# Smart Planner: AI generation settings

> Enable the Smart Planner and set how the AI builds your schedules: analysed history, legal constraints, data modules and default prompt.

Source: https://help.shyfter.com/en/settings/smart-planner-ai-generation-settings  
Category: Settings  
Updated: 2026-07-01

## What the Smart Planner does

The **Smart Planner** is Shyfter's artificial intelligence scheduling assistant. From your history and the settings below, it automatically generates an optimised schedule that you remain free to adjust. You describe your business in a prompt, the AI proposes, you validate.

Go to **Settings ⚙️ → Schedules → AI**. The screen is organised into three sections: **General settings**, **Constraints & rules** and **Modules & data**. Remember to save after your changes.

## General settings

- **Enable the Smart Planner**: enables the AI to automatically generate optimised schedules based on your team's needs and preferences. As long as this option is disabled, automatic generation is not offered in the schedule.
- **Default prompt**: the text that guides the generation. Describe your business and your habits (type of establishment, footfall, key positions). A **Generate a suggested prompt** button automatically offers a prompt suited to your account, which you can then adjust.
- **Only create free shifts**: the AI generates only unassigned shifts. It analyses the history (schedules, skills, sections) but assigns the shifts to no one. Handy to prepare a framework that you then assign yourself.
- **Analysed history (months)**: the number of months of history the AI examines to detect your habits. The longer the history, the more the proposals reflect your recurring patterns.

## Constraints & rules

- **Check availabilities**: excludes from the generation employees on leave, absent or unavailable.
- **Respect the hours/week from the profile**: limits the generated hours to each employee's contractual maximum.
- **Strict skills**: only assigns a shift to employees who have all the required skills.
- **Optimise costs**: favours the least expensive options during the generation.
- **Respect legal constraints**: applies the legal rules (minimum rest of 11 hours, maximum 6 consecutive days, weekly limit).
- **Use contractual schedules as fallback**: in case of insufficient data, relies on the employees' contractual schedule templates.

## Modules & data

These options tell the AI which dimensions to take into account to distribute and size the shifts.

- **Use sections**: places employees in their usual sections.
- **Use skills**: takes the employees' skills into account.
- **Use quotas**: calculates staff needs from the hourly quotas and the forecast revenue.
- **Base on revenue**: adapts the number of people scheduled according to the forecast revenue.
- **Use weather**: adjusts headcount according to the weather forecast and external events.
- **Affinity between people**: takes into account affinities between employees to favour good combinations.
- **Fair distribution**: distributes the shifts fairly among the employees.
- **Fair rotation (turnover)**: alternates the demanding schedules (evenings, weekends) among the employees.
- **Learn from corrections**: takes your past manual changes into account to improve future proposals.

> Several of these modules depend on external data: revenue and quotas assume revenue tracking (manual or via a connected POS), weather relies on the establishment's location. The more complete your data, the more relevant the generation.
