> For the complete documentation index, see [llms.txt](https://docs.enlyze.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.enlyze.com/en/integrations/grafana/advanced-api/03b-advanced-variables.md).

# 04 Advanced Variables

In this tutorial you'll build a dashboard that shows OEE metrics and production runs per machine. Panels repeat automatically for each selected machine:

<figure><img src="/files/ZFgJwS6k2eyuuJmIc4kF" alt=""><figcaption></figcaption></figure>

## What you'll learn

* Multi-select and "Include All" for variables
* Automatically repeating panels (Repeat)
* Hidden variables for time rounding (OEE API)
* Per-machine OEE with POST queries
* Brief overview: time series data via the API

## Prerequisites

* [Working with Variables](/en/integrations/grafana/advanced-api/03-variables.md) completed
* Understanding of dashboard variables, JSONata, and API queries

***

## Multi-select and Include All

In [Part 1](/en/integrations/grafana/advanced-api/03-variables.md) the `machine` variable held a single value. For a dashboard showing multiple machines at once, you need multi-select.

1. Create a new dashboard with a `site` variable as in Part 1
2. Create a variable `machines` (plural) with these settings:

| Setting       | Value      |
| ------------- | ---------- |
| Variable type | Query      |
| Name          | `machines` |
| Label         | Machine    |
| URL           | `machines` |
| Root selector | `$.data`   |

3. Add under **URL Query Params**:

| Key    | Value     |
| ------ | --------- |
| `site` | `${site}` |

4. Columns: `name` (Name), `uuid` (UUID)
5. Under **Selection options**:
   * Enable **Multi-value**
   * Enable **Include All option**
   * **Sort**: Alphabetical (asc)
   * **Refresh**: On dashboard load

You can now select individual machines from the dropdown or choose **All** for every machine at the site.

{% hint style="info" %}
Note the variable name `machines` (plural). This signals that the variable can hold multiple values and avoids confusion with the single-value `machine` variable from Part 1.
{% endhint %}

***

## Repeating panels (Repeat)

Instead of creating separate panels for each machine, let Grafana repeat panels automatically.

### Configure a row with Repeat

1. Create a **Row** (click **Add** > **Row**)
2. Click the gear icon on the row
3. Set **Title** to `${machines:text}`
4. Under **Repeat for**: select the `machines` variable

Grafana now creates a separate row for each value of the `machines` variable. Within each row, `${machines}` is replaced by the UUID of the respective machine.

<figure><img src="/files/ZFgJwS6k2eyuuJmIc4kF" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
Panels inside a repeat row automatically use the current iteration value. While `${machines}` normally contains multiple UUIDs, inside a repeat row it resolves to exactly one UUID per iteration.
{% endhint %}

{% hint style="warning" %}
You can only edit the **first** repeated element — whether row or panel. The remaining copies are automatically regenerated from the first element on every refresh.
{% endhint %}

***

## Hidden variables: time rounding for OEE

The `productivity-metrics` endpoint expects timestamps aligned to 10-minute intervals. Grafana's built-in time variables (`${__from}`, `${__to}`) return arbitrary timestamps. The solution: custom variables that round the selected time range to 10-minute intervals. Since these variables are only relevant for queries, they are implemented as **hidden variables** — they don't appear as dropdowns in the dashboard. If needed, you can include their value using `${start}` or `${end}` in panel titles for transparency.

### Create the start variable

1. Open **Dashboard Settings** > **Variables** > **+ New variable**
2. Configure:

| Setting       | Value                       |
| ------------- | --------------------------- |
| Variable type | Query                       |
| Name          | `start`                     |
| Hide          | **Variable** (fully hidden) |
| Source        | Inline                      |
| Data          | `{"time": ${__from}}`       |
| Root selector | see below                   |
| Refresh       | **On time range change**    |

3. Root selector (JSONata expression):

```
(
  $time_millis := $number($.time);
  $time_rounded := $time_millis - $time_millis % 600000;
  $fromMillis($time_rounded)
)
```

`600000` milliseconds = 10 minutes. The formula rounds the timestamp down to the nearest 10-minute interval.

### Create the end variable

Create a second hidden variable `end` with the same settings, but:

| Setting | Value               |
| ------- | ------------------- |
| Name    | `end`               |
| Data    | `{"time": ${__to}}` |

{% hint style="info" %}
**Refresh: On time range change** is critical here. Without this setting, `start` and `end` would only be calculated when the dashboard loads and wouldn't update when you change the time range.
{% endhint %}

***

## Per-machine OEE with Repeat

Now you'll combine repeat rows with the `productivity-metrics` endpoint.

