IBM Planning Analytics

IBM Planning Analytics Tips & Tricks: Impact of Consolidations in Feeder Statements

May 29, 2018 by Revelwood Leave a Comment

This is a guest post from Revelwood’s Stephen Ellis.

Did you know writing feeders in TM1 is one of the more difficult tasks for any developer? One of the questions I am asked most often is “How do I know if I’m feeding this correctly?”

Today’s post will help you visualize the impact of utilizing consolidations in feeder statements. It will also help you interpret and review a feeder statement. These skills will help you develop rules within Planning Analytics. It can also assist you in reviewing your feeders with an eye for overfeeding.

The first example is for the feeder [‘Actual’,’Jan’] => [‘Forecast’,’Jan’]. This should be read as “Actual for January feeds Forecast for January.” Below is a visual representation of this feeder:

Learn about feeder consolidations in IBM Planning Analytics

Pretty straightforward, right? An individual month in one version feeds that same month in a different version (or Scenario).

If the feeder changed to [‘Actual’,’Jan’]=>[‘Forecast’,’Year’], which is read as “Actual for January feeds the Year for the Forecast,” what is the impact?

Learn the impact of consolidations on feeder statements

How did that one feeder statement create twelve separate feeders? Planning Analytics interprets any declaration of a consolidation in a feeder statement as an identification of every lowest level element within that consolidation. To clarify, in this example “Year” is the sum of all individual months.

Let’s review the previous statement. The feeder statement I previously declared was “Actual for Jan feeds the Year for the Forecast.” This can be rewritten as “Actual for Jan feeds all months within Year for the forecast.” Let’s do a few more examples to solidify this concept.

The next example is for the feeder “[‘Actual’,’Year’]=>[‘Forecast’,’Jan’]” This is read as “The year of actuals feeds January for Forecast.” Knowing that year is a consolidation of all months, we can change the interpretation to: “All months within year for Actual feeds January of Forecast.” Please see below representation.

Learn how consolidations impact feeder statements

This feeder statement also created 12 separate feeders, except that they all point to the same data point (January Forecast). In practice, this is an example of overfeeding, which can negatively impact the performance of your Planning Analytics application.

The last example is [‘Actual’,’Year’]=>[‘Forecast’,’Year’]. This is read as “The year of actuals feeds the year of Forecast,” and could be rewritten as “The months within Actuals feeds the months within Forecast.” This statement creates 144 (12×12) feeders, or the full combination of months. Below is an abbreviated representation of this feeder statement.

For illustration purposes, I’ve provided the first four months as impacted by this example.

With this understanding, you can read, write, and review your own feeder statements!

IBM Planning Analytics is full of new features and functionality. Not sure where to start? Our team here at Revelwood can help. Contact us for more information at info@revelwood.com. And stay tuned for more Planning Analytics Tips & Tricks weekly in our Knowledge Center and in upcoming newsletters!

IBM Planning Analytics Tips & Tricks: Planning Analytics Workspace Chart Types

May 22, 2018 by Revelwood Leave a Comment

This is a guest post from Revelwood’s Susan Musselman.

Did you know charts and graphs are an excellent way to illustrate relationships in data and highlight trends in IBM Planning Analytics Workspace? They are visualizations of data that allow end users to quickly analyze a large amount of data. To make the biggest impact, it is important to choose the right chart for the use case of the data. Following is a list of chart types available within Planning Analytics Workspace, along with use cases where one type might be better than another.

Area Charts

Chart types in IBM Planning Analytics Workspace

An area chart emphasizes the magnitude of change over time. It is good for showing trends. Example: show how sales values have changed over a number of years.

Bar Charts

Bar charts are used to compare different items, or to show a comparison of items over time. Compared to Column charts, Bar charts are good for displaying long data labels. An example would be a Top 10 Salesperson chart where the full names are listed.

Stacked Bar Charts

Stacked bar charts are designed to help you simultaneously compare totals and notice sharp changes at the item level that are likely to have the most influence on movements in category totals.

