• Reduce time spent reading through multiple case notes
• Create allocations faster with conversational interface
• Automate repetitive data entry tasks

• Identify patterns and trends across multiple notes
• Detect urgency levels and risk indicators
• Get actionable recommendations based on data

• Natural language interaction with AI chatbot
• Structured, easy-to-read summary reports
• Intelligent validation prevents errors

To access the AI summary feature:

• Navigate to the Operations tab
• Open the Case Notes section
• A purple circular button with an AI brain icon appears in the top-right corner
• The button is visible only when case notes exist for summarization
• It becomes disabled during summary generation to prevent duplicate requests
• A loading spinner is shown while the AI processes the notes

The AI summary feature enforces strict permission rules:

• Requires CaseNote_View permission
• Users can only summarize notes they are allowed to view
• If permissions are insufficient, an error message is displayed
• Permission checks happen both on frontend and backend

The summary respects all current filters and settings:

• Users may filter by:
  • Carer - Specific employee or all carers
  • Client - Specific client or all clients
  • Date Range - Custom time period
• Users may also search using the Search field
• Apply desired filters before clicking the AI brain icon
• Only the filtered/visible case notes will be included in the summary

• Once filters are applied, click the AI brain icon
• AI processes all visible case notes based on filters
• Processing time varies based on number of notes (typically 5-15 seconds)
• Loading indicator shows progress
• Summary modal opens automatically when ready

• Title: "Case Notes Summary" styled with a purple theme
• Badge: "Powered by AI" indicator shows AI-generated content
• Close Button: Standard X button to close the modal

Each section provides structured insights from the analyzed case notes:

Executive Summary

• High-level overview of all notes
• Major events and milestones
• Overall tone and sentiment of the notes
• Expanded by default for quick access

Key Observations

• Significant findings and patterns
• Behavioral changes noted by carers
• Health concerns or improvements
• Emotional patterns observed
• Notable incidents or events

Care Activities

• Summary of care actions performed
• Medication administration records
• Personal care activities
• Mobility support provided
• Other assistance rendered

Progress Notes

• Client improvements or changes over time
• Positive improvements identified
• Areas of regression or concern
• Stability or fluctuations in condition
• Responses to interventions

Action Items

• Recommended next steps
• Follow-ups required
• Risk mitigation strategies
• Communication needs
• Additional assessments suggested
• Specific carer instructions

Patterns & Trends

• Recurring themes across notes
• Repeated complaints or concerns
• Frequent incidents of similar nature
• Consistent behaviors observed
• Ongoing challenges identified
• Daily routines and preferences

Displays important context about the generated summary:

• Total Notes Processed: Helps users understand how much data the AI analyzed
• Date Range: Shows the time period covered by the summary
• Urgency Level: AI detects if the notes indicate low, medium, or high concern
• Confidence Score: Indicates how confident the AI is in its interpretation
• Note Types: Categories of notes included (Incident, Medication, General, Discharge, etc.)

• Based on current filters (Carer, Client, date range, search text, pagination)
• Only includes notes the user has permission to view
• Respects role-based access controls
• Maintains data security throughout process

The AI model performs sophisticated analysis:

• Reads every note in the filtered set
• Identifies patterns, themes, concerns, and care activities
• Extracts important details and context
• Organizes information into structured sections
• Applies natural language understanding to interpret meaning

The AI creates a comprehensive report:

• Output is a multi-section report with clear organization
• Each section focuses on a different aspect of the client's care journey
• Content is written in natural, professional language
• Summaries are concise yet comprehensive

AI calculates important summary metrics:

• Number of notes processed: Total count of analyzed notes
• Time period covered: Date range of included notes
• Urgency indicators: Risk level assessment (low/medium/high)
• Confidence score: How certain the AI is about its analysis
• Types of notes: Categories included in summary

The Work Allocation System is used to schedule carers for NDIS (National Disability Insurance Scheme) and CHSP (Commonwealth Home Support Programme).

