Earlier this year, McKinsey published a manufacturing study describing the future of American manufacturing. They predicted increased opportunities due to an uptick in demand, new technology, and value chain optimization. Though many news headlines forecast the ongoing decline of U.S. manufacturing due to globalization and technology, the years ahead may hold potential for manufacturing revitalization.
With shorter product life-cycles and more demand for choice and customization, there’s greater complexity for manufacturers to navigate than there was in the past. To address what this means for manufacturing in Central Texas, we looked at the first two main points of the study for deeper examination.
The first point reflects new opportunities and imperatives from rising demand and fragmentation, industry 4.0 technologies, and shifting value. The second point addresses three scenarios for 2025 that translate trends into opportunities, potentially boosting GDP by up to $530 billion over current trends.
To get a better understanding of what these technology challenges and workforce implications mean to the Texas manufacturing community, we sought insight from two experts. We spoke to Selene Crosby, previously part of the manufacturing resource team at the Texas Manufacturing Assistance Center (TMAC) program and a Lean Six Sigma expert, and Scott Parker, an engineering manager at Southern Folger Detention Equipment Company based in San Antonio. Here’s what they had to say about what the next few years hold for manufacturing in the Lone Star State:
What challenges do you see with adopting digital tools?
McKinsey calls this a prediction, but I think it’s here, and it’s been here for a while. Customers are expecting us to decrease our cycle time and incorporate more customization into the product. With digital on top of us— and we’re talking about the Internet of Things, machine learning and instant analysis—what it comes down to is how good is your data and information flow. If you think about it, I can’t make the right things if I don’t have all the information. Connecting my machine to a source of data isn’t going to help me if I don’t have the data— when I need it, from whatever source it comes from. I think this is always challenging for manufacturers. If you think about the way the process works, sales is collecting a set of information, and it goes to engineering, and goes back to sales for more data. McKinsey said it in their study: to capitalize on technology, companies have to start capturing, integrating and analyzing data flows from across operations and ecosystems.
I do agree that more and more customers want customized equipment that fit their needs. How manufacturers approach that needs to shift. Most manufacturers want to build the same products over and over. When they don’t get to build them (to a certain extent) they’ll charge customers an arm and a leg, and take a long time to do it. The market is changing so that can’t happen. Manufacturers have to respond in the shortest time frame possible and get the customer the product that meets their criteria as best as possible. The shorter your cycle time is, the better off you are, or the more margin you can make on your product. The thing that’s slowing everything down is the upfront work from sales or marketing to engineering to purchasing to the shop. Manufacturers need to make the upfront information flow match up with the speed of the shop.
How would tech help you solve/alleviate your needs?
The more accurately you’re cutting, shaping, tooling or assembling, the better. That’s always been true, but the challenge is that we tend to go buy an IT system (ERP and CRM) that promises to make data flow faster. But the system is just a tool. I know how to use a hammer, but that doesn’t make me a carpenter. I have to understand how data flows through my systems and processes before I even know what system I need or how those systems need to be interconnected.
Creating those data systems is hard work and ideally you’re designing the system to maximize how data flows through to where it needs to be. It’s not an easy fix. As it changes and customer demands change, robust processes keep it fresh and alive. Once you understand how data flows, you can think about advanced analysis and machine learning and all the advancements.
I believe in getting the most bang for your buck in what you use, information wise. Most of what we design is in a 3D CAD tool. But then we revert back to the way everything has been done, communication-wise, for 100 years. We provide a drawing on paper, make multiple copies for various groups, and they try to decipher what it means. There’s a back and forth with that.
Most companies steer away from robotics and 3D engineering data. Everyone thinks there’s too much variation in the parts supplied to the robot. There is a retranslation of the data from the source it comes from into what you can use in terms of robotics or AI. In my opinion, you have to build your system to take advantage of data. The rapid prototyping industry is one industry built to take advantage of that, but other than that the companies/industries are few and far between.
What tech capabilities do you look for that aligns with priorities?
What Lean Six Sigma would tell you is to start with the value stream map. It contains your physical process flows, but it also contains your information flows. People often start with the physical flows and under-emphasize the information part. They should have equal weight. If I was doing it for the purpose of understanding how digital could help me, I would flip that on its head and ask, “what information do I need to get, and when do I get it?” Map it back to the point where it’s available. Ask yourself how digital technologies can improve that.
Start with all the information you’re going to need to make it right the first time and the source. Does that come through the sales process? Engineering process? External data about inventory? Where is the source of that data element? What information is flowing here that I absolutely have to have and take it to a granular level? It’s no good for a machine to read your data if you don’t have data for it to read. So before we start upgrading machines, we need to map what they need to know.
