Manufacturing

Data Integration is still the biggest hindrance to Industry 4.0

Share this post:

Manufacturing plant operators have been working for years on data integration infrastructure in the plant floor. Yet despite the effort, availability of the right data for use cases continues to be an unresolved challenge for the success of Industry 4.0 programs. So what’s the issue?

This isn’t because plants aren’t collecting data. In many cases, they are. But integration of disparate data sets is the  key challenge and it’s more complicated than you would think. To help overcome it, IBM Watson IoT is adding a new third-party offering from Telit: deviceWISE Factory Edition.

The challenge: leveraging hidden and vanishing data

Manufacturing plants have long collected and stored alerts, fault codes, product counts and some time series parameters in their historian. But while these data are enough for calculating KPIs, they may not be sufficient to implement many of the core predictive use cases that are foundational to Industry 4.0. Without predictive use cases, plant staff miss much of the basic value promised by Industry 4.0. They cannot easily be proactive in foreseeing issues to optimize yield, performance and equipment availability.

The good news is most of the data necessary for asset and production optimization are hidden in programmable logic controller (PLC) registers and machine controllers. However, only 10-20% of these data are actually captured in SCADA and historians – the rest simply vanish from PLC registers after each cycle.

Task cycle time is a good example of vanishing data that could add value if properly captured. This metric is an important predictor of machine and task performance in discrete manufacturing plants, but very few are recording it.

90% of data necessary for asset and production optimization is not captured

90% of data necessary for Asset and Production Optimization is not captured

Apples to oranges: juggling disparate systems

At first glance, the answer seems simple: manufacturers should just be collecting the vanishing data. So why don’t manufacturing plants simply extract the relevant data out of their PLCs, rather than allow them to vanish from cycle to cycle?

The reality is that a typical plant floor has multiple PLC and controller brands, many speaking different protocols. These disparate protocols do not allow for uniform data collection and storing methods, and the data become siloed as a result.

Resolving the difference with traditional methods is expensive and time consuming. We have observed that such data integration projects can take up to 60% of Industry 4.0 program budgets, and up to 60% of project time. Issues are compounded when we add the northbound integration of data with specific cloud solutions, making IIOT industry solution projects expensive and unattractive.

Connectivity costs up to 60% of IIoT projects and up to 60% of overall project time to implement

Connectivity costs up to 60% of IIoT projects and up to 60% of overall project time to implement

deviceWISE and Watson IoT have the answer

What if, instead, we could give you a different solution? This is where deviceWISE Factory Edition from Telit can help. This third party offering can connect to more than 25 PLC brands – all speaking various protocols – and directly map the PLC registers without writing a single piece of code.

deviceWISE can directly ‘peek-in’ to the PLC registers of over 25 PLC brands with specialized device drivers, and communicate with native protocol to collect the data from those registers. All without writing any code. With deviceWISE we can filter the data on the edge, have rule-based aggregation and interactively configure north bound integration to send the data to Watson IoT platform in near real-time.

Once the data are collected and filtered, they can feed into Industry Solutions like IBM Production Optimization – a new, AI-driven, cloud-based offering that delivers Industry 4.0 use cases. IBM Production Optimization leverages plant data to predict and pinpoint production losses, identify root causes, and prescribe optimized remedies to address these issues.

So, by combining the capabilities of deviceWISE with IBM Production Optimization, plant operators can put PLC data to work for predictive analytics and optimization use cases. The resulting insights, properly applied, can maximize throughput and eliminate production waste – driving down costs.  The result? Manufacturing plant operators are free to focus on the use cases that the data deliver.

Find out more about deviceWISE and IBM Production Optimization

If you’d like to learn more about IBM Production Optimization and deviceWISE, you may find these resources interesting:

The Offering Leader and the brain behind IBM Plant Performance Analytics offering

More stories
By Brian Massey on November 19, 2018

Smart, connected fabric: how AI, IoT and Mitsufuji create safer working environments

How can you keep your company relevant and vital in today’s connected world? Can you leverage technology and analytics to advance overall safety for employees? When do you use IoT technology to gain insights into employee’s activities in near real time? How do you provide notifications and preventative measures for their safety? These are questions […]

Continue reading

By Lauren Longhi on November 9, 2018

The top 5 takeaways from the Industrial IoT World conference

From October 29-30, companies from across industries gathered in Atlanta, GA for the Industrial IoT World conference to accelerate their projects “from inspiration to implementation.” As a leader in the IIoT space, IBM was invited to share its expertise. I joined IBM five months ago, and was excited to see our capabilities in person. Here […]

Continue reading

By Binny Samuel on October 22, 2018

Why data science experimentation drags out production optimization programs

Production optimization programs aim to increase production throughput and eliminate waste by leveraging data insights. But what happens when data identification – and creating analytical models relevant to use cases – becomes a job in its own right? Can out-of-the-box templates both support production optimization and reduce the need for data science experimentation? This blog explores […]

Continue reading