BEN: The manufacturing work and work in other industries is really beginning to shift from people doing physical tasks close to where a product is being made, to people supervising a process, really working in a more control room environment where they're orchestrating machines.

DANIEL: Hello, everybody. Welcome to MIT Learn. You're listening to Why This Matters, a podcast that aims to connect abstract ideas with real world impacts. I'm your host, Daniel Braconnier. Today in the studio, we have Doctor Ben Armstrong, the Executive Director of the Industrial Performance Center here at MIT. And together, we're going to find out why this matters. But before we get to this, thanks for being here.

BEN: Thank you. Thanks for having me.

DANIEL: Tell me about, your role here a little bit more at MIT. What's this, "Industrial Performance Center."

BEN: The Industrial Performance Center was founded by a multidisciplinary group of faculty in the late 1980s, early 1990s that came out of an MIT wide commission on productivity. And the goal of that commission, which was formed by then MIT president was to understand how American manufacturing can compete with the rise of Japanese manufacturing and other, manufacturing in other countries that were really eating the Americans' lunch at that point. And, a number of MIT faculty thought this is a really important issue for us to understand that we have something to contribute as scientists, engineers, social scientists.

And what they ended up doing is, is coming together to study the issue from a number of different angles. Was there a technological problem that was facing US manufacturing that led it to fall behind? Was there a workforce problem? Was there something in the capital markets that was really troubling U.S. manufacturing? And their conclusions, they, summarized in a book called Made in America, which laid out a course for how U.S. manufacturing might have been rebuilt in the 1990s. And unfortunately, a lot of the great ideas in that book didn't come to pass, and US manufacturing continued to decline after that point.

And the decline actually got worse. It accelerated in, in the early 2000's. But nonetheless, the Industrial Performance Center came out of that project. And over the past 35 years, we've been doing, similar research, like the research that started the Center, research on manufacturing, competitiveness, globalization, how automation affects the workforce. And our core has always been, in really heavy industries, industries like automotive, manufacturing, aerospace, manufacturing, metals manufacturing. But we've branched out a bit. Some of our research now focuses on logistics. We have projects in generative AI that cut across industries looking at how generative AI is reshaping work.

So our our scope's expanded somewhat, but the, the core pillar of our work has as long been manufacturing.

DANIEL: I mean, really important. So were you a part of the creation of this or how did you how did you end up becoming the Director?

BEN: Well, you know, I was in diapers when the creation was happening. So. No, but when I was, you know, I started my career in, in Silicon Valley. I was, graduating college at a time when the tech industry was really on a meteoric rise. And the most exciting job, I thought, and my peers thought, was it places like Google where I, I went to work first and really tried to understand, well, how is this company running? What's so exciting about this company? Of course, you see the rainbow colored beanbags and the ping pong tables in the offices. And I was curious what was happening in. I went, and I learned a lot, but I also learned that that wasn't for me.

That wasn't the business environment for me. It wasn't where I felt like I was really making a contribution or really feeling satisfied with the work that I was doing. So I moved on to, a startup. And when I was at the startup, I started working on a project that was trying to build something physical. Of course, at the time, in the early 2010's, every company that I could see in Silicon Valley was building something digital. It was focused on software. The costs were very low. There's a lot of venture capital for new software ideas. Mobile apps, software as a service. But not many people were building anything physical.

And I came from the Midwest, growing up around the automotive industry. And I really liked the idea of, you know, building physical things that could be sold for some real, measurable margin. And, I and some, some peers wanted to build this physical thing. And we realized it was really hard to build it in the US. So it's an electronic device with some remote sensing capability. And everyone was telling us, you just got to go build it in China. You got to find someone who can set you up with, essentially a supply chain in Shenzhen. That's the way you built it, and it got me interested. So why is it so hard to build something so basic?

It felt like, so basic and so essential to the modern world, like, electronic device in the US. And that led me, to come to MIT and start studying this problem of what really happened to American manufacturing? And as a graduate student at MIT, I started getting involved in that Industrial Performance Center where I am now.

