How the manufacturing industry can unlock the value in data

The manufacturing industry continues to suffer as a result of the disconnect between data collection and data usage, hindering or altogether eliminating valuable use cases.

Despite the massive amounts of data collected, the industry continues to be a laggard amongst its peers, showing the lowest levels of task automation in data management.

With so many connected assets, manufacturers see the value in collecting operational data. However, the raw data collected is not providing the solution it was intended to offer; that of a decision-making engine.

Much of this has to do with the fact that data, in its raw form, is nearly useless, or at the very least, offers support for only the most simple use cases. These “primitive” use cases, which are more suited to less-data-mature organisations, are also paramount to building a foundation for rapid and continuous value creation.

But first, why is it so hard for manufacturers to get usable data?

A skills gap and data silos

Insufficient skills and capabilities is the top challenge to capturing value from manufacturing data, according to a survey conducted by the Forum and BCG.

Manufacturing continues to struggle from both an ageing workforce with quasi-tribal knowledge and a difficulty in finding new workers with the right technical chops.

In addition, data silos arise in many forms, both horizontally and vertically within organisations. Data can be locked in machines, have a software dependency, be department-siloed, or stuck in “time” – outdated due to manual collection.

Manufacturing equipment is complex, resulting in hundreds of distinct data points that change constantly. To provide effective tools for analysing data across these distinct systems, the data must not only be un-siloed but also transformed into a common data model so it can be understood.

Then, legacy equipment makes it more difficult to capture accurate, usable data from machines. In addition, these assets tend to be complemented by a legacy infrastructure and cloud reticence. As more assets and systems are connected, the infrastructure crumbles.

A solution? Incremental value use cases with data maturity

As manufacturers continue to overcome the challenge of collecting and using data effectively, they can work towards incrementally valuable use cases. These more advanced use cases are the result of a more mature data infrastructure that supports the collection, transformation and accessibility of data.

Furthermore, advanced use cases must be supported by an aligned ecosystem of solution providers, rather than a simple point solution or reliance on internal resources. This ecosystem includes machine builders, service providers, technology solutions and system integrators.

Each partner offers unique perspectives and capabilities that drive complex use cases of far greater value than the more simplistic ones.

A simple use case exemplifies how manufacturers who are early in their data journey can begin using data:

Real-time measuring and monitoring

Being able to identify production challenges accurately and having the ability to make this information accessible and visible to those who make decisions is the first step to mitigating problems. Think of this as stage one in data usability.

Having overcome the challenge of initially collecting accurate, real-time data from equipment, manufacturers have instant visibility into production performance. Even if they choose to do nothing more than monitor the stream of data coming in, there is immediate value to be had.

Take SilencerCo for example, a manufacturer of firearm suppressors. Management was interested in capturing real-time data from machines in order to set realistic expectations via baselines as well as improve decision-making.

The live data pulled from their equipment and operators is not only displayed on the shop floor but also accessible remotely, allowing management to react accordingly, while giving operators a sense of accountability.

Qualitatively, the instant visibility of data gave SilencerCo answers to their most basic, yet imperative, questions. It also gave them the opportunity to drill deeper, analysing the data to identify the root cause of a primary downtime event.

With data visibility, SilencerCo achieved an 8% increase in machine utilisation, a 200% improvement in good part production, and eliminated 11,500 hours of unplanned downtime.

Algorithms, predictive analytics and automation

Enter BC Machining, a manufacturer in North Carolina offering aluminum machining services to clients in diverse industrial niches. It is a more advanced data use case. However, the goal remains the same: enhance decision-making, and make decisions faster, even autonomously.

With a system in place to support data collection and visibility to production problems, BC Machining wanted to take further advantage of their data but they were beyond stage one of data maturity, with an established infrastructure to support their data and use it effectively.

However, BC Machining continued to run equipment at 200% capacity, constantly breaking their machine tools in the pursuit of profit, and accepting the resulting cost of scrap parts and time lost as a byproduct of success.

They required a more complex solution that would enable them to collect equipment data at a higher frequency (1kHz). This influx of new data points provided BC Machining with the ability to develop an algorithm to predict when a machine tool would fail. Furthermore, the problem is communicated to the machine control in order to stop the machine automatically.

This advanced use case, employing predictive analytics, has been able to eliminate nearly 100% of scrap parts, as well as the time operators spend sorting scrap. Operators and machines can in turn spend more time producing quality parts. The result has been $72,000 in annual savings per machine.

Industry 4.0 has made a promise of revolutionising manufacturing with advanced solutions, but to get started with data, manufacturers must first make sure they can collect accurate equipment data and make it available to decision-makers.

Only then can they unlock rapid value creation, while delivering the opportunity to deploy advanced data use cases with an ecosystem of partners in the future.