How to Pragmatically Accelerate Plant Growth with the IIoT

In a recent webinar, experts from PTC and Deloitte revealed the best practical IIoT practices to speed up manufacturing performance. They continue the discussion with this Q&A.

Recently, IndustryWeek hosted a PTC-sponsored webinar called Pragmatic Paths to Accelerate Manufacturing Performance with Industrie 4.0. The presenters, Kent Eriksson, a Senior Advisor in PTC’s IoT Transformation Advisory Practice, and Stephen Laaper, a Digital Supply Networks leader in Deloitte Consulting’s Strategy & Operations practice, had too much information about the best, most practical ways to adopt the Industrial Internet of Things (IIoT) didn’t have time to answer every listeners’ questions. It’s such a vital component to the future of manufacturing success, one that there is much to learn about, that we all felt these questions could not linger. So here they are, and even if you couldn’t attend, there is much to glean from these experts’ responses.

Q: Where should we draw distinctions between Industrie 4.0 and IoT, and Big Data and its connectors?

Eriksson: Industrie 4.0 for me is the connotation of the next century of technology development. We have seen glimpses of what will be possible. For me the two main topics are everything being connected and computer power moving to be smart, i.e. smart, connected products and operations as described on the innovation platform schematic architecture.

It's my opinion that a person cannot take action from the big data alone. If we get 30,000 readings from sensors shown to us, it will most likely mean nothing. Maybe visualization with trend curves can help, but often that does not help. It needs to be synthesized. For example, if machine learning synthesis indicates that Machine “X” will likely break down in five days due to the vibrating bearing. That provides three actionable data points that the MRO team can act faster on than seeing a big data lake of 30,000 readings.



Q: What are the new scenarios for global purchasing (digital foundry B2B service is the product) with suppliers and with their own organization?

Laaper: How progressive companies are viewing their Digital Supply Network within Industrie 4.0 is absolutely in an end-to-end fashion. In many cases, the predominant potential value comes from in the DSN, sharing information in real-time across the traditionally siloed and disconnected parts of the Supply Chain.

Q: How do you apply these concepts in smaller operations? Where funding is limited for automation and the systems infrastructure to help analyze processes digitally?

 Eriksson:The industry of Germany is very much smaller operations, what they call “Mittelstand.” The Industrie 4.0, RWTH Aachen Campus, etc. are actually primarily to help smaller companies, because they do not have the investment abilities of large companies. Smaller firms need to get going faster and more pragmatically.

One example is in the center of smart services in E4TC, we did help the oldest German family firm to make their close to 500 years old button machines smart and connected.

Q: Is there a formula that can be used to estimate on a broad basis the cost to move to this digital platform? For example, for every million in sales you would need to spend X% for X number of years to realize this digital transformation platform. Is there such a formula?

Eriksson: Generally, it’s much less than the return. The costs of collaborative robots and additive manufacturing are of course considerable. However, when you look on Digital Lean with a software platform, the investments are comparably small. The reason for Lean Thinking from Toyota was that they were in a very bad financial shape and did not want to be dependent on loans or others. That is what I call a “burning platform” by the way. Their situation made them strive for perfection, step-by-step, in a very cash and capital efficient manner.  Digital Lean with an innovation platform is exactly the same. Start small and have a clear proof of value of the first sprints, and they will basically self-fund your next steps.

Q: What are today's concepts, tools and practices to apply to existing intelligent system records so that critical info is surfaced quickly and clearly to appropriate Operations & Management layers without overloading the OT/IT systems?  This is essentially the equivalent to "Alarm Management" in complex Process Automation Systems, but likely with more latency.

 Eriksson: A very good question in the context of the schematic architecture of the innovation platform as described during the webinar. We talk about the innovation platform to wrap and extend. The functions you have today, like historians and SCADA watchdogs, that work well should be kept. However, most operations can improve to strive for perfection, a leading idea in Kaizen.

We believe there are role based interactions with room for improvement. Often when persons need information from several IT/OT systems, this is where an innovation platform will catalyze the digital transformation. It also allows right-time dispatch of information to the person that needs it, or even find out and propose/execute counter measures. The latter is what we mean with intelligent. A main thread is that we help the personas to get better performance from the top executives to the individual operators. How it will be done is case by case. If you take a look on most shop floors across plants in a company, you will find quite a heterogeneous mix of different hardware with various ages and vendors. We have seen that a large SW that tries to do all often do a poor job in this environment, hence our thinking on wrap-and-extend is very different making the software more modular and tailored to your specific situation.