DANIEL: Yeah, there it is. So you mentioned you you grew up in the the Midwest. Was your was your family connected to manufacturing at all or how did you kind of have that perspective from from being out there?

BEN: Well, in two ways. The city where I was born, what had a manufacturing legacy in the automotive industry. It was a city that really grew with the rise of manufacturing in the Midwest around Chicago. And, by the time that I was growing up, it was nowhere to be found, right? There was a casino downtown. There wasn't really any factory jobs or, you know, middle class jobs where you could support a family like the the lure of, you know, the postwar 1950s and 60s. So manufacturing had gone, but it was sort of in the legend of the place. So a lot of deindustrialized towns like that around where I grew up. And then the second piece was that, my dad was in the automotive industry and was, in sort of this constellation of suppliers and distributors, at the lower tier below the OEMs.

So what he did was he sold a number of different automotive components to, lower volume customers like John Deere or Caterpillar. But it was the same kind of components that would go into Ford or GM. So, I had a vague sense growing up of A, what these parts looked like and what a warehouse floor looked like, and also what the ecosystem was. It wasn't just the Fords and the GM's. There were hundreds if not thousands of other companies that were really important that fed this whole ecosystem. That was a critical part of the economy, and that those small companies, like the ones, the one where my dad worked, was just as important as the the large companies that made the headlines.

DANIEL: Do you feel like that experience got you into science and engineering, or was was it some other thing that sparked that interest?

BEN: Well, I think like many kids, what your parents are involved with, you want to run away from before you run back to. So I think maybe going into Silicon Valley from the Midwest is like trying to understand if the grass is really greener elsewhere. And then, I'm very glad that I sort of came back home, at least if not geographically and certainly in the subject matter that, I do feel connected to manufacturing in a lot of ways, and I'm happy to go back and discuss the issues I work on with my family who, who know more about them in many ways than I do. Awesome.

DANIEL: Our listeners will probably guess that we're talking around or about manufacturing today. So before we get into more topical questions in your mind, what is manufacturing?

BEN: So the traditional definition of manufacturing is anything to add value, to to a physical good by somehow transforming it.

DANIEL: Right.

BEN: So I'm butchering the textbook definition. But really there is a process in which you take certain materials and components and turn them into something that's more valuable. It's almost like a biological process where in our in our bodies, we constantly have different, kind of biological components that are coming together to do something that those individual components couldn't do on their own. And that's a lot of what, manufacturing is about. It's about, you know, reshaping metal and plastic and composite into something that then can go into something else, that can go into something else that then becomes valuable.

To individuals and to society and to industry. One of the... You know, I think why I'm hesitating is that I've begun to think about manufacturing even more broadly than the traditional sense of steel mills and car factories and plain assembly and, you know, toy making. And I'm thinking about it in two ways. One is I think we are more and more looking at mining and the cultivation of raw materials as a critical part of manufacturing, that actually taking things out of the earth and making them more valuable follows a very similar process, requires similar technologies as traditional manufacturing, and to think about them holistically, makes a lot of sense that that the mining upstream of, of core materials and the processing of them and some mining companies are vertically integrated, such that they do mining and manufacturing.

And our thinking about that as hand in glove. And then the second piece, which is important, not just because, it's technically very interesting, but also because it's environmentally important, is this process of remanufacturing. Sometimes it's thought of as maintenance and repair. Maybe it's repair and overhaul. So in the automotive sector, in consumer products, in aerospace in particular, there is a big industry in taking existing components that are worn or torn or twisted and making them usable again, and that in I began to study that with a colleague, Zack Cordero here, who's in the Aero Astro department.

Interesting how new technologies might make that process. The process of remanufacturing more efficient, because you can imagine to set up an automated line for, a car that's going to look the same every time and control that line. It's a feat of engineering, but automation works really well when you're doing the same thing over and over and over again. But what if the part that you're making, or the part that you're assembling is slightly different every time, and you can't predict exactly how it's different, that it requires not just traditional automation technology, but a software layer on top of it that might need to help the machines adapt to slightly different environments. And that's a really exciting technical problem with with great potential benefits.

