How to Get Contextualized Data Where It Matters
When it comes to data collection, who are you really serving?
That objective often gets lost amid the OT/IT alignment discussions. Anyone who has embarked on a digital transformation project is likely familiar with the data silos that exist between their OT and IT departments. But we don’t spend enough time talking about how to make that data usable for the line of business. Our line of business colleagues (and their systems of record) are the ultimate customer.
The use of IoT-enabled devices is increasing the availability of operational data. IDC has projected there will be 41.6 billion IoT devices in the field generating 79.4 zettabytes of data by 2025. These devices include machines, sensors, and cameras as well as industrial tools. To truly make that data usable, we need to merge this data with information from other systems and provide context for line of business users. In an industrial environment, these users include quality, maintenance, engineering, R&D, regulatory, and product management.
In the past, these functional areas received information through weekly and monthly reports. In some rare cases, the functional systems were integrated with custom coding and manual data cleansing. That’s not an easy feat when you consider the number of disparate OT applications running without standard architectures, tags, or naming conventions.
It’s a time-consuming process that can lead to failed digital transformation strategies. Consider that 36% of IT and operations professionals responding to an IDC 2021 Future of Operations survey indicated that they failed to reach a return on investment for their machine learning (ML) or artificial intelligence (AI) implementations. (Source: IDC Analyst Brief, How IT Can Effectively Build and Scale Operational Data Pipelines, May 2022.)
IT Pain Points
Cumbersome integration efforts put further stress an already-strained IT department. IT professionals are not industrial data experts. They get caught between the desires of operational decision makers and the capabilities of process technology teams. With IT workers in short supply, you can’t afford additional turnover due to burnout and low morale. In fact, 75% of IT executives responding to a 2021 Gartner survey agreed that talent availability was a main adoption risk factor for the majority of IT automation technologies.
It’s easy to see how talent availability can impact Industry 4.0 deployments when you think about the inherent challenges with digitalization initiatives. As IDC noted in the aforementioned Analyst Brief, “Operations data is unlike most enterprise data, as many organizations have discovered when they attempt to apply traditional IT data operations tools and techniques to it.” IT teams must overcome several hurdles for successful IoT implementations. Some of these barriers include:
- Multiple systems and data formats at the OT level, including process historians, which may have varying site-to-site configurations.
- Existing data collection may include a mix of manual logging and technology-based processes. Also, experienced workers may use tribal knowledge to execute various tasks, leaving IT with undocumented processes to reference.
- Unlike traditional back-office applications, operational data can’t be easily integrated through APIs.
- Operational data is a mix of structured and unstructured data sets and unique data types, which may include real-time machine data, transactional data, or time-series historical data, for example. Machines generate this data at a rate and volume that’s often unsuitable for traditional IT data storage systems.
- Operational processes change much more frequently than enterprise job functions, which creates problems when systems are connected via hard-coded or direct API integrations.
The DataOps Advantage
To overcome these hurdles, IDC’s Jonathan Lang (Research Manager, Worldwide IT/OT Convergence Strategies) suggests that organizations need “a purpose-built data abstraction layer to act as an operational data operations pipeline and transact between OT systems and the cloud, where a bidirectional connection to the line of business is created.” You also need the ability to deliver source data to consuming applications independently without the need for unwieldy, direct connections.
At HighByte, we refer to this abstraction layer as Industrial DataOps. The DataOps layer is best deployed at the edge to securely collect data and contextualize it into standard models for distribution across on-premises and cloud-based applications. Models correlate the data by machinery, process, and product and present it to the consuming applications in a single payload—in the format and frequency they require. A DataOps solution delivers contextualized data where it matters: In the hands of IT and line of business professionals. This is a critical step forward for organizations as they look to leverage operational data and maximize the ROI from their digital investments.
That value will be realized through line-of-business success and productivity improvements within the IT department. Download the IDC Analyst Brief, “How IT Can Effectively Build and Scale Operational Data Pipelines” to learn more about this topic.
About the author
This article was written by Torey Penrod-Cambra, the Chief Communications Officer of HighByte, focused on the company’s messaging strategy, market presence, and ability to operationalize. Her areas of responsibility include marketing, public relations, analyst relations, investor relations, and people operations.