Digital transformation: An IIoT journey to analytics maturity
A fully-augmented factory allows process experts to manage by exception rather than by reaction. Once companies can manage processes by exception, they have created a smart factory, according to a company with a Control Engineering Engineers’ Choice Award product.
- Understand how to get operations under control with visibility into root causes.
- See how plants can be driven by data with communication gaps that slow improvements.
- Learn how to connect systems to get deeper operation performance insights and optimization, creating a smarter, digitalization approach to create an augmented factory.
Industrial process manufacturers have been capturing time-series data for years and storing it in databases (historians). While the collection process is essential, the raw data is meaningless for improving operations.
The question then becomes: How do you reach the full potential of captured data to improve operational performance? Companies just starting their digitalization journey go through a series of technical advancements over time. Each step represents a growing digital, and even analytical, maturity.
Plants move through this journey from a state of capturing data to analyzing it for continuous improvements. Companies can use the following model to go from capturing data to a fully functioning, augmented facility. Consider the model shown in the Figure.
Under control without visibility into root causes
Industries use supervisory control and data acquisition (SCADA) systems and distributed control systems (DCSs) for controlling complex production processes that signal the control room in case of upsets. This allows process experts to take corrective action. To get a deeper dive into and find the root cause of the problem, process engineers often use a spreadsheet to search for anomalies in the sensor-generated data.
Although the company has begun its “controlled factory” step of the model, industries must invest in a historian. The stored captured data gives process experts more insight through the trend client. For large amounts of data, though, the historians lack an easy way to quickly analyze useful data.
Driven by data, but communication gaps slow improvements
Companies have often turned to a data scientist to help analyze time-series data. However, communication gaps between data scientists and engineers slow the process. Engineers likely would spend significant time explaining the process so a data scientist can run the calculations.
Today, there are better ways to put the power of data analysis in the hands of operational experts. This is the next step in the digital journey.
Modern solutions use data science capabilities to predict potential root causes for process anomalies. When processes run optimally, engineers can visualize the process on a dashboard and use overlays to create ideal fingerprint profiles. In turn, this allows operational experts to monitor performance and create soft sensors. This is the “data-driven factory.”
Because engineers can make data-driven decisions, they can contribute to improving business objectives.
Connecting systems to get deeper operation performance insights, optimization
When engineers gain value from analyzing time-series data, their next step in a digitalization journey is to unlock data silos. Companies can get deeper insights into operational performance from the data residing in third-party systems.
True to its name and straight to the point, this is the “connected factory.”
Industries looking to connect systems will need assistance from their information technology departments. During this stage, information technology (IT) experts connect repositories and an analytics platform. This allows easy access to all contextual data in those applications. While the time-series data helps process experts get a broader perspective on operations, the contextual analysis enables them to get a much deeper look into production performance.
A fully-connected factory helps industries achieve business objectives. This includes reducing carbon footprint, controlling product quality, and increasing uptime.
A smarter, digitalization approach creates an augmented factory
The end goal of a digitalization journey is to create an “augmented factory.” Although the term sounds complex, the model has been used to explain other processes outside the industrial business world for centuries.
Think of an augmented factory as the ultimate smart factory, or one that’s fully embraced Industry 4.0.
As data insights and trends become available, process experts can begin to automate tasks and use machine learning (ML) and artificial intelligence (AI). Adding these layers allows them to decrease repetition, create anomaly detection models, and get prescriptive recommendations to take corrective actions. Process experts also can use the connected solutions to plan the optimal time for maintenance.
A fully-augmented factory allows process experts to manage by exception rather than by reaction. Once companies can manage their processes by exception, they have created true smart factory.
KEYWORDS: Augmented factory advice, Engineers’ Choice Awards
How are you using data analytics to improve operations?