If we're able to reuse materials that we would have otherwise scrapped.

DANIEL: Oh yeah. So this is different than recycling, I guess. From or would would the terms be interchangeable in this sphere?

BEN: Well, recycling is quite different in the sense that when you're recycling something like let's take what Redwood Materials does with a battery, there would be a difference between recycling, which is extracting the valuable raw material from an object, and then using that raw material in a new manufacturing process. What MRO facilities do in aerospace, for example, is they might take an engine that's been on a plane working for a long time. That is, you know, warped or needs an overhaul, and they'll make it as good as new effectively. Or they'll take components and they'll run them through a new machine or turbine blades and run them through a new process to make them as good as new, so that you don't need to necessarily fab a new blade.

This a long standing set of processes. But, at least my hypothesis is, and we're still figuring this out. And there are some folks in, in Germany actually at the Karlsruhe Institute Technology who are doing great work in this area. They have what they call the circular factory, where they're taking all sorts of different components and equipment, and they're trying to design processes to take these used components and bring them back as good as new in domains where this hasn't necessarily been done historically, like in aerospace.

DANIEL: So with that, you were mentioning automation for its do you feel like that's the kind of the special next step to make this manufacturing possible? Is there or is a lot of it still very hands on?

BEN: Well, I think that's the starting point is certainly hands on, because you could have someone with a lot of craft based skills go in and kind of re-machine and tweak a piece to make it work. I mean, this is the traditional hobbyist who might be able to make a used lawn mower, you know, add a few new components and then make it hum better than ever. Those sorts of skills certainly exist. The only issue is that's not more cost effective necessarily, than adding more capacity to an original production line. So the economics of it are really the challenge of how do you make this re-manufacturing or the circular economy, essentially have more economic return than the original manufacturing economy.

And it's not going to be true for all components. For example, cheap, replaceable components. It might not make sense to repair those, but there are enough high value components. And this is what, folks in, in Europe where there's a, you know, higher cost of carbon and more interest in sustainability, they're banking on if you do raise the cost of carbon, for example, which is more intensive in original manufacturing, you then bring in, it becomes a lot more attractive to invest in re-manufacturing in more domains.

DANIEL: Yeah, it gets to, I guess, industrial policies. Across the world. I know you've done a lot with US focused stuff, especially here in the, the, Center here at MIT. Do you how do you feel the connection of. Because an industrial policy can feel pretty like up above everybody that might be just working like a, a metal shop, bending some sheet metal or something. But how do you feel it connects down to the local economies, around here?

BEN: Well, the the main goal of industrial policy, if we can have one kind of unifying goal is to, is to shift the structure, industrial structure of the economy, to make some industries, to give some industries more advantage and some industries less advantage oftentimes. So industrial policy historically has been either to incentivize innovation in new industries and make those new industries more attractive. So people go into those industries with the idea that they maybe pay higher wages or are more competitive. And over the last four administrations, it's really the Obama administration. Trump won Biden and now the current Trump administration, there's been a concerted effort through a variety of tools, different tools for each administration, actually to incentivize the revitalization of American manufacturing.

And those policies have been based on a few simple ideas. One is that what we learned between 2000 and 2010 was that the costs of losing manufacturing are pretty dire. When you let go of manufacturing, you lose a lot of middle class jobs. You really hurt certain communities and pretty sharply, you make it harder for our national security apparatus to maintain its technological advantage and advantage in war fighting material. And then finally, you lose your ability to innovate on the technological frontier, because if you're building, you know, the next generation drones or if you building, the next sort of, electronic device you're designing in the US, but then you have to go and build it and really figure out how to iterate in another place.

