Suppose we have a massive amount of sensor data available. Good right?
Yes… but there is still a caveat. A critical concept which is unknown to or misunderstood by many at both IT as well as OT side: that concept is context.
Context is what anchors digital data to the physical world. It transforms raw numbers into real-world meaning by linking sensor values to products, batches, machines, and events. It’s how data becomes a story.
Fundamentally, all data has context, eg. temperature sensor data. By adding more data we can expand the context. Context starts really simple: just a tag name, maybe in combination with a description and the units of the measurement (eg bar, m3/h or °C) is already context.
But that context needs to be expanded. Expanding is important to ask more/more complex questions to data and thus leverage the answers/insights.
The key in expanding the context lies in integrating data sources.
For example combining data from a process historian (a database optimized to store sensor data from the SCADA or PLC systems), with a Manufacturing Execution System (which typically holds product, recipe or batch information), and adding context to the long, cryptic lists of raw tags.
This is where the magic happens. Suddenly, the data becomes information: understandable and actionable.
Data users across your organization now have access to meaningful information they can use to solve problems and make decisions with confidence. Without the need to manually integrate data in a spreadsheet (which is not only very time consuming but also extremely error prone) or the need to first make sense of cryptic tag names like L15.B1.T01A.PV.
Introducing meaningful context is a major leap forward for many companies. One that’s often talked about but rarely achieved. It’s not just about having the right tools; it’s about making data accessible and meaningful. By adding context to raw operational data, organizations empower internal teams (who already understand the processes and problems) to explore and solve business challenges more effectively. Whether it’s process engineers, reliability experts, or data scientists, they can now act on data that reflects the real world more clearly and intuitively.
Context comes in several forms. Asset context explains where sensors or equipment sit within the physical plant, making cryptic tag names understandable and usable. Production context connects data to specific products, batches, and shifts, allowing for insights into quality and efficiency. Maintenance context links performance with repair history, revealing patterns that support predictive strategies. Other valuable layers include financial context (costs of energy or raw materials) and quality context (how parameters affect final output):
Asset Context
Asset context is the foundation for understanding how physical components in your plant are interconnected. It transforms cryptic tag names into meaningful information, providing clarity about where a sensor or device is located and how it fits into the larger operational system. It is the digital representation of your physical assets. This context is often sourced from engineering or master data systems, where detailed information about equipment and infrastructure is stored. For example, in the cookie factory, asset context could reveal that a temperature sensor is installed in a specific section of the baking oven, which itself is part of a series of ovens within the plant. With this context, operators can immediately identify and address issues at a granular level, ensuring smoother operations and faster troubleshooting. Exposing asset context to all data users is something that has been technically possible in many tools for many years, but rarely used in the past. Too many still believe that it is enough to understand their cryptic tag names and even see it as a ‘badge of honor’ to know them all. Now, that might be fun and true, but it makes the data simply inaccessible for those who do not.
This is where the ISA95 Hierarchical Model(Enterprise > Site > Area > Line > Cell) comes into play
Production Context
Production context bridges the gap between raw data and the manufacturing process it represents. It ties data to specific products, batches, materials, and even the teams or shifts responsible for production. This context is typically stored in Manufacturing Execution Systems – sometimes directly in an ERP – and is invaluable for answering questions like, What product was being made? Which batch did this data come from? Who was overseeing production at the time? For instance, in our cookie factory, production context allows us to link the start and end times of a batch with the temperature data from the ovens. This makes it possible to calculate the average temperature and total gas consumption for a specific batch, providing actionable insights to optimize both quality and efficiency.
ISA-88 provides a standard framework for modeling and controlling batch processes, including recipes, equipment, and procedural control.
Together they align enterprise-level planning with batch execution.
Maintenance Context
Maintenance context adds another layer of understanding by highlighting the relationship between equipment performance and maintenance activities. This one is overlooked by most companies today as it requires them to have a digital representation of their maintenance activities and can typically be found in ERP or more dedicated CMMS or Computerized Maintenance Management Systems (or on paper or in that one custom Access Database developed 19 years ago). It helps to correctly track KPIs such as Overall Equipment Effectiveness (OEE) and identifies patterns that may link unplanned maintenance events to specific process conditions. This context is particularly powerful when combined with structured, timestamped data-something that is still a challenge in many facilities where technicians rely on pen and paper. For example, analyzing maintenance context in a cookie factory might reveal that certain temperature settings in the oven lead to increased wear on components, allowing for proactive adjustments and reduced downtime. Accurate and consistent maintenance context enables organizations to shift from reactive to predictive maintenance strategies.
Other Potential Contexts
Beyond these core contexts, additional dimensions can bring even greater value to your operations. Financial context, for example, integrates cost data such as energy prices or raw material expenses, enabling precise calculations of production costs. Similarly, quality context connects process parameters to product outcomes, shedding light on how factors like temperature or mixing time impact the quality of the final product. These contexts allow for deeper analysis and optimization, ensuring that decisions are informed by a comprehensive understanding of both processes and outcomes.
By expanding these contexts, organizations can transform scattered data points into a cohesive, actionable narrative. This is the step that turns raw data into meaningful insights, laying the groundwork for smarter decision-making and a truly data-driven culture.