AI Scheduler
ShiftAI's core feature is its intelligent scheduling engine that understands natural language and creates optimized schedules automatically.
How It Works
The AI Scheduler uses advanced language processing to:
- Parse your request - Understand what you need in plain English
- Gather constraints - Check employee availability, skills, and preferences
- Optimize - Find the best assignment that satisfies all requirements
- Validate - Ensure no conflicts, overtime issues, or compliance violations
Natural Language Commands
Basic Syntax
Just type what you need like you're texting a colleague:
"Create a schedule for next week"
"I need 2 servers on Saturday night"
"Schedule Tom for Monday morning"
Specifying Time Ranges
| Format | Example | |--------|---------| | Relative dates | "next week", "tomorrow", "this weekend" | | Specific dates | "December 15", "Dec 15-21" | | Day names | "Monday", "every Friday" | | Time of day | "morning", "evening", "9am to 5pm" |
Role-Based Commands
"I need 3 baristas and 1 cashier for morning shift"
"Schedule 2 nurses for night duty"
"Make sure there's always a manager on site"
Employee-Specific Commands
"Schedule Alice for Tuesday and Thursday"
"Don't schedule Bob this week - he's on vacation"
"Maria prefers evening shifts"
Advanced Commands
Complex Requirements
You can combine multiple requirements:
"For next week, I need:
- Weekday mornings (7am-3pm): 2 staff
- Weekday evenings (3pm-11pm): 3 staff
- Weekends: 4 staff all day
- At least 1 senior employee on every shift"
Constraints and Preferences
"Create a fair schedule where everyone gets similar hours"
"Avoid scheduling anyone for more than 5 consecutive days"
"Try to give Sarah the hours she requested"
Modifications to Existing Schedules
"Add an extra server to Friday dinner"
"Swap Tom and Jerry's Saturday shifts"
"Cancel all shifts for next Monday"
Command Reference
Schedule Creation
| Command | What It Does | |---------|--------------| | "Create schedule for [date range]" | Generates a new schedule | | "Auto-fill [date range]" | Fills empty slots intelligently | | "Copy last week's schedule" | Duplicates previous schedule |
Coverage Management
| Command | What It Does | |---------|--------------| | "Need [N] [role] on [day/time]" | Sets coverage requirement | | "Add coverage for [period]" | Increases staff for a time | | "Reduce to [N] staff on [day]" | Decreases coverage |
Employee Assignment
| Command | What It Does | |---------|--------------| | "Schedule [name] for [time]" | Assigns specific employee | | "Remove [name] from [day]" | Unassigns employee | | "[name] is unavailable [period]" | Marks temporary unavailability |
Shift Operations
| Command | What It Does | |---------|--------------| | "Extend [name]'s shift by [hours]" | Lengthens a shift | | "Split [name]'s shift" | Divides into two shifts | | "Merge morning and lunch shifts" | Combines adjacent shifts |
Understanding AI Decisions
Why Did the AI Schedule This Way?
Click on any shift to see the reasoning:
- Availability match - Employee was available
- Skill match - Employee has required skills
- Preference match - Aligns with stated preferences
- Fairness score - Balances hours across team
When the AI Can't Fulfill Requests
You might see messages like:
- "Cannot satisfy constraint: insufficient available staff"
- "Warning: This would exceed overtime limits for [name]"
- "Conflict: [name] is already scheduled at this time"
When you see these messages, try relaxing constraints or checking employee availability settings.
Optimization Settings
Fairness Level
Control how the AI distributes shifts:
- Strict - Very equal distribution (±2 hours difference)
- Balanced - Reasonably fair (±5 hours difference)
- Flexible - Prioritize coverage over fairness
Preference Weight
How much to consider employee preferences:
- High - Try hard to match preferences
- Medium - Balance preferences with needs
- Low - Prioritize business needs
Cost Optimization
Enable to minimize labor costs:
- Prefers employees with lower hourly rates
- Avoids overtime when possible
- Suggests schedule changes to reduce costs
Cost optimization should be balanced with fairness. Over-relying on it may result in some employees getting very few hours.
Tips for Better Results
Be Specific
| Less Effective | More Effective | |----------------|----------------| | "I need people" | "I need 3 servers" | | "Schedule for the weekend" | "Saturday and Sunday, 10am-8pm" | | "Add more staff" | "Add 1 more cashier to Friday evening" |
Provide Context
The more the AI knows, the better it schedules:
- Keep availability data current
- Set skill levels accurately
- Note preferences in employee profiles
- Update constraints regularly
Iterate
Don't expect perfection on the first try:
- Generate initial schedule
- Review AI suggestions
- Provide feedback with adjustments
- Regenerate or manually tweak
Troubleshooting
"No valid schedule found"
Causes:
- Not enough available employees
- Conflicting constraints
- Over-specific requirements
Solutions:
- Check availability settings
- Reduce coverage requirements
- Remove or relax constraints
Uneven Distribution
Causes:
- Limited availability from some employees
- Skill requirements restricting options
- Preference settings favoring certain employees
Solutions:
- Enable strict fairness mode
- Review skill assignments
- Adjust preference weights
Slow Generation
Causes:
- Very complex requirements
- Large team size
- Many constraints
Solutions:
- Simplify initial requirements
- Generate in phases (week by week)
- Reduce optional constraints
Integration with Other Features
Works With
- Availability Management - Respects all availability settings
- Shift Templates - Can apply templates automatically
- Time-Off Requests - Considers approved time off
- Skill Tracking - Matches skills to requirements
Feeds Into
- Notifications - Alerts employees of assignments
- Time Tracking - Schedules become trackable shifts
- Reports - Schedule data appears in analytics
Best Practices
- Keep data clean - Accurate availability = better schedules
- Start simple - Begin with basic requirements, add complexity gradually
- Review before publishing - AI is smart but not perfect
- Communicate changes - Use built-in notifications
- Learn the patterns - The more you use it, the better results you'll get
What's Next?
- Employee Availability - Set up availability correctly
- Shift Swapping - Handle post-schedule changes
- Notifications - Configure schedule alerts