SPACE

Appearance

TM1 data ​

SPACE AI lets users explore and analyze connected IBM Planning Analytics data using natural language. Ask a business question, investigate a result or variance, or describe the data you need without having to construct the underlying TM1 query yourself.

The assistant can use SPACE-controlled tools to inspect the model, identify relevant dimensions and elements, query TM1, and analyze the returned data. Data access remains governed by SPACE connection access and IBM Planning Analytics permissions.

Asking questions about TM1 data ​

Start with the business question you want to answer. For example:

  • Show me revenue by region for the last six months.
  • Compare Actual and Budget for operating expenses.
  • Which products had the largest decline this quarter?
  • Show headcount by department and location.

When useful, include details such as the cube, period, scenario, measure, or other selections. If the required context is unclear, the assistant can inspect available model information or ask for clarification.

You can continue with follow-up questions without rebuilding the entire query. For example, after reviewing revenue by region, ask the assistant to focus on one region, compare it with another period, or investigate the contributors to a variance.

Analyzing results and variances ​

SPACE AI can use retrieved TM1 data to compare periods, scenarios, organizational units, products, accounts, and other model dimensions.

For example:

  • Explain the largest Actual versus Budget variances.
  • Which regions contributed most to the decline?
  • Compare this quarter with the same quarter last year.
  • Break this variance down by product.

The assistant can progressively refine the analysis by retrieving additional data as needed. If understanding a result requires examining the calculation behind it, SPACE AI can also inspect the relevant model logic. See Models and logic.

MDX and data retrieval ​

SPACE AI can construct and work with TM1 MDX using available TM1 syntax and functions. Users normally do not need to write MDX themselves: describe the required data in natural language and let the assistant build the appropriate query. Advanced users can also provide specific MDX requirements when useful.

When retrieving TM1 data for the AI Assistant, SPACE returns a manageable portion of the result together with its row, column, and filter context. By default, SPACE retrieves up to 50 rows and 20 value columns at a time.

For larger results, SPACE supports pagination so the assistant can retrieve additional data as needed instead of loading an unrestricted cube result at once.

MDX data retrieval is read-only. Creating or modifying model objects is handled through separate supported SPACE tools.

Visualizing results ​

SPACE AI can visualize returned data as charts when a visual comparison or trend makes the results easier to understand.

For example:

  • Show Actual and Forecast revenue by month as a chart.
  • Visualize the largest budget variances by department.
  • Chart the monthly trend for these results.

You can ask for a visualization as part of the original question or after reviewing the returned data.

Saving analysis results as a TM1 view ​

After analyzing TM1 data, you can ask SPACE AI to save the latest results as a cube view. This turns a useful analysis from the conversation into a reusable Planning Analytics object.

For example:

  • Save these results as a view called Revenue Variance Analysis.
  • Save the last results as a private cube view.
  • Create a public view from these results called Monthly Forecast Review.

Saved views can be reused in IBM Planning Analytics Workspace or accessed again through SPACE and SPACE AI.

Views are private by default unless you explicitly request a public view. Creating the view requires the appropriate permissions in the connected Planning Analytics environment.

Views and subsets ​

SPACE AI can work with TM1 views and subsets as part of data exploration and development workflows. It can create or update both private and public views and subsets using supported SPACE tools.

When creating a view or subset, SPACE AI creates a private object by default. Ask explicitly for a public view or subset when the object should be shared.

For example:

  • Create a private view showing Actual and Budget by month for this department.
  • Create a subset containing the active cost centers.
  • Create this as a public view so other users can use it.

Available actions remain subject to the connected environment and the user's permissions.

Data access and security ​

SPACE AI does not bypass IBM Planning Analytics security. TM1 authentication and object and data permissions continue to determine what a user can access when the assistant queries a connected model.

SPACE connection access also determines which configured environments are available to the user. See Connections and Access for SPACE configuration, or Users and access for analyzing and troubleshooting TM1 user access.

SPACE AI can also work with supported external data sources, such as Google BigQuery, alongside connected Planning Analytics data. This allows a conversation or analysis to use information from multiple configured connections where appropriate.

See Connected data for more information.