Eight Types of Value-Adding Data Context in Manufacturing

[Erkan Teskancan]

## Eight Types of Data Context that Add Value in Manufacturing

Manufacturing data is worthless without context. Here's crucial information on what to pay attention to in order to turn digital junk heaps into valuable insights.

It's not always easy to distinguish between data and junk. In 2013, an engineer accidentally threw away a laptop hard drive containing private keys, losing access to 8,000 Bitcoin worth approximately $751 million. This shows how even valuable data can seem like junk without context.

On the other hand, many data pools, databases, or historians contain plant data that is actually worthless but appears valuable. This is because all data without context is junk.

### Data Context Enriches Raw Data

Data context, or data contextualization, is the process of enriching raw data with additional information such as timestamps, identity information, physical location, and semantic information to make it more meaningful and usable. This transforms data from meaningless noise into actionable information.

Without context, it's impossible to trust, interpret, compare data, or automate decisions. Neither SCADA, ERP, nor business intelligence systems can function properly.

### Eight Key Types of Data Context

1. Time Context: Placing data in the correct chronological order is critically important. Without time, the sequence of events, cause-and-effect relationships, and predictive analyses cannot be performed.

2. Identity Context: Information about which asset, device, module, or sensor generated the data. More detailed identity information is needed as one moves closer to operational systems.

3. Data Lineage: Describes the data's source, collection method, path, and transformations. Important for audits and artificial intelligence validation.

4. Location Context: Physical and logical location information affects the meaning of data. For example, temperature changes in different plant areas can affect quality.

5. Semantic Context: What the data means, its function, and domain. Converts numerical data into meaningful terms.

6. Process Context: Information about which operation was being performed at the moment the data was collected. Enables root cause analysis and optimization.

7. Measurement Context: Numerical properties of the data, such as units, precision, calibration, and quality codes.

8. Organizational Context: The structuring of data across the enterprise, hierarchy, standard tag structures, and integration models.

### The Importance of Context

In the Industry 3.0 era, the importance of context was low; data like temperature was directly transferred to SCADA. However, with Industry 4.0, Smart Manufacturing, and Artificial Intelligence, data is integrated into all enterprise systems. Insufficient context leads to faulty analyses, inconsistent reports, slow troubleshooting processes, and misleading dashboards. In AI applications, this significantly damages the value chain.

### Data Context Providing Devices

Most industrial gateways on the market only collect and transmit data, without providing rich context. Some, however, offer context modeling capabilities:

• Basic Protocol Converters: Transport data but cannot add context.
• Edge Compute Frameworks: Provide programming capabilities for adding context.
• Industrial Data Hubs: Support complex context modeling.

Alternatively, Real Time Automation's RTConnect A-B PLC Historian product offers both data time-series collection and rich context. With easy installation, configurable storage up to 1 TB, and multicast protocols, it is an ideal solution for plant operations, maintenance, and process engineers.

### RTConnect A-B PLC Historian Features

• Tag capture and normalization from multiple A-B PLCs
• User-defined model population and instant publishing
• No subscription, license, or third-party software required
• Customizable data storage up to 1 TB
• Supports SQL, HTTP, FTP, WebSockets, USB, MQTT, and email protocols
• Direct integration with InfluxDB

Contextualized data enables faster analysis, accurate KPIs, efficient energy use, reliable data pipelines, and standard integration in manufacturing processes.