You lose some of the capacity to really scale the business in the United States. And we've seen more and more that businesses that started with the design in America and build in Asia strategy just ended up being businesses that were scaling more dramatically in Asia. So that recognition led the US to try to claw back some of some manufacturing activity, not necessarily taking supply chains that were already established in Asia around, let's say, garments, shoes, even electronics, and then like, re-shore them totally. There's been some talk about that, but that's really hard. I think the goal, primarily what I see as the goal is to ensure that when there are the next wave of physical things that need to be built, the most advanced physical things that need to be built, that we have the capability to, to build them and to compete, to build them, effectively in the United States.

And that's what the goal of industrial policy has been. How does that matter for American workers, American factories? It hopefully will make it easier for them to access knowledge to do their job better. It will give more incentives for them to attract very skilled workers because manufacturing jobs will be more attractive if these policies are effective. And it will hopefully. And this is, I think, a more controversial, point is it will make, you know, the raw materials, you know, the materials necessary do manufacturing, you know, equipment, infrastructure, other technology more accessible to them. And that's one of the real disadvantages that American manufacturers have today is that equipment costs a ton.

It takes a long time to access, and you don't necessarily have a skilled workforce to operate effectively. And in manufacturing hubs, like I was just in, in southern China, it's a totally different story, right? The equipment is, is highly affordable. It's accessible everywhere. And that allows you to move much faster and a much lower cost.

DANIEL: You brought up two things that it that seem interesting, like the. I guess the societal view of manufacturing as a career. In the US, at least from what I know, seems a little little lacking. Like there's a huge gap between between a workforce, the age of new people coming in out of, say, college or or high school. And then you get the really trained individuals who are, you know, doing time during their retirement at some, at some places, whereas and you bring up in an over in Asia, it seems like that's, almost just a normal like it's just, oh yeah, I can totally do that if I wanted to do that. And how do you where do you see the big differences and why we're kind of there?

I know there's a lot of outsourcing, but what else do you think is affecting that?

BEN: It's, it's a... It's a it's a long story. Not that I need to spend a lot. I mean it's been happening for a long time. So this isn't something that just came up since Covid where there was a big labor shortage, a lot of focus on, you know, why aren't people going to the trades? Why aren't people going to manufacturing? But this is a story really, that started in, the 1980s and 1990s. So we've had now, you know, at least a generation and a half of people who didn't look at manufacturing jobs as attractive. And one of the reasons was that you looked at this industry and it had declined from being 25% of the economy, or 25% of the workforce, to 10% of the workforce, like, am I really going to go into an industry where it looks like the bottom's falling out and factories are closing, and maybe my parents friends who were in the industry lost their jobs?

I don't think as a as a parent, I would want to send my kids into that industry and, it doesn't you don't really want to get on a sinking ship, I think is how a lot of people felt, about the time. So what you saw as a result was the workforce in manufacturing became progressively older because people stuck around in a, an industry that they liked, that I think, for a lot of people is very satisfying to build things and to see the fruits of your labor going out the door. But they weren't backfilling those workers as they were retiring with, with new, with new skilled workers. That manufacturing is essentially a last resort for a lot of people.

And I think there are a few other things that were happening during that time. One is that manufacturing went from being the best possible job, wage wise, for someone without a college degree to being a job equivalent to a lot of jobs in retail or logistics. So in 1960, if you're a manufacturing worker without a college degree, you're on average earning 40% more on a factory floor than you would earn in an alternative career. Today, you earn about 2% more on a factory floor than an alternative career. Massachusetts, you earn less. You could earn more at a lot of other places, like an Amazon and Walmart.

Target, maybe not Walmart. So anyway, so, so so this is a big transformation that manufacturing went from this high wage really, I think respected career to being a career that's lumped in to a lot of other, you know, low wage work that specialized for people without a college degree. So I think there's there's both a rational decision from, from young individuals who say, I don't want to go into an industry that's declining. I don't want to go into a job that's comparatively low wage. That also requires me to invest in skills, whereas I might be able to just jump into another, career on the job and figure it out.

