Starting a Digital Manufacturing Plan, Part 1
What is currently on the market that will help a company get all the great things the IoT and Industry 4.0 are said to provide? This is probably one of the most damaging questions that lead to the most failures in a digital manufacturing plan. It’s like a chicken-and-egg problem: You need data to make informed decisions, but you need to make an informed decision on what data you need to collect to make that happen. This series will share how to start a successful digital manufacturing plan that won’t involve any products but, if done right, will leave you knowing exactly what products will yield success for your specific applications.
First, support needs to come from the top down. If the higher levels of a company support change, everyone under them will be more likely to work towards it or voice thoughts. However, often the top-level of executives or managers might not know details of the problems hindering the company’s growth. Therefore, change should happen from the bottom up.
When we think of a foundation, we think of something that is set in stone and difficult to change. Incumbent inertia seems inherently built into the word foundation, but industry needs to be flexible, scalable, and consider future-proofing.
The better you understand that a digital plan is only a representation of reality, the easier it is to see the importance of having accurate models. Without accurate models you might only increase failure. (Source: University of Alabama, "Approaches to Modernism")
How can we build off something that isn’t tangible and needs to adapt quickly? Let’s remember the digital world is only a model of reality. While everyone keeps talking about data, keep in mind what data means. The above image is not a pipe, but an image of a pipe. A solid foundation in digital planning comes from this idea. If you understand what the digital model represents in reality and what the numbers on the HMI mean, and you’re listening to the story being told to you by the data, you can have a dynamic, robust foundation.
There is a lot of hype and money being thrown around; we see companies connecting things and reading case studies. But while the new Industrial Revolution seems good for business, and it can maintain a competitive edge, I also hear Aidan Quilligan the global lead of Industry X.O Practice at Accenture, saying “78% of companies that tried to scale its digital plan fail to make expected earnings.” With this in mind, I thought it is time to get back to engineering fundamentals. The following is a simplified Six Sigma DMADVV format in an even simpler who, what, where, when, how format. There is a link to a free downloadable template at the end of the article that follows this format.
Before you start, make sure you first define the problem. “You have to find the pain point and what is needed to alleviate it. If there is no problem, there may be no reason to invest money into what could end up being bells and whistles,” says Joel Young, CTO of Xirgo Technologies.
“There must be a need; whether it’s for safety, cost reduction, or to improve efficiency, a need is the first step,” adds April Ankrum, senior product manager at TURCK.
Whether you are looking at this article on account of production problems, needed improvements, or curiosity, make sure pain points are documented and reviewed regularly.
Step 1: Who
Digital manufacturing offers many benefits, but it can be difficult to get everyone on the same page. A company’s greatest asset is its people. Use your people’s knowledge and expertise to determine the best solutions for them and the company.
Who is willing to work with you? Who is willing to sit at the table and listen? Step one isn’t concerned with what products are on the market and what can they do; step one is the data chicken-and-the-egg question. Companies have more data than they realize. People have an archive of data in their head, so start with your people’s knowledge. Recently we are seeing more data retiring from companies before it can be downloaded.
People are retiring with tribal knowledge—data—in their head. Should your digital plan also include things like AR/VR to save and download that tribal knowledge? If all you do if record it and can’t figure out how to use it to support training new people, AR/VR recording and training become more bells and whistles that look cool but will never be able to gain desired results.
Rockwell Automation and PTC show how software on a tablet can use augmented reality to show the location of a driver not operating correctly. By having knowledgeable software built from OEMs and tribal knowledge new employees could see where the problem is and access all the data, documentation history, and possibly augmented videos to troubleshoot problems or maintenance of equipment.
When trying to integrate new technology, everyone’s support and concerns are important. Communication on all levels are imperative. Building trust with your digital manufacturing plan will help open your people’s data to you that is imperative to building a more accurate digital model.
Who also includes users and customers. We hear about customization and personalization, but this isn’t just about having a monogrammed machine. It’s about doing business with a company being an end-to-end experience. You are selling an experience, not just a product. Again, this is about people and bringing intimacy and humanity into the process. Having a connection to your manufacturer, your order, knowing where it is, and what’s happening offers a lot value.
Know who is sitting at your table and who your customer is.
Step 2: What
What do the people at your table bring to it? What resources do they have, what are their goals, objectives, and concerns? Business managers want to see a return on investment. IT is concerned with security. In addition, IT may already be juggling a complex system with years of upgrades and getting legacy networks communicating with new networks. Workers seeing new technology may be concerned with job security. Finally, operations technology is concerned about production and quality.
We need to connect the analog before we connect the digital, so start by connecting with your people.
Step 3: Where
But where is the best place to start? Here is where we turn from the people and go back to engineering fundamentals. Where do you start? Where is the value? Where are available resources—power, Ethernet, Wi-Fi, etc.
Additionally, understand the limitations of the resources. For example, Wi-Fi was never meant for industrial internet. It doesn’t have the bandwidth or robust design an application might need. However, it might be okay to start with. Also, there is a lot of talk about 5G, but if you are waiting for technology to exist so you can adopt it you’re going to be operating behind the curve.
5G sounds great, but what are you using it for? Because there are a lot of applications where 2G or maybe even Wi-Fi will work. If you start taking advantage of prioritizing data, partitioning, and multiple networks today, when new technologies are available and you’ve analyzed them, you can upgrade if necessary.
One of the greatest tools will be to build an analog and digital workflow. Just like an analog flowchart to pin point bottlenecks and root causes of problems, know your digital flowchart. This will show you what type of data from what machines would give you the most benefit. If done properly, comparing the analog and digital flow charts should show the best place to start getting data, what type of data, and where it needs to go.
It is better to move effectively than fast. You need to know your analog-real world workflow before you go digital. If you don’t have an accurate real-world model; you might be using advanced digital technology to streamline and accelerate your failures. Generate an accurate, real-world, or analog flowchart then generate a digital flowchart. Find all the places you can get data from or where data would have the most value. If done well, combining the analog and digital flowcharts will act as a map as you expand your digital plan. Do not rush this step, and make sure the workers and people with their “boots on the ground” can see this map and add their own comments.
One of the biggest problems is that we are moving from machines that were specific to machines that are general or versatile. This is because things move too fast for machines or people to do or be one thing. We must be flexible. Survival of the fittest is not the strongest or the smartest—it is the company able to adapt to its environment quickest.
Where is the valuable data, where does that data need to go, and finally, where should the data be coming from? For example, if a camera sends an alert that indicates a problem in production. How much distance is between that digital error information to the analog process causing it, and how much product does that mean is potentially now waste? Your analog/digital map will help you see where you are getting data, and where you should be getting data to move effectively—not just fast.
Important Note: We are in an ever-changing world and many companies are designing a digital plan for a single point in time. Our universe and factories are dynamic and ever-changing. Stop planning and thinking that it is possible to integrate a new technology, then hit cruise control.
In part 2 you will see continue the DMADVV format with…
Measuring and Analyzing
- When you need data (digital and analog cycle rates)
- Why are you doing this, and review other options to prevent tunnel vision
Verify & Validate
- How can current products work with your people?
In addition, part 2 will cover what to look for in products, and how to handle scaling and future-proofing.