entire time period you are analyzing. These dates in the DAC Source System Parameters are crucial for the building of Day Dimension, flattening of Exchange Rates, Cost Lists, and KPI (Key Performance Indicator fact) calculations.

  • The DAC Source System Parameters — ETL Date Format, ETL Analysis Start, ETL Analysis End parameters, and the List of Values — ETL_UNSPEC_DATE must be defined in the same data format. If one is changed, the others must be changed accordingly.

  • List of Values must be defined appropriately. If there is no appropriate entry in List of Values, the strings that depend on List of Values in the Oracle Business Analytics Warehouse will not be translated.

  • There must be exchange rates defined for the currencies your organization deals with. If the appropriate exchange values are not found, the ETL process uses the ETL Unknown Exchange Rate defined in the DAC Source System Parameters.