I think there's there's something, though, that that young, people entering the workforce are missing, which is that the opportunities for upward mobility in manufacturing, the ability to gain new skills. And one thing that we have in the US and it manifest other places as well, like Germany or Japan or China, is that people who really invest in developing skills in manufacturing can earn a great career and can really rise into positions that are they're considered engineers, even know they don't have a college degree. There's a big share of, folks in manufacturing who have engineering titles but don't have a four year degree, and they earn really good wages, wages that are far higher than they can earn in other domains.

So the good jobs in manufacturing still exist. They're just hard to hard to find. And they're, might take more work, particular for people without a four year degree. So, so that's the challenge where, essentially it's the the door to those good jobs is hard to open, but it's still there. And I think the, the goal for, for policymakers, for, I think institutions like MIT that work with employers and help with training is just to make that door to good jobs, manufacturing easier to open and maybe a little wider so more people can fit through.

DANIEL: Do you do you feel like that door is opening?

BEN: I don't think we have the evidence. I mean, you know, I was I was quite optimistic after Covid, so so we had a pretty bad labor market situation in the, around 2018 before Covid. So, there was a shortage, far more job openings in manufacturing than there were people who wanted those jobs. You talked to most manufacturers like, I can't find people to hire. I want to grow and my customers want me to grow. I just can't find the people to do the work that got astronomically worse. Triple, triple the job openings after Covid, right 2021. But what we have in 2021 is really a mobilization of private sector capital and public officials that saw Covid and said, oh no, we can't let this happen again.

We really need to support, a new workforce agenda and new, factory infrastructure in the US. So not only job openings tripled, the investment, new factories, you know, tripled as well and really coincide. So there's this real energy behind the growth of manufacturing. And I think an excitement around manufacturing jobs at places like, you know, new battery factories making electric vehicles or places like TSMC in Arizona, which were different kinds of jobs than people had associated with manufacturing. And, unfortunately, in beginning in 2025, we saw what had been growth in manufacturing investment and growth in manufacturing employment really taper off.

And last year was a year of manufacturing decline for the first time in quite a while, really excepting that, you know, like the Covid spikes and troughs. So I'm still optimistic that there's still a consensus that we need more manufacturing in the US. And I think this this short term decline will will give way to longer term growth. And we'll begin to figure out these workforce issues. But, we have a long way to go.

DANIEL: So I, I guess thinking back to wanting to, you know, try out Google. And that might have felt like the cutting edge at the time for a lot of people, like getting into CSS jobs or, or more software oriented type of things. What's the what do you feel the cutting edge of manufacturing is nowadays? That would be an exciting, enticing thing for people to get into that, that line of work.

BEN: Yeah. And just to clarify, I'm not an engineer, so I was not going to Google with that. Maybe I would have liked to Google a lot more because it was interesting. There's this hierarchy where the, the, the engineers are, are royalty for the company like that, and the rest of the people are really kind of figuring out their own roles and identities. And in manufacturing, you have some companies that are similar where you have certain kind of classes of workers that are really given a lot of authority and, creative license. And then factory floor workers are often in at least American companies. I don't want to generalize because there are a lot of great American companies.

They give a lot of discretion to factory workers. But one of the big differences between, let's say, a Germany or Japan in the US is the, is the level of, creativity, discretion, respect given to the workers on the factory floor that I think, that American factories in general have a lot to, to learn from. What is the American factory job? I think two, two answers to this. One is geographically, we're in a bit of a pickle with American manufacturing. It used to be very clear where the hub for manufacturing was. It was along the Great Lakes in places like Cleveland and Detroit and Pittsburgh. Today, if an MIT graduate wants to go into manufacturing, there's no clear American region to send them to, right?

Maybe if they have a very specialized industry, we can send them somewhere. But if we're really being honest with ourselves, we would probably send them to Shenzhen in China to to figure things out and to really learn and work for, fast paced company that would, satisfy their interests. I think that needs to change. I think we need to figure out in the US where is going to be the hub for the most exciting, creative, talented people in manufacturing to go to build companies to help other companies grow. Our American version of what Detroit was in the early 20th century. And that's missing today. Frankly, the closest thing we have to it is in Southern California with startups coming out of SpaceX and Tesla.