Column Charts

Similar to bar charts, but with a vertical orientation instead of horizontal. Column charts are good for displaying data sets with negative values. A good example would be a Regional Profit comparison chart where some regions have a loss.

Stacked Column Charts

Similar to column charts, but it shows multiple items in each column. Example: showing 4 quarters of Orders by Region.

Line Charts

Line charts show trends over time. The chart can show just one line or multiple lines, but the number of lines should be limited so the chart does not get too busy.

Map Charts

A map chart displays data based on geography. A good use of this chart type would be to show relative Revenue by State.

Packed Bubble Charts

packed bubble chart_ chart types in IBM Planning Analytics Workspace

This chart is a way to show relational value without regards to axes. The bubbles are in different sizes and colors. The x-axis represents one measure, the y-axis represents another measure, and the size of the bubbles represents the third measure. A good use would be Product Sales within Product Types. This chart is best for positive numbers.

Pie Charts

Shows each piece as a component of a larger part. Good for showing percentages as part of a whole. Does not show changes over time.

Point Visualizations

A point visualization uses multiple points to show trends over time. It is similar to a line chart, but without the lines; only the data points are shown.

Radial Charts

Shows progress toward a goal. Comparisons can be made by having multiple circles. Too many will be busy and ineffective.

Radial Bar Charts

This chart is pleasing to the eye, but it may be difficult to interpret. It is similar to a bar chart, but it is displayed in a circular fashion.

Tree Maps

A tree map identifies patterns and exceptions in a large, complex data set. It should not be used with negative numbers. It shows relationships among large numbers of components by using size and color coding in a set of nested rectangles.

Word Clouds

A word cloud is a visual representation of text values. The more frequently a text string occurs in your data, the larger the string appears in the word cloud.

When choosing which type of chart to use, don’t forget the purpose: to make it faster and easier for end users to understand the data. The less explanation it needs, the better.

IBM Planning Analytics Tips & Tricks: Regions with Planning Analytics Rules

May 15, 2018 by Lee Lazarow Leave a Comment

Did you know that comments are an important piece of good rule development in IBM Planning Analytics? Comments help to describe your logic and can make it easier to navigate through many lines of code. A special type of comment in Planning Analytics entails the definition of a region. The use of regions allows you to expand and collapse sections of rules in the Planning Analytics rule editor.

A region is defined by adding a comment that starts with #Region and ends with #EndRegion. The start of the region also allows you to describe the region via a RegionName.

Here is an example of a rule that includes a region definition. Note that the first line includes both the #Region comment and a description of the section.

Regions with IBM Planning Analytics Rules

Here is the same rule in the Planning Analytics rule editor. Note the grouping symbol (the triangle) next to line 6.

IBM Planning Analytics Tips & Tricks: Regions with Planning Analytics Rules

The section can be collapsed to make your rule easy to read. You can hover the mouse cursor over a collapsed region to show the contents as a tooltip.

This simple approach will allow you to make your rules easier to navigate and read when scrolling through many lines of code.

IBM Planning Analytics Tips & Tricks: The Data Hold Feature in Planning Analytics Workspace

May 8, 2018 by Revelwood Leave a Comment

This is a guest post from Revelwood’s Shane Bethea.

Did you know that you can hold (e.g. lock) a data point in a view in IBM Planning Analytics Workspace simply by typing the word ‘hold’ into a cell? To do this, create a view to do data entry. The below view has the first 4 months locked, allowing entry into May through December.

Next, enter a value into the consolidated year to spread evenly. I have entered 800 into the year 2016. This spreads 100 to May through December.

To place a hold on a particular month, type the word HOLD into the cell (I chose May) and you will see a symbol next to the data point.

Now change the consolidated year to a different value and the hold month will not change.

You can release the hold by right clicking on the cell and selecting ‘release hold’ or by simply typing the word RELEASE into the cell. The hold symbol will no longer be visible.

