Practical Industry 4.0

“Work smarter, not harder.” In 1712 the first productive steam engine was invented and improved until 1801 when the first steam-powered locomotive was driven through the streets of Camborne by Richard Trevithick. Between 1719 and 1764, three inventors improved weaving and spinning so that by 1785 Edmund Cartwright’s power loom could break onto the scene. Samuel Morse invented the telegraph in 1844 to allow quick communication over long distances, and in 2007 Steve Jobs revealed the game-changing iPhone. Innovation followed us across the centuries and will continue to push us into the future. We have seen three industrial revolutions come and go, so what exactly is the fourth, and when is it coming?

A quick Google search suggests that we are currently in the midst of Industry 4.0. Industry 4.0 focuses on cyber-physical systems, leveraging technology like internet of things, and networked big data. People and technology have never been more connected than we are today. We have wireless capabilities to gather information, send it to the cloud, and then pull it back down to our phones or computers when needed. Artificial Intelligence is getting more reliable, and we are capturing more data than before. But what good does that do for our business?

Scott Rogers, Technical Director at Noble Plastics, defines this new industrial revolution as the ability to elevate productivity and profitability through the use of data and analytics autonomously. Automation and connectivity were part of the third industrial revolution. Our ability to autonomously collect, analyze, and act is what Industry 4.0 is all about. “If we’re saying we are going to have an Industry 4.0 initiative, then some degree of autonomy has to be our goal. Otherwise, we’re just calling it something that it’s not.”

Mirriam-Webster defines automation as “the state of being operated automatically” while defining autonomy as “the quality or state of being self-governing.” Google may say we’re in the ‘midst’ of Industry 4.0, but we’ve only begun our trek through this new phase of innovation. Autonomy is not achieved through a set of rules that give rise to an IF-THEN, ELSE decision-making methodology; this is automation. In this case, we have already determined a limited number of outcomes based on rules, and the rules don’t change unless we change them. The discussion of autonomy requires a larger form than this N-sight allows, but let’s start with the basics; as we begin this journey, what are some of the important concepts and pitfalls to watch out for?

First Step: Gather Data and Context

Most of us are collecting some amount of data related to production or process performance. Equally or perhaps more important is the collection of contextual information. Intelligent decision-making requires context. Molding cycle time has little value without knowledge of the associated product. The data is even more valuable if we also know things like material, molding machine size, average wall thickness, etc. More context, more value.

If providing data and context is a good idea for our future intelligent and autonomous systems, it should also be a good idea for the autonomous systems we have in-house today; people. By making sure that our workers have usable access to multiple data sources with appropriate context through a single interface, they can work more productively today. We can also observe and characterize how our current intelligent systems react to the data received. In this way, our efforts provide immediate benefits while preparing for the future.

Next Step: Silos and Centralization

After collecting all this data, what do you do with it? Because we gather data using multiple systems from multiple vendors, it is natural that we have multiple repositories. The challenge lies in devising a strategy that facilitates the automatic and useful integration of data from disparate sources that are continually changing (e.g., versions and functionality). While challenging, we know this is possible, as evidenced by internet search engines.

There are sources that suggest the best solution to address siloing of data is to use a single provider for all information-based systems. While this may seem like an easy and attractive solution, the probability that one source will offer best-in-class performance and innovation across an I4.0 landscape is low. Conversely, there are more narrowly focused systems that suggest the value of their performance is high enough to preclude the need for integration. Noble’s goal is to create a fluid and integrated landscape that allows the use of best-in-class systems to drive manufacturing performance.

Integration and Analytics

Once a data collection and access strategy are implemented, how do we gain insights? In some cases, a simple graphical representation can give a boost to what we perceive and how long it takes that perception to happen. In other cases, the relationship between data and our desired results requires extensive calculations and may require data taken over an extended period. While cutting-edge efforts require less common expertise and computational requirements, there are Business Intelligence (BI) tools available that offer very useful capabilities with little or no training requirements. Notable platforms are Domo, Tableau, and PowerBI. For our purposes, Noble uses Domo. BI tools can also virtually centralize your data by pulling information from multiple platforms for analysis and display.

If you are ready to take the next step into machine learning, do some research into Industrial IoT platforms offered by companies like Microsoft and Oracle. Oracle’s IoT Intelligent Applications platform allows you to connect your data sources and build a wide variety of predictive analytics without writing any code.

Ultimately, you need a useful representation of your data available and accessible by your team and technology so that you can make the best decisions as early in your process as possible.

Autonomy

That sounds super neat, but how do we meet the autonomy requirement of Industry 4.0? Currently, Noble is like a lot of you. Our manufacturing systems are not autonomous, but we’re setting the groundwork to get there. We wouldn’t have our smartphones without the groundwork Samuel Morse laid out for Alexander Graham Bell, who made Steve Jobs’ 2007 reveal of the iPhone possible.

And that is the kicker; we need time. No matter how much we want it, no matter how hard we think about what an autonomous manufacturing industry looks like, it will take time. The good news is that our ability to track data is only getting better. That means our calculated insights will become more accurate. Once we have cracked the code on true AI-generated insights for autonomy, we’ll be ready. Don’t forget we still have landlines in the age of smartphones.

If you are interested in seeing more of our Industry 4.0-focused N-Sights you can submit your email to our mailing list!