And I think in in Southern California, you have companies like Anduril and SpaceX and a number of new investors. Mostly related to the defense industry, but a lot of really exciting companies. And despite being a, a very high wage labor market, there's been, a, bounty of really great manufacturing activity there and excitement about learning how to make things. I was surprised I went to visit a startup in that region. And they were doing work with composites, and they were doing some complicated welding work. And I asked because I assumed, given this region, that they were going to face labor shortages and have a hard time finding those people.

And they said, well, they're expensive, but we have a lot of them in this region because of the, excitement for people to to work in these sorts of places that are, bringing a little bit of this Silicon Valley, fairy dust. We're building companies that are going to change the world except in hardware. And that appeals to a wider skill set than just, computer science and, of course, six majors Yeah. At MIT. Yeah. So I think if there is a region, even though it's quite a narrow set of interest, it would be Southern California. If there are a set of companies, the real paradigm shifters were Tesla and SpaceX.

Those companies have have lost some of their luster, I think, for a variety of reasons. But, there is a family of companies now that have been founded by Tesla and SpaceX. So just think about manufacturing, and hardware differently than legacy companies like, let's say a Ford or General Electric. And that I think are exciting to recent graduates.

DANIEL: So what are the ways in which they think differently about manufacturing?

BEN: So there are, there are three and I don't think this just comes from Elon Musk. And I've, you know, the the maybe you think of it like the Walter Isaacson deification of his ideas. I think it comes from the, types of people who came together to build those companies, people who have backgrounds in the hard sciences, who have backgrounds in in Tesla case, in semiconductor or manufacturing, as well as in automotive manufacturing and what they, they really do three things. One is they're willing to move much faster than, peer companies. So the the way that Tesla ended up being more productive than their peers wasn't to build a new kind of assembly line that moved much faster.

It was instead to have much shorter product cycles, so that when other companies were spending a lot of labor hours developing and iterating the next version of the car, Tesla was already done, and they were into production. So they can afford to make more mistakes in production, with a much tighter, product cycle. The second is not being afraid to make mistakes, right? Where Tesla was willing to change various, elements of the vehicle on the fly as they were figuring out how to improve it. The design, it wasn't locked in in the same sense. And I think part of this was, because they were so early as a company, they had to be willing to make those mistakes.

But this is really importing some of the software company philosophy, of of failing fast and adapting that. It was, I think, very exciting and ended up, working, in their ability to, to learn a lot from the Roadster and then learn a lot from the, Model S and so on until they get to the Model Y and they have the best selling vehicle. Right. So so that was very rapid, iteration for a startup to achieve that in such a short period of time. I think is really, remarkable. And then, finally, I think the, the investment software. So, so Tesla and now Rivian and some other companies have really restructured what the manufacturing workforce looks like for their respective industries.

And that's with a a much larger investment in software capability. So they can build internal custom software tools. They can use the data that their machines are throwing off much more effectively. And they can be be more flexible, which is really what the, the superpower that software brings to manufacturing is, is flexibility. And for a lot of other even large manufacturing companies that don't have these, these capable I mean, they have internal software teams, but they rely a lot on external expertise and external vendors. They don't necessarily get the same flexibility out of their software.

DANIEL: So with the software that I'm sure a merger that's a lot of automated and maybe some robotic stuff on the line with automation, I know that's a big thing with when when people think about manufacturing nowadays. Is is are we entering a space where we kind of know where that fits along with someone who's actually on the shop floor? Or is how is that changing that position, changing?

BEN: Well, it varies a lot product to product, industry to industry. So there are certain things that we've really figured out how to automate. So, producing certain motors, you know, welding the body of a vehicle, there are extremely impressive automated lines and systems that, we're really, you know, that manufacturers are great at and that, there's very little human intervention, there's human supervision. So there's a control room with people figuring out how to adjust. And part of the challenge is when something goes wrong, how do you minimize the downtime of that automated system? And then there are other, there are other operations, particularly final assembly of finicky devices where there are different angles where you have to screw things in and the pieces are delicate or, flexible.