IBM Planning Analytics Tips & Tricks: GMT vs Local Time

May 1, 2018 by Lee Lazarow Leave a Comment

Did you know that TM1 offers an easy way for you to convert GMT times to local times? Do you get tired of seeing your message log records in GMT? Are you frustrated that you have to convert times from GMT to local time? Does the concept of daylight savings time annoy you? Well, there’s not much that TM1 can do about that last question, but it does offer a solution to the first two.

TM1 contains many of your configuration settings within a file called tm1s.cfg, but there is also a second file called tm1s-log.properties. The properties file is designed to control message levels that are logged by TM1, but it also contains a configuration option to define the TimeZone.

One of the parameters in the tm1s-log.properties file is:

log4j.appender.S1.TimeZone=Zone

where Zone can be set to either GMT or Local.

If the parameter is set to Local, the log messages will have timestamps based on the local time of the machine where the TM1 server is running.
If the parameter is set to GMT, the log messages will have timestamps based on Greenwich Mean Time.
If the parameter is set to something other than GMT or Local (or if the parameter is not set at all), it defaults to GMT.

The properties file must be located in the same folder as the tm1s.cfg file and you must restart your TM1 server after changing this setting.

This setting will help you make your log files easier to read when looking at time stamps within the various lines of the log.

IBM Planning Analytics Tips & Tricks: Ensuring Cube Logging is Reset in TM1

April 24, 2018 by Revelwood Leave a Comment

This is a guest post from Revelwood’s Susan Musselman.

Did you know you can easily reset cube logging in TM1? A previous tech tip was written to discuss when to enable and disable logging. As stated in that posting, it is best practice to turn cube logging off while processing TurboIntegrator processes. Typically, logging is turned off in the Prolog and turned back on in the Epilog. But if an error occurs when the process is run, it never reaches the Epilog and logging never gets turned back on.

Initially, logging is set by right-clicking on cubes and placing a check in the logging box for each cube that needs logging turned on.

Learn how to ensure cube logging is reset in TM1

When logging is turned off via TI process, the checkbox is turned off.

One way to ensure that settings aren’t permanently “lost” is to reset logging in a chore that runs on a regular schedule (nightly or more frequently, if desired). To do this, set up an attribute in the }Cubes dimension to define which cubes should have logging turned on and manually populate the attribute with “YES” for each cube that should be turned on for logging. In the following example, the attribute is called “LogChore” and two cubes are defined to have logging enabled.

Then create a process with “None” as the source that will loop through the cubes and set log changes to 1 when the attribute is populated. Here is the code to put in the Prolog tab.

iCount = 0 ;

iSize = DIMSIZ ( ‘}Cubes’ ) ;

WHILE ( iCount < iSize ) ;

iCount = iCount + 1 ;

strElement = DIMNM ( ‘}Cubes’ , iCount ) ;

IF ( SUBST ( ATTRS ( ‘}Cubes’ , strElement , ‘LogChore’ ) , 1 , 1 ) @= ‘Y’ ) ;

CUBESETLOGCHANGES ( strElement , 1 ) ;

ENDIF ;

END ;

The final step is to set the process up as a chore and schedule it to run nightly. Any cubes that may have accidentally had Logging turned off will be turned back on.

IBM Planning Analytics Tips & Tricks: Captions

April 17, 2018 by Lee Lazarow Leave a Comment

Did you know TM1 allows users to create alias names for elements within your model? This is a great approach that allows you to give different names to the same element. For example, account 601 can also be displayed in reports as “601 – Comp Expense”.

Many clients have asked me for an easy way to do something similar with dimension and cube names since many of these are often created with some form of prefix (example: a glbProduct dimension or staffEmpInfo cube).

Planning Analytics allows you to create an alias called “Caption” within both the }CubeAttributes cube and the }DimensionAttributes cube. This setting will allow you to display a cube or a dimension using “friendly” names when displayed in your PAW books.

The following example shows a model containing a dimension called bpmCountry. The value within the }DimensionAttributes cube is set to “Country”.

Working with captions in IBM Planning Analytics

As a result of this setting, any references to the dimension will appear in PAW as “Country” instead of “bpmCountry”.