Q:How should higher ed change their curriculum to get their workforce ready for AI   How do you see jobs changing in the next 10 years?

Laaper This really needs to be a two-fold solution in terms of the “type” of workforce that will be required.  On one hand, we’ll need a significant number of resources whom are highly skilled in analytics, engineering, and mathematics.   But balancing that, companies are also looking to augment their workforces with an increasing number of people with “soft skills”, many with degrees in physiology and humanities.  This balance is viewed as critical to the appropriateness and execution success of the AI solutions which will be developed as many of these solutions are related to enhancing and augmenting human performance as opposed to replacing it.

Eriksson: I see many universities in Europe ramping up the need of advanced analytics engineers. As mentioned during the Q&A, companies need more roles. Job in 10 years, in my opinion will be decided from policy makers. There is an abundance of opinions from AI creating more jobs than destroying to no traditional jobs at all. We are facing the larges challenge mankind have seen, but stay positive and remember the wisdom of the science fiction writer Arthur C Clarke: “The goal of the future is full unemployment, so we can play.”

Q: How advanced is the digital twin in the commercial aircraft industry?

Laaper: The digital twin is something that has been around the aircraft industry for quite some time (although we’ve only called it the “Digital Twin” in our more recent history.  The application of the digital twin is most prevalent in complex mechanical assemblies such as aircraft jet engines, however those same principals are being widely applied (along with the Digital Thread) for airframe assemblies as well.   We’ve helped to describe the story of an aircraft manufacturing who are using these principals in their new model design in a short video series which can be viewed here.

Eriksson: The term digital twin has a background in aviation from NASA quite long ago. We do think there have been technology limitations that today are gone by applying what we call PLM-720. It is basically having more digital twins for specific purposes. In the commercial aviation, we have many example, one concrete are hand held tools, like track and trace. We showed an example with Bosch at Liveworx 2015 Stuttgart on track-and-trace. You have a digital twin to a hand held tool. It knows there it is related to the aircraft frame and where in the installation sequence the operator us. It makes the right torque curve. It also registers the work performed. The quality step can be omitted since the digital twin has recorded a much more reliable result than any post checks. Now, operating airplanes and the equipment (like the engines) is more the same challenge as the next question.


GE's Smart factory Digital Twin Model

Q: In the commercial aircraft manufacturing industry, how will some second and all third-tier suppliers be able to adapt to digitalization/digital platform.

Stephen: The largest impact for 2nd and 3rd tier suppliers in many industries will be around the real-time sharing of data and information, not only about product specifications and the latest details on product engineering, but the sharing of the latest information sourcing, lead-times, quality, and production readiness.   As tier 1 suppliers are increasingly seeing this level of integration, it will flow downstream in the supply base.

Q: Could you give me an example where and how this IoT / Industry 4.0 can be applied in Food Processor companies (specifically in the manufacturing part of the supply chain)?

Laaper: While there are a number of applications for IoT/DSN/I4.0 in the Food & Beverage industry, some of the most prevalent we are seeing are around raw materials savings being realized by finer control of our input materials on production processes.   As we’re able to collection and process more and more data real-time from our production equipment, we can fine tune the processes to better achieve the expected quality/quantity, with a higher degree of precision.

Eriksson: It can be applied in multiple places. I discussed with food distributors in South Africa on the topic of addressing their challenges towards the customer, in reference to our retail offers. However, it can also be inside the process industry. We have examples of measuring the aggregated throughput on the packaging factory owner and even the machine packaging makers that want to provide smart services to those factories.


How can we get the papers from Harvard?

Eriksson: Please find your copies at

Q: Can 3D printing and modelling be done for samples with a fabric for quick submits of apparel designs?

Eriksson: I think it can or will be possible. We have commercial examples of making tailor made single unit pair of shoes, but the main volume is still off the shelf shoes. Patterns, colors and individual design options are possible, if it will be with 3D print, AR or other means for sales/prototypes can be discussed. See more on retail here.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.