Those are really hard. And there is a lot of manual labor that remains. But most processes are somewhere in between where you have, a lot of human labor and maybe some capability to automate. And it's maybe, a borderline call of whether the automation is worth the sweat that would go into it. And one of the difficulties for American manufacturers is that automation isn't just the cost of the robot or the equipment, it's often the cost of the knowledge to figure out how this robot plugs in. And that's more costly. You know, some of my colleagues have estimated, yeah, 2 to 3 times more costly is the integration of the automation than the automation itself.

Now, where we're at a disadvantage in the United States is that, that knowledge is harder to come by than in other places with a lot more density of manufacturing. So if you go to a factory in a place with tons of manufacturing capability, you have automation engineers who do this all the time that they know integration, they can do it faster and they're already on staff. So it's not it's not necessarily an additional cost. You're kind of expecting to do that integration. Whereas in the US a lot of companies, they rely on a third party to do it. And you need to set that up. Then you maybe you need to wait for the equipment because the equipment's coming from overseas.

So the whole process of, of doing automation becomes a lot more complicated, a lot more costly. And there are just plenty of reasons to say, no, I'm going to do it how I know how to do it. So we end up with what we have today, which is only about 12% of American factories with any robotic automation at all. A lot of them have specialized equipment. They know how to automate in the ways that are industry standard. But pushing the envelope, really making dramatic improvements has been it's been quite hard for for companies - automotive industry like you asked about Tesla. Very different story. But 50% of all robots in in deployment are in automotive.

And a lot of the big automotive processes, are quite automated. I would say Tesla's advantage doesn't necessarily come from more automation. It comes from more vertical integration, where Tesla was willing to make more parts of the car that they thought they could make, and control the quality of better, rather than rely on suppliers. That might have, you know, variable quality issues or they might not be able to, it might not be able to move as fast as Tesla is moving and changing.

DANIEL: Yeah. I've toured some facilities around in the Cambridge area and I remember one from a sheet metal bending plant. They had this it was from maybe 100 year old just like five, 500 ton press, and that was their workhorse. Every other machine was either just as old or or, maybe a little younger, but pretty much just classic, huge cast iron machines that, no sense of automation over there. They, they had maybe a, forgive if I forget the acronym, but is it a DRO that controls location on a, on a mill?

BEN: or are you talking about, like, probing? Yeah. Yeah. Like they had automated probing to look. Yeah.

DANIEL: They had, they had that on like one machine out of all their stuff that they did and everything was manual and, and do you so you mentioned like Tesla does all the vertical integration of their parts and everything. Did do you think the I guess the pressure of automation kind of being the next step for a lot of manufacturers is that is that founded or is is there really still a place for like just because the machines are old? But if the product still good, is there a way to to make that that work? BEN Yeah. I mean the the only way that manufacturers in the US with far higher wages and other costs than manufacturers elsewhere can compete is by becoming far more productive.

Two ways to become more productive. One is to invest in technology that helps you do the same kind of thing faster than you could do it otherwise. The second is to invest in skills where the people figure out how to do things faster and faster. And then the, the, you know, the magic is when you can do both. Right? Have more skilled people, working with better technology. So we talk about automation. I think the most people's minds goes directly to, to robots that are, you know, a physical arm doing something a human arm might have done. But that's not the only kind of automation. Not by far. There could be automated inspection that uses vision systems, which is becoming more and more popular in manufacturing.

There could be software that helps do changeovers where people don't need to reprogram, you know, manually in a system. So, so there are all sorts of different technologies that what they're trying to do is help a human do a task faster. So the human in in manufacturing, the human is always in the loop in one way or another. And the question is, can you help them do more tasks faster? So I think the in the long term and this is what, you know, I was teaching a class in the fall that was on this theme is manufacturing work. And work in other industries is really beginning to shift. It's a slow shift. It's beginning to shift from people doing physical tasks close to where a product is being made, to people supervising a process, really working in a more control room environment where they're orchestrating machines and they're trying to ensure that the machines are doing what they need to do, and they're adjusting them as appropriate and trying to make them more flexible.