Using captions in IBM Planning Analytics

This simple approach will allow you to make your Planning Analytics components more user friendly and easier to read.

IBM Planning Analytics (built on the TM1 engine) is full of new features and functionality. Not sure where to start? Our team here at Revelwood can help. Contact us for more information at info@revelwood.com. And stay tuned for more Planning Analytics Tips & Tricks weekly in our Knowledge Center and in upcoming newsletters!

IBM Planning Analytics Tips & Tricks: TurboIntegrator Logging

April 3, 2018 by Lee Lazarow Leave a Comment

Did you know you can easily control your logs, making them easier to navigate? Your TM1 server tracks data transactions made in the system. When a cube value is changed, TM1 records the change in a transaction log file named Tm1s.log. The values in the log can be used as an audit trail and can also be used as part of a data restoration approach in the event of a server crash.

While logging is a great auditing tool and an excellent point of reference, these logs can quickly become large and hard to navigate. This post shows you various ways to control your logging approach and offers some recommendations on when to disable it.

}CubeProperties

The }CubeProperties control cube includes settings to define how each cube performs within the model. One of these settings is called “LOGGING” and the value determines whether the results of data changes are logged in the tm1s.log file. A value of “YES” states that data changes will be logged and any other value (“NO”, blank, etc.) states that changes will not be logged.

The following example shows logging turned on for the bpmFinance_Metrics cube and turned off for the bpmFinance cube.

This setting is designed to serve as a default value for the cube.

TurboIntegrator

Although each cube has default values assigned to it, you can temporarily overwrite the setting via the following TurboIntegrator function:

CubeSetLogChanges(Cube, LogChanges);

The “Cube” parameter defines the name of the cube. The LogChanges value can be set to 1 to enable logging and 0 to disable logging.

Recommendation

So now that we know how to enable and disable logging temporarily, the next step is to determine when to use it. Logging is a great approach when you need a mechanism to track user entry, but it is not needed when loading data from an external source system that can easily be tracked and restored.

It is recommended that logging be disabled in the prolog of a TurboIntegrator process and re-enabled in the epilog within the following data processes:

Loading Actuals from an external source system such as a database or a file
Replicating data in your system (example: making a snapshot of forecast data into another scenario)

Summary

The proper use of logging can help minimize the amount of disk space needed in your environment and may also improve the efficiency of your data load processing.

IBM Planning Analytics Tips & Tricks: False options within an IF statement

March 27, 2018 by Lee Lazarow Leave a Comment

Did you know that there are three different ways to define the false section of an IF statement? The approach you choose will impact how the cell calculates a result. In this blog post, we discuss these options and when to use each one.

TM1 uses a standard approach to the concept of an IF statement:

IF (Argument, True Result, False Result)

This approach is great when you have specific values for both the true result and the false result, but what happens in situations where we only care about the true result? The required parameters do not allow us to ignore the false result, so what do we do?

TM1 Rules allow for three different options in the False Result section:

CONTINUE: This is designed to tell TM1 to continue reading the rule until the next applicable calculation should be used.

STET: This is designed to tell the rule to stop and leave the cell empty.

A specific value (example: the number 0): This is designed to tell the rule to stop and put a specific value into the cell.

So when should you use each option?

CONTINUE – use this statement when there is another rule that may apply to the cell. For example, you may have a generic rule that looks at all forecast values but also have a more detailed rule that looks at a specific account within the forecast. The use of CONTINUE will allow TM1 to check the forecast rule while still giving the option of calculating the specific account.

STET – use this statement when you want the rule to “end” while giving users the ability to enter a value. For example, you may want to have a cell calculated for ACT months while keeping it write-able for planning months.

A specific value – use this when you want the rule to “end” without giving users the ability to enter a value. For example, you may want to have a bonus calculated for managers while ensuring that no bonus amounts can be entered for non-managers.

As you can see, each of these options perform a specific function… so be sure to think about which approach is best for your rule.

IBM Planning Analytics Tips & Tricks: Impact of Consolidations in Feeder Statements

Read more IBM Planning Analytics Tips & Tricks blog posts: