Process definition

​

What is a process definition?

​

Key characteristics of process definitions

​

Understand definitions and instances

​

Manage the process lifecycle

​

Work with core components

Process definition navigation menu

​

1. Process Designer

Process Designer canvas showing swimlanes, nodes, and process flow

• Node palette: Access all BPMN node types (User Tasks, Service Tasks, Gateways, Events, etc.)
• Swimlanes: Organize process steps by role or department for access control
• Canvas controls: Zoom, pan, and navigate large process diagrams
• Sequence flows: Connect nodes to define the execution path
• Visual feedback: Real-time validation warnings and configuration status

​

2. UI Designer

UI Designer showing component library and interface preview

• Component library: Pre-built UI components (Forms, Inputs, Buttons, Tables, Cards, etc.)
• Drag-and-drop design: Build interfaces by dragging components onto the canvas
• Live preview: See how your UI will appear to users in real-time
• Platform-specific design: Configure separate UIs for Web, iOS, and Android
• Data binding: Connect UI elements directly to process data
• Theme support: Apply consistent styling across your application

​

3. Data Model

Data Model showing types, structures, and relationships

• Type definitions: Create reusable data structures with typed fields
• Visual relationships: See how data objects connect and reference each other
• Input parameters: Define what data the process expects to receive
• Output parameters: Specify what data the process returns
• Example values: Include sample data for testing and development
• Validation rules: Ensure data integrity with type checking

​

4. Settings

Settings interface showing process configuration tabs

• Process Name: Set the unique identifier with naming constraints
• General: Platform selection, reporting, task management, expiry time
• Permissions: Configure swimlane roles and access control
• Task management: Integration with Task Manager plugin
• Data Search: Define searchable and filterable process data keys

​

5. Start a process for testing

• Automatically generates input forms based on your defined input parameters
• Validates required data before the process starts
• Creates a test process instance in your current environment
• Helps you test and debug during development

​

6. Access additional operations

Process definition contextual menu

Clear Cache

Clear cached process definition data to force a refresh from the database.When to use:

• After making significant changes that aren’t reflecting in the UI
• When experiencing unexpected behavior or outdated data display
• To troubleshoot synchronization issues

This only clears the cache for the current process definition in your browser session.

Audit Log

View comprehensive change history for the process definition.Information tracked:

• Timestamp of each change
• User who made the change
• Application version
• Subject and event type
• Subject identifier
• Status of the change

Features:

• Filterable entries for easy tracking
• Complete audit trail for compliance
• Version comparison capabilities

Audit documentation

View History

Access the version history of the process definition to see all changes over time.

Select a version

View changes

Capabilities:

• Compare different versions side-by-side
• See what changed between versions
• Track evolution of the process definition
• Identify when specific changes were introduced

This shows the version history within the project versioning system, allowing you to see how the process evolved across committed versions.

Delete

Remove the process definition from the project.

Delete process definition

Use with caution: Deleting a process definition is a significant action that should only be performed when:

• The process is no longer needed
• You’ve verified that no other processes reference this one as a subprocess
• You have appropriate backup or version control

Impact:

• The process definition is removed from the current wip version (you cannot delete a process definition that is part of a committed version)
• References from other processes may break
• Historical data of process instances is not affected
• Can be recovered from previous committed versions if needed

​

Define your data model

​

Why define a data model?

• Type safety: Prevent errors during process execution by defining data types
• Auto-completion: Get intelligent suggestions in UI Designer based on defined keys
• Validation: Ensure data consistency across nodes and actions
• Documentation: Use the data model as living documentation of your process data structure
• Reusability: Reference data structures across multiple nodes and actions
• Testing: Include example values to test without creating mock data

​

Configure input and output parameters

​

Access parameter configuration

• Enable process start validation
• Generate automatic input prompts for testing
• Support Call Activity nodes and subprocess integration

• Return data to parent processes
• Enable synchronous subprocess communication
• Replace manual “Append Data to Parent Process” actions

Configuring input parameters

​

How to configure parameters

1. Navigate to Data Model → Input/Output Parameters.
2. Click Define parameters.
3. Configure the schema:
   • Add the parameter name and data type.
   • Mark parameters as required or optional.
   • Add descriptions and example values.
4. If you have multiple start or end nodes, select the specific node from the dropdown.

​

Key features

• Automatic validation of required parameters
• Type checking ensures data integrity
• Change propagation: Deleted/renamed attributes automatically update in mappings
• Multiple nodes support: Different parameters for different start/end points

​

Organize processes with swimlanes

​

Benefits of using swimlanes

• Improve visual clarity: Show which team or role is responsible for each process step
• Control access: Restrict who can execute, view, or self-assign tasks in each lane
• Ensure compliance: Enforce separation of duties by distributing tasks across different roles
• Optimize processes: Identify bottlenecks by seeing which roles handle the most tasks

​

Add swimlanes

1

Open the process

1. Access the FlowX.AI Designer
2. Open an existing process definition or create a new one

2

Access swimlane menu

Identify the default swimlane and select it to display the contextual menu.

With the contextual menu, you can easily perform various actions: add or remove swimlanes, reorder them, or configure their settings.

3

Add swimlane

1. Choose the desired location for the new swimlane (above or below existing lanes)
2. Click the add swimlane icon to create the new swimlane
3. Give the swimlane a descriptive name (e.g., “Front Office”, “Approver”, “Back Office”)

4

Configure permissions

After creating a swimlane, configure which roles can access it in the Permissions section of process settings.

​

Configure process settings

​

Set your process name

• Process definition name: The unique name for your process

• Use descriptive names that clearly identify the process purpose.
• Follow a consistent naming convention across your organization.
• Use underscores or hyphens instead of spaces.
• Keep names concise but meaningful (for example, Customer_Onboarding or Loan_Application).

