IIoT health check: 6 signs you need Industrial DataOps
“This story was originally published by HighByte
and has been republished here with permission.”
An executive for an industrial products company once told me even though his factories are full of similar equipment, he still struggled to access meaningful data from the machines. Each one of the plastic injection molding machines had a different way of presenting the data. That meant the company needed to customize coding for every piece of equipment to obtain meaningful insights.
It’s a common scenario in many industrial environments, where plants may have hundreds of PLCs and machine controllers on disparate machines generating operational data that is unintelligible to the data scientists who must make sense of it.
This is where Industrial DataOps comes in. It provides a way to standardize data using common models, or object-oriented approaches, to integrate and manage information coming from multiple sources. Here’s a closer look at the top six signs it’s time to consider an Industrial DataOps architecture for your company.
1. IIoT projects aren’t scaling because of data interoperability issues.
Translating and analyzing data from hundreds or even thousands of machines can leave you with tens of thousands of data points. Without a common protocol or standardized way to present the data, manipulating or handling each individual data tag can be incredibly time consuming. Engineers solved machine communication issues years ago with protocol translators but making sense of different data structures is another challenge. Modeling data for each system is inefficient and costly. There’s no simple way to manage individual machine models and roll them out across multiple devices and systems. Industrial DataOps resolves this problem with a standard set of models that combine multiple data points into a single flow.
2. You’re writing and maintaining custom scripts to leverage IIoT data.
Many manufacturers quickly ramp up IIoT projects only to find themselves mired in “technical debt.” Technical debt is a term that programmer Ward Cunningham coined in the early 1990s to describe a scenario in which people deploy software using the fastest, easiest code available with little regard for future technology needs. Unfortunately, IIoT projects often stall because manufacturers took this same approach and are constantly rewriting code each time they need to make a change on the factory floor like replacing machines, optimizing processes, or adding new product lines. Integrated systems that are not tied to standardized models are prone to disruptions due to cumbersome and time-consuming scripting. An Industrial DataOps solution that enables manufacturers to build standardized models is more adaptable and resilient to future changes.
3. Data scientists are spending more than 50% of their time finding, massaging, and preparing data for analytics.
A disconnect often exists between Operations Technology (OT) and the IT teams. In the past, PLC protocols, such as Modbus, were sufficient to communicate codes to the OT team. But a numerical code doesn’t give IT folks much information to work with. Data scientists shouldn’t have to waste time manipulating the data. Industrial DataOps makes the information available in a usable, self-descriptive format for data scientists so they can present it to diverse functions throughout the organization for strategic decision making.
4. The OT team is backlogged with data requests.
Another sure sign that it’s time to consider an Industrial DataOps solution is a backlog of data requests to the OT team from cross-departmental stakeholders to grant access to and explain machine data. It means the IT team isn’t receiving contextualized information in a timely fashion and that they’re overly reliant on OT personnel, who may have configured the systems, to explain what certain codes mean. Industrial DataOps automates the analytics process to free the OT team from manual data reports and to increase availability to data scientists.
5. You’re paying high, variable cloud storage fees for raw industrial data without a strategy for how the data will be used.
Pricey cloud-storage fees are particularly frustrating when you don’t have a clear vision for how to use the data. Many of these storage options come out of the consumer tech world and weren’t made for industrial-sized applications. Industrial DataOps was designed for industrial data needs. In an Industrial DataOps configuration, the data may be processed close to the machinery, in an on-premises data center, or in the Cloud depending on your unique requirements. The solution runs close to the device and feeds cloud applications only their required data and at the frequency or condition specified thus reducing high, unpredictable cloud storage fees.
6. You’re unsure who has access to operations data internally and externally, which creates security issues.
IIoT connectivity can create security concerns. Bringing more devices online may leave information vulnerable to hackers and other malicious threats. Allowing vendors direct access to machines may improve uptime but may introduce cyber risk. Exchanging data using the built-in security of OPC UA and MQTT reduces potential attacks. Also, taking a proactive approach of maintaining outputs by connection allows administrators to implement higher-level management and security than typical pub/sub broker architectures and open, unmanaged API access.
Launching an IIoT program without Industrial DataOps is like operating on only two legs of a stool. You may have the sensors and the analytical tools but without the ability to transmit and prepare the data, you’re going to fail. Industrial DataOps bridges the gap between the collection and analysis of the data so manufacturers can make more informed decisions.
If you’re struggling with one or more of the challenges discussed in this post, let’s talk. Schedule a demo with me to see HighByte Intelligence Hub in action and receive a personalized health check on your data architecture.