### OEE Stat panel

Create a **Stat** panel inside the repeat row:

| Setting       | Value                                       |
| ------------- | ------------------------------------------- |
| Method        | POST                                        |
| URL           | `machines/${machines}/productivity-metrics` |
| Root selector | `$.data`                                    |

Expand **Parsing options & Result fields** and add a **Column**:

| Selector             | Title        | Type   |
| -------------------- | ------------ | ------ |
| `productivity.score` | productivity | Number |

Expand **URL options** and configure the request body:

| Setting           | Value                                                           |
| ----------------- | --------------------------------------------------------------- |
| Body Type         | Raw                                                             |
| Body Content Type | **JSON**                                                        |
| Body Content      | `{"datetime_ranges": [{"start": "${start}", "end": "${end}"}]}` |

{% hint style="warning" %}
Make sure **Body Content Type** is set to **JSON** (not Text). The ENLYZE API rejects requests with `Content-Type: text/plain` (HTTP 422).
{% endhint %}

Configure the panel:

* **Unit**: Percent (0.0-1.0)
* **Color mode**: Background
* **Thresholds**: Red (base), Yellow (0.6), Green (0.8)
* **Value Mapping**: `null` → "NO DATA" (Grey)

### Why Columns instead of Root Selector?

It would be tempting to extract the score directly in the root selector, e.g. `$.data[0].productivity.score`. This works as long as the API returns data. However, if a machine wasn't running during the selected time range, `$.data` is an empty array. Accessing `[0]` then fails and Grafana shows an error instead of an empty panel.

The solution: set the root selector to `$.data` and extract the value via a **Column** (`productivity.score`). If there's no data, the column returns `null`. You can then display this `null` value as "NO DATA" using a **Value Mapping**.

### Additional OEE panels

Create panels for the remaining metrics using the same pattern:

| Panel        | Root selector | Column Selector      |
| ------------ | ------------- | -------------------- |
| Availability | `$.data`      | `availability.score` |
| Performance  | `$.data`      | `performance.score`  |
| Quality      | `$.data`      | `quality.score`      |

Configure a value mapping for `null` → "NO DATA" in grey on all panels. Not all machines provide quality data, and machines without production in the selected time range show "NO DATA" for all metrics.

<figure><img src="/files/8bqyQ8tQ69vcpt6V3sVr" alt=""><figcaption></figcaption></figure>

### Production runs table

Add a **Table** to the row showing the production runs for the respective machine:

| Setting       | Value                                                                     |
| ------------- | ------------------------------------------------------------------------- |
| Method        | GET                                                                       |
| URL           | `production-runs`                                                         |
| Query Params  | `machine=${machines}`, `start=${__from:date:iso}`, `end=${__to:date:iso}` |
| Root selector | `$.data`                                                                  |

Add columns: `start` (Start, Timestamp), `end` (End, Timestamp), `production_order` (Order, String), `productivity.score` (OEE, Number).

***

## Excursus: time series data via the API

For time series data (e.g. temperature curves, throughput) the ENLYZE API offers the `timeseries` endpoint:

```json
POST /timeseries
{
  "machine": "uuid",
  "start": "2025-12-08T00:00:00Z",
  "end": "2025-12-09T00:00:00Z",
  "variables": ["variable-uuid-1", "variable-uuid-2"],
  "resampling_interval": 60
}
```

The `resampling_interval` parameter (in seconds) reduces the data volume. Without resampling, queries can return very large datasets.

**Limitations:**

* The API paginates responses (max 100 entries per page). The Infinity plugin only supports automatic pagination experimentally.
* For time series visualizations, the ENLYZE data source (the dedicated Grafana plugin) is the better choice, as it handles pagination and resampling automatically.

***

## Tips

* **Variable Refresh**: "On dashboard load" for master data (sites, machines). "On time range change" for time-dependent calculations like the rounded timestamps.
* **Variable order**: The order in Settings determines the evaluation order: `site` → `machines` → `start` → `end`.
* **Debugging Repeat**: If repeat panels don't appear, check that the variable is correctly selected under **Repeat options** on the row and that it contains multiple values.
* **POST Body**: The body is specified as a string. Variables like `${start}` are replaced before sending. Ensure correct JSON syntax with quotes.

***

## Next steps

* [**Production Status Dashboard**](/en/integrations/grafana/production-dashboards/04-production-status.md) — Constant variables and live status displays for multiple machines


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.enlyze.com/en/integrations/grafana/advanced-api/03b-advanced-variables.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.