​

Configure general settings

General settings tab

​

Select your deployment platform

• Web Only: Use this option when your navigation areas are defined exclusively for web. The process is optimized for web usage and doesn’t provide the same functionality on mobile devices.
• Mobile Only: Use this option when your navigation areas are defined only for mobile. The process leverages mobile-specific features and provides an optimal mobile experience.
• Omnichannel: Use this option when your process has navigation areas defined for both web and mobile platforms. This ensures users have a consistent experience regardless of their platform.

​

Enable read-only access

Read-only process configuration toggle

• Supervision and oversight: Let supervisors review process instances without making changes.
• Audit and compliance: Let auditors view process data and history without risk of modification.
• Information sharing: Let stakeholders view process status without granting edit permissions.
• Training and demo: Show process flows and data to trainees without allowing changes.

1. Enable “Allow Read-Only Process Configuration” in the General settings tab.
2. Configure read-only permissions at the swimlane level in the Permissions tab.
3. Use permission-based expressions in UI Designer to control element visibility and interaction.
4. Enable or disable the toggle at the node level to configure read-only access for specific nodes.

​

Enable reporting

​

Configure general data

• Supervision messages: Display important notices about account supervision or approval requirements.
• Document generation status: Show immediate feedback about document processing.
• Process-level notifications: Present critical information that needs to be visible from the start.
• Banner messages: Display persistent notifications at the top of the process interface.

{
  "notifications": [
    {
      "identifier": "Supervision",
      "show": "true", 
      "componentIdentifier": "MESSAGE",
      "displayOptions": {
        "flowxProps": {
          "text": "Account opening requires supervision. The client's account and products will be active after receiving necessary approvals.",
          "messageType": "warning"
        }
      }
    }
  ]
}

​

Set expiry time

​

Set permissions

Permissions settings tab

​

Configure swimlane permissions

​

Default permissions

• Execute: Allows users to perform actions in the swimlane
• Self-assign: Enables users to assign tasks to themselves

​

Integrate with Task Manager

Task management settings tab

​

Available configuration options

{baseURL}/appId/buildId/processes/resourceId/instance

• Define the URL dynamically using configuration parameters (for example, ${genericParameter}).
• Set the URL at the process data level using business rules.

https://demo-angular.devmain.flowxai.dev/

​

Configure data search

Data Search settings tab

​

Index search keys

1. Click the + button to add a new search key.
2. Enter the key path (for example, application.client.firstName).
3. Save your changes to apply the indexing configuration.

• Use dot notation to access nested fields: parent.child.fieldName.
• Keys must match your data model structure exactly.
• Matching is case-sensitive.

• application.client.firstName
• application.client.lastName
• application.status
• application.dateOfBirth

​

Apply best practices

​

Follow configuration standards

• Choose Web Only for desktop-focused applications with complex interactions.
• Choose Mobile Only for mobile-first experiences optimized for touch.
• Choose Omnichannel only when you need both platforms. Maintain separate processes when platform requirements diverge significantly.

• Use descriptive, consistent names that clearly identify the process purpose.
• Follow a naming convention like [Department]_[Function]_[Action] (for example, HR_Onboarding_NewEmployee).
• Avoid special characters beyond the allowed set: [ ] ( ) . _ -.
• Keep process names under 50 characters for better readability in lists.

• Define your data model structure before building process flows.
• Use meaningful key names that describe the data they contain.
• Organize related data under common parent keys (for example, customer.firstName and customer.lastName).
• Include example values in the data model to help with development and testing.
• Avoid deeply nested structures—limit nesting to 3-4 levels for better performance.

​

Configure swimlanes effectively

• Create swimlanes that mirror your organizational structure or process ownership.
• Assign clear, descriptive names to swimlanes (for example, “Front Office”, “Back Office”, or “Supervisor”).
• Configure appropriate permissions for each swimlane (execute, view, self-assign).
• Enable Allow multiple executors when multiple users with different roles need concurrent access.
• Limit the number of swimlanes—too many can make processes difficult to visualize. Aim for 3-5 maximum.

​

Optimize performance

• Expiry time: Set appropriate expiry times to prevent stale process instances from accumulating.
• Task management integration: Only enable when needed—it adds overhead for task tracking.
• Reporting integration: Enable selectively for business-critical processes that require analytics.
• Search indexing: Index only the keys you need for searching—excessive indexing affects performance.
• General data: Keep general data lightweight—it loads on every process start.

• Don’t store large binary data in the process data model.
• Use references to external systems for large files or documents.
• Clean up temporary data that’s no longer needed as the process progresses.

​

Follow security best practices

​

Configuration checklist

​

Manage process versions

​

Understanding versioning

• Draft versions: Work-in-progress changes that you can edit freely.
• Committed versions: Finalized snapshots that become immutable.
• Branches: Parallel development paths for different features or environments.
• Version comparison: Tools to compare settings and configurations across versions.

​

Make changes to process definitions

​

Understand the impact on running instances

• You can safely deploy new process definition versions without affecting active users.
• Long-running processes complete using their original definition.
• New process instances use the latest deployed version.
• To apply changes to running instances, you need to migrate them (typically not recommended).

​

Troubleshoot common issues

​

Get help with other issues

1. Check the audit log to see recent changes that might have caused problems.
2. Review process validation warnings in the Process Designer.
3. Test with a simple process to isolate whether the issue is process-specific or environmental.
4. Check Platform Status to ensure all FlowX services are healthy.
5. Review logs in your FlowX.AI deployment for detailed error messages.

​

Related resources