Creates Allocation Between:

• Employee (Carer) → Client
• With defined start date/time and end date/time

Use Cases:

• When you know exactly which carer will provide the service
• Scheduled client visits with specific carers
• Regular recurring care schedules
• Personal care, medication support, mobility assistance

Example:

The Department Allocation System assigns entire departments to clients. This is useful when the specific carer is not yet known or when the department will choose the appropriate staff member later.

Creates Allocation Between:

• Department → Client
• With defined start date/time and end date/time
• A carer can be assigned later by the department

Use Cases:

• When the specific carer is not yet determined
• Department-level resource planning
• Flexible staffing arrangements
• Pool allocation where any qualified staff can be assigned

Example:

To open the chatbot:

• Navigate to the Operations tab
• Open the Scheduler section
• This page displays all existing work allocations
• A purple AI button with a spark icon appears in the top-right corner
• The button is visible only if the user has permission to create allocations

The chatbot enforces strict permission rules:

• Users must have Work Allocation - Create permission
• Only authorized users can create employee or department allocations
• If permissions are missing, the chatbot displays an appropriate error message
• Permission validation happens before the chat interface opens

• Click the purple AI button to open the chat panel
• Chat panel slides in from the right side of the screen
• Chatbot greets you and asks for allocation type
• You can close the panel at any time using the X button

After creating allocations, users can filter the list by:

• Client: Filter by specific client
• Carer: Filter by specific employee
• Status: Active, completed, cancelled
• Type: Employee or department allocation
• Notes: Search in allocation notes
• Dates: Filter by date range

These filters help users quickly locate and review specific allocations after creation.

• Title: "Work Allocation Assistant"
• Subheading: "Create allocation via chatbot"
• Refresh Button: Restart the conversation
• Close Button: Close the chat popup

• Chatbot prompts: "What type of allocation would you like to create?"
• User types either "Employee" or "Department"
• Chatbot waits for correct input and confirms selection
• Example: "I need an employee allocation"

When a user types an employee name (e.g., "james"), the assistant:

• Searches for matching employees in the system
• Displays a numbered list if multiple matches exist
• Waits for the user to choose the correct one by number
• After selection, confirms: "Got it. I've selected James Anderson."

When a user types a client name (e.g., "alice"), the assistant:

• Searches for matching clients in the system
• Displays a numbered list if multiple matches exist
• Waits for the user to choose the correct one
• After selection, confirms: "Got it. I've selected Alice Johnson A."

Once the client is selected, the assistant asks for:

• Start date: When the allocation begins
• Shift start time: Time of day the shift starts
• Shift end time: Time of day the shift ends
• Repeat pattern: Whether the allocation should repeat

This ensures all essential scheduling data is captured before proceeding.

Users can enter time and date in natural language, such as:

• "5pm to 4am this Thursday"
• "Next Monday 9 to 5"
• "Tomorrow evening"
• "Starting Friday 8am until 2pm"

The assistant interprets this into structured data:

It then confirms the interpretation before moving forward.

Users can specify repeat patterns naturally:

• "Yes every week for another 3 months"
• "Repeat daily for 2 weeks"
• "No repeat" or "Just once"
• "Every Monday and Wednesday for 6 weeks"

The assistant converts this into:

• Frequency: Daily, weekly, specific days
• End date: Calculated from the duration (e.g., 3 months from start)

It then confirms: "This allocation will repeat weekly until May 12th, 2023."

Before creating the allocation, the assistant summarizes:

• Client: Name and ID
• Employee: Name and ID (or Department)
• Start & End Time: Complete datetime range
• Repeat Pattern: Frequency and end date (if applicable)

It asks for a final "yes" to proceed with creation.

If an overlapping allocation exists, the assistant alerts the user:

The user can then adjust:

• Dates or times to avoid overlap
• Choose a different employee
• Choose a one-time allocation instead of repeat

Example user correction: "Ok allocate only for this Thursday"

The assistant re-confirms the updated details and checks for conflicts again.

Once confirmed and validated:

• The assistant creates the allocation in the system
• Displays a success message: "✅ Allocation creation completed."
• Shows allocation details for reference
• Offers a shortcut button: "Create Another Allocation"

Users can either:

• Click "Create Another Allocation" to start fresh
• Close the chat panel
• View the new allocation in the Scheduler list