> 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/production-optimization/data-quality.md).

# Data quality

Before ENLYZE can calculate reliable metrics such as OEE or the [Maximum Demonstrated Speed (MDS)](/en/concepts/performance/maximum-demonstrated-speed.md), the underlying data has to be sound. The **data quality card** shows you, for each production run, how reliable the data basis is and what causes poor data quality.

## The data quality card

You find the data quality card in the **production run details**. It summarizes the data basis of a production run in three values:

* **Data quality**: a percentage that rates the overall data basis of the production run. A low value is highlighted in red.
* **Missing machine data**: the share of the production run's runtime for which no machine data is available for the performance value, shown as a duration and a percentage.
* **Overlapping booking**: the share of the production run's runtime during which another booking runs on the same machine, shown as a duration and a percentage.

<figure><img src="/files/9WvI8kry940P553F1vyy" alt="Data quality card in the production run details"><figcaption><p>The data quality card in the production run details</p></figcaption></figure>

## How data quality is calculated

Data quality combines two factors:

* the **data coverage**, the share of the production run's runtime for which machine data is available, and
* the **overlap share**, the share of the production run's runtime that is overlapped by other bookings.

```
Data quality = data coverage × (1 − overlap share)
```

A production run reaches 100 % only when machine data is available for its entire runtime and no other booking falls within its time period.

## Why data quality matters

ENLYZE assigns every machine data point to the production run that is active at that moment. This is the only way to determine which product was manufactured and at what performance.

Two problems disrupt this assignment:

* **Missing machine data** means that no measurements are available for part of the runtime. That period cannot be evaluated.
* **Overlapping bookings** mean that two production runs claim the same period on the same machine. ENLYZE then cannot unambiguously determine which product is being manufactured, and some metrics (such as the MDS) cannot be calculated for that period.

{% hint style="info" %}
Low data quality is not an error in ENLYZE but a signal about the data basis. Overlapping bookings originate, for example, from incorrect bookings in the MES or ERP system; ENLYZE only surfaces the errors.
{% endhint %}

## In this section

* [**Analyze overlapping production runs**](/en/production-optimization/data-quality/overlapping-production-runs.md): See which production runs overlap in time on a machine, and how.