> For the complete documentation index, see [llms.txt](https://docs.myrspoven.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.myrspoven.com/how-the-ai-saves-energy/comfort-schedule.md).

# Comfort schedule

How myCoreAI relaxes comfort requirements during unoccupied hours.

Comfort schedule lets myCoreAI relax comfort requirements when a building is unoccupied. Holding office-hours comfort around the clock wastes energy on empty floors. Nights, weekends, and early mornings can run with wider temperature bounds and reduced ventilation without anyone noticing.

It comes in two variants: a **manual** schedule configured per building, and **Automatic Comfort Schedule (ACS)** that learns occupancy from the building's own CO₂ sensors. They can also be combined. ACS applies its setbacks on top of whatever bounds the manual schedule has in force.

## Manual comfort schedule

The manual variant is a weekly schedule of comfort bounds, configured per building in BackOffice. Operations defines time slots and the min/max bounds that apply in each. myCoreAI receives those bounds as time-varying limits and optimizes within them.

The manual schedule is explicit and predictable. Bounds change exactly when the schedule says they do, every week. That is also its limit. It does not adapt to how the building is actually used.

In myPortal 2.0, the [Occupancy](https://docs.myrspoven.com/use-myportal-2.0/occupancy.md) page shows the building's comfort schedule against its CO₂ pattern and AHU run times, and lets you align and edit its intervals.

## Automatic Comfort Schedule (ACS)

ACS removes the guesswork. Instead of a configured schedule, myCoreAI infers occupancy from CO₂ sensor data. When people arrive, CO₂ rises. When they leave, it falls. From that pattern, myCoreAI builds a weekly occupancy schedule per air handling unit (AHU) at 15-minute resolution and applies setbacks during the hours it has learned are unoccupied.

### How occupancy is learned

* myCoreAI analyzes roughly the last 90 days of CO₂ readings for the zones served by each AHU.
* For each weekday pattern, it finds the morning CO₂ rise, typically 05:00–11:00, and the evening fall, typically 14:00–22:00.
* It marks the time between those events as occupied and builds a weekly pattern per AHU.
* Schedules are recomputed continuously, so usage changes work their way into the learned schedule within weeks.

Different AHUs can get different schedules. A building where one AHU serves a 24/7 gym and another serves offices will not be treated as uniformly occupied.

### What changes during unoccupied hours

Three setbacks are configured per building. Each is independent.

| Setting                   | Applies to                  | Effect during unoccupied hours                                                                           |
| ------------------------- | --------------------------- | -------------------------------------------------------------------------------------------------------- |
| Indoor temperature offset | Indoor temperature bounds   | Bounds widen by ±X °C. The building may drift further from target.                                       |
| Supply-air offset         | AHU supply-air temperature  | Setpoint may drop by up to X °C.                                                                         |
| Ventilation offset        | AHU pressure/flow setpoints | Setpoint may drop by up to X%, or rise with a negative value, never below 30% of the configured maximum. |

A setting of `0` disables that setback. If all three are `0`, ACS still computes schedules but applies no changes.

### Fail-safes

ACS only relaxes bounds when it is confident the building is empty. When in doubt, it assumes occupied.

* If CO₂ data is missing, too short, or too flat to show a clear pattern, the schedule defaults to fully occupied.
* A weekday pattern must show at least 4 occupied hours to be trusted. Otherwise that AHU falls back to fully occupied.
* Ventilation never drops below 30% of the configured maximum, so baseline air exchange is always maintained.
* Individual signals can be excluded from ACS. This is useful for zones that must hold conditions around the clock, such as server rooms or labs.

{% hint style="info" %}
The learned schedule follows day-of-week patterns only. Public holidays run on the schedule of their weekday, so a holiday Monday is treated as a normal Monday.
{% endhint %}

## Interaction with other features

* [**AI Freedom**](https://docs.myrspoven.com/how-the-ai-saves-energy/ai-freedom.md) enables ACS with a guaranteed setback. Comfort bounds widen by at least 0.5 °C during unoccupied hours, even if no indoor temperature offset is configured.
* [**Adaptive min/max**](https://docs.myrspoven.com/how-the-ai-saves-energy/adaptive-min-max.md) can restrict how far heating-circuit supply temperatures may fall. Keep both values at `100` unless there is a specific reason not to.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.myrspoven.com/how-the-ai-saves-energy/comfort-schedule.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