And that control room environment is in some ways more the work of a traditional engineer, but it also requires some in-situ experience. You need to know what it looks like when a process is going well on the factory floor, and you need to know all the different varieties of how it might break down so you can respond quickly. So there's this mix of knowledge that's really required. But I think the goal for a lot of places is to make that shift from lots of workers close to a process who need to be hands on to probably a smaller number of workers per the value add. Right? Maybe. I imagine as manufacturing grows more workers overall, but each worker is going to be responsible for a lot more value that's created in the factory.

I think that really clears up a lot of stuff when people are thinking about automation in manufacturing. Is there any other like, misconception that you think, people would need a little bit more information?

BEN: Well, I think there's competing narratives about what manufacturing is. And part of it's because there's high variety in, in factories, that you might have seen in your visits and that my colleagues and I have certainly seen when we go out and, and visit factories, is that there are some factories that are really focused on quality and reliability in the sense that a successful day is to produce exact copies of a thing many, many, many, many times, and to hit a target. And you could think of this as, some people call it, you know, copy exactly. Manufacturing or like, set it and forget it manufacturing. And that's important.

It's important to know how to produce something extremely reliably with very little variation many times over. But that set of skills is insufficient for the US to compete in manufacturing. There's this other type of manufacturing, which is the type that we talked about with Tesla, which is much more inventive, where you're trying to figure out how to produce something and produce it, slightly better, slightly better, slightly better over time. And you're very flexible to introduce new varieties of the product and, produce the product in new ways and, maintain reliability so that you're still within spec, but that you're, you're really changing the process over time.

It's not a copy exactly type of, of manufacturing. And that second type is prone to a lot more error and requires a lot more, you know, I would say controls in other ways, not the traditional controls. Engineering. Yeah. But that type of manufacturing is going to be required if if the US is going to be prepared to, to make new things and figure out how to make new things quickly. It's really that manufacturing is focused on learning, in a way that's different from the copy exactly manufacturing. Now, I'm sure that that comment will have some pushback from folks in this, maybe that that feel part of the copy exactly camp.

Because I'm not trying to, to devalue that type of manufacturing at all because there's a there's a variant to that where manufacturing focused on continuous improvement. One of the big, mantras in among U.S. manufacturers has been lean, which is a philosophy that came from the Toyota Production System and is roughly describes continuous improvement through respect for people. And you can have something that you need to make a million times in exactly the same way, and run a continuous improvement process on that product to find it, cut out steps and cut out waste and, give people more discretion of finding ways to make that product exactly the same, but more efficiently.

I think that's great. And that would be a way of doing the copy exactly model better. But there still needs to be, and it needs to be the ability to, to learn how to do new things quickly and be comfortable with high levels of, of variation. I think, if we're going to compete with some of the, very, you know, very powerful and impressive manufacturers elsewhere in the world.

DANIEL: Yeah, that makes a lot of sense. All right. Well, so we're nearing the end. I'm, I'm trying to start this TLDL segment, so too long didn't listen. And if you could address our camera over here to our listeners and maybe give a quick summary about, you know, why we what we talked about today, matters and what you're looking forward to in this type of type of area.

BEN: Well, if you care about, building new kinds of products like drones or the next generation of, vertical takeoff and landing aircraft or nuclear energy, all of those new products rely on us figuring out how to manufacture new things in new ways competitively. And that's why we should care about getting better at manufacturing.

DANIEL: Thanks again for joining us. Thanks so much for, for our listeners still curious, which I'm sure some of you are, you can learn more about manufacturing and a whole lot more for free at learn.mit.edu. I would recommend, if you're a junior level undergrad and curious about manufacturing, there's the course number 2008X Fundamentals of Manufacturing Processes is a great course to get started online with that. And this has been the MIT Learn podcast, Why This Matters. I'm Daniel Broconnier. Thanks for listening.