Manufacturing should be moving more to a virtual world. The drawing process could be eliminated so you’re dealing with only 3D solid models or a representation of a CAD model. From that data laser cut or machined parts could be generated. Robotic welding or robotic assembly of parts could also occur from that same 3D data. We have already removed the majority of layout from the welding process, locating parts from the 3D model (i.e locating features, tab and notch, etc). If you locate parts in the 3D model, then all the engineers have to do is get the parts welded in place. You don’t even need a person to translate your designs from paper. Metal forming, machining and welding could flow through to assembly with minimal human interface. Taking the virtual world concept to the other end of the process, automation of the sales and design process could yield whole model and fabrication package based on minimal input from a salesman. The paradigm needs to shift away from how it’s been done since Henry Ford.
What does this mean for smaller manufacturing sectors?
It’s an interesting dynamic because we have the globalization of supply chains and a new more protectionist trade dynamic that are in play right now. Smaller manufacturers can benefit from the agility of being local. It can mean a couple of things: the need to understand the local market and the demand, Texas or regionally, as well as the larger more global market. They need to take advantage of the agility of having smaller organizations. That means understanding what adjacent market you can play in, and what strategic partnerships within your area you can sustain. That’s understanding where you can fit. A lot of people believe cycle time is the differentiator, and if you can get that to be faster than anybody else, and you’re local, then you’re ahead.
The small manufacturers need to understand how to turn something from a six to eight week lead time into a two week cycle time. Amazon has changed the way people buy products and the expectations that they have. That same mentality has even entered into the customized equipment industries. Manufacturers need to change what they perceive as a good lead time. There are three elements that you can compete with: price, quality and lead time. Quality has to be built into your system. The current paradigm is that price is going to be high, and lead time is going to be long because of customization. In the future, quality needs to be there but the price will have to correspond with the value, and lead time needs to be shorter. It has been said that you can only have two out of those three (price, lead time and quality). The new paradigm needs to be that all three are achievable and the lead time needs to head toward the minimum.
What are the implications for the workforce?
We’re moving in the right direction with apprenticeships and trade schools. In Innovation Engineering, we always say, “if you’re willing, I can make you able.” A lot of jobs say that they want 20 years of experience in some particular skill or industry, but those people are retiring. And the industries are changing anyway. You need a robust training program.
Especially with small companies, they’re very worried about investing in that because other people will steal the people they’ve trained. They feel like they can’t compete against bigger companies. That goes in hand with a culture that rewards loyalty. How do I bring willing people up and give them enough autonomy and intrinsic motivation to be loyal to this business? If you could do that, then you probably don’t have to worry about it.
There’s an art to how to do training effectively. It’s content, but also delivery— what’s the right delivery for the skillset you’re trying to build? Texas has resources to help with this. San Antonio Manufacturing Association (SAMA) is partnering with community colleges to do more technical training and the Austin Regional Manufacturer Association (ARMA) is as well. TMAC offers a variety of training including process improvement training. A lot of companies like COX manufacturing have also started apprenticeship programs.
Candidates really value a company that’s willing to invest in that. I’m currently helping one company build out an internal Analytics College. The younger employees are really hungry for things that build their capability, and if you’re willing to invest, you will build loyalty.
The upfront work is going to have the same type of elements in the workforce. All that is going to stay around, but there’s going to be a whole new level of coordination. A salesperson might have to look at an engineering model. There’s a different set of tools that the same people could handle. I don’t see a lot of change in personnel or skill-set except for training — how do you work with this new dataset that I’m generating for you? Moving away from drawing packet to instructional movies using monitors or iPads is what your current shop personnel wouldn’t have. A lot of information can be sent to an iPhone, and everyone knows how that works. The industrial equipment you use to produce these things, like a press brake—an operator still needs to know this information. But what he needs to look at to create his parts—that’s where I see the changes coming.
What are the next steps moving forward in the Central Texas community?
I personally would encourage manufacturers to get involved in their network like ARMA or SAMA because they can hear about these types of initiatives and how they’re going. There are peer groups they can be part of. They can hear about people who tried to do information flow mapping, and people who bought systems and what they’re disappointed in. I really think we should be learning from each other. There are strong networks where manufacturers can get advice and help and they know the resources that the region can tap into, from consultants to vendors. If you don’t know what to do next about digital, that’s where I would start.
The first step is that the owners of these manufacturing firms need to realize that things need to change. You need to get the upper management of the company to realize that the processes are too slow. For example, we are using 1980s vintage technology in the programs that help us run our sales, engineering, purchasing, and shop, while very slowly increasing technology around it. But the whole system needs to work smoothly where everything needs to integrate together. I don’t think you’ll find any small to medium manufacturers who have good visibility of flow of information or products all the way through the system. The company itself should be a system — not siloed, which is the way many companies that I have experience with currently operate.