Meredith Westafer: Bringing systems thinking to complex problems

Meredith Westafer is an engineer at heart. When faced with complex and ambiguous problems, she advocates for taking a holistic approach that views the issue from a broad lens: Let’s not just optimize one specific piece of the supply chain but the whole entire system.

But when it comes to her own career path, Westafer’s biggest move was spurred not so much by systematic thinking but by a whim and a desire for a new challenge. Visiting San Francisco for the first time, Westafer decided to tour the Tesla factory near there. While on the website to sign up for the tour, she noticed the “apply” button. Intrigued, she decided to clean up her resumé and submit it. 

The next thing she knew she was helping to design the company’s first “gigafactory”—a term used for manufacturing facilities that make components and products associated with electrification and decarbonization. Almost nine years later, Westafer is now Tesla’s director of factory design and has played a central role in designing all of the company’s gigafactories as well as being involved in strategic planning for the automotive and clean energy company.

Westafer, along with Intel’s Jackie Sturm, was recognized at CSCMP’s 2023 EDGE Conference in October 2023 with the inaugural Outstanding Women in Supply Chain Award. The questions and answers below are modified from a Q&A session on the show floor with Supply Chain Xchange Executive Editor Susan Lacefield. A video of the full presentation and conversation can be found on Supply Chain Xchange’s website (www.TheSCXchange.com). 

What are your key responsibilities in your current position?

I run the factory design team for Tesla. I started at Tesla’s first gigafactory in Nevada, when it was just a big, flat piece of land. Basically, there were 14 of us working in one trailer. It was not enough people to build a factory, but we pushed forward. And now we are starting construction on our sixth gigafactory in Monterrey, [Mexico]. It’s insane how fast we moved. In any typical car company, we'd still be on the first one. But we've been able to incorporate lots of new changes along the way. 

My team does the material flow and the layout for those factories, which essentially means we're like the city planners for the factory. So, there's manufacturing teams that do the design of each of the individual manufacturing lines. And then we have a wonderful supply chain team that handles all the sourcing and logistics. My team makes it all work as a system. We intersect with supply chain at the docks, warehouses, and logistics yards. All of these things need to be planned into the overall factory to make that system work really well. 

That can be a huge job. When we're talking about something like our Texas gigafactory, it's like Disney World. It's a massive place that has tons of people coming in and out on a shift schedule. We have all of these peaks, all of these ebbs, and all these flows. And then we also have all of this steady logistics traffic that comes in and out of the site at the same time. So, we really have to think about all of that upfront when we design the factory. 

And if we're doing site selection, we're also looking at road and infrastructure design. We need to plan for the supply chain of the future, and that can be really hard to predict for a future factory when you don’t know exactly where your customers are going to be or where your suppliers are going to be based. 

What are the one or two things that keep you up at night? 

Something that keeps me up at night in terms of our factory design and relates back to the supply chain is what happens at our docks. Within the four walls of our factory, we are doing a crazy amount of automation. We have some of the world's most amazing automation teams designing bespoke systems to do all of our manufacturing. We design all of our products, most of our controls, and a lot of our software in house. Yet if you go to our warehouse, it's full of forklifts. It’s full of forklifts bumping around, unloading pallets off of trucks, there’s splinters everywhere, and there’s dirt flying. It’s dangerous. It’s the industry standard, unfortunately. 

I know there are companies doing truck unloading automation, but it’s really difficult to replace forklifts at the scale of an auto manufacturer. Because you need precision. Let’s say we have an autonomous forklift, [automated guided vehicle], or some other piece of automation. The [autonomous] forklift needs to be able to come into the trailer and see the load and the environment with the sensors that it has, and say, “Okay, this is safe.” But how many trailers have you seen where there's smashed pallets everywhere? It happens. It's totally normal. But that’s not a suitable environment for automation.

You need to have packaging that works really well with automation. If you’re talking one or two really specific parts coming from a local supplier, that’s fine, automate away. But our supply chain is so complex. Our suppliers are spread everywhere, and we're using thousands of parts to manufacture our cars. The cost for automation-friendly packaging to come from all of those suppliers would be astronomical. So, we have a supply chain that is spread across the world, and it all is coming into our docks, and that’s why we need to have those stupid forklifts unloading those trailers. 

The answer to this problem isn’t really straightforward, and that’s why it keeps me up at night. We need to continue to foster our supplier relationships, keeping them very close so we can enable cost-effective fleets of automation-friendly packaging.

We also need to continue to change truck fleets for the future, so they are not like they are now. For example, side-unloading trailers interface better with automation versus going all the way up and down the long axis of a rear-unloading trailer. This is the kind of thinking that we need. But it’s frustrating to me because we can't easily change, right? I can’t just say, “Oh, I want the entire U.S. to please have side unloading trailers for me right now.” I mean it might look really good on paper, but it would take eight years, and we would have to buy the entire fleet, etc. It just becomes a global problem at that point, and you really have to think about it that way. 

That leads to a good question: How can you work better with your suppliers on challenges such as using returnable packaging that would make automation easier? 

For us, it’s really about talking with our suppliers and being partners with them as early as we can. Tesla moves very fast. We don't always have the time to say, “What's our five-year plan? Okay then, let's go work with this vendor, and we’ll come up with something.” We tend to move a lot faster than that. So, we need suppliers that can move fast with us, which is tricky, especially with some of the suppliers that are still working under the old ways of thinking about auto manufacturing. 

So, when we're talking about a new site and new factory, we need to partner with our suppliers to be as localized to our factories as possible, when we have volumes that support that. It’s important to note that we're not talking about setting up shop for something like one little production line, we're talking about setting up shop for a world power of auto manufacturing. We sort of ask our suppliers to come take a risk with us, and we'll grow together. Let's figure out how to do something that's new and different. Something that's not just the way that we've always done it before. Let's make the future of supply chain, which requires risk; it's not going to be easy.

There is a dearth of talented technicians and frontline workers in the manufacturing and supply chain space. How can we start to build up that talent pool and attract people to these jobs?

At Tesla, we have a lot of need for automation technicians. I mean we can buy the entire world's supply of six-axis robots—let’s just to put that on the table, that is a limiting factor for us. Sometimes there are literally not enough robotics solutions in the world for us to do some of the projects that we have. So, we have to have a lot of robotics engineers, and we need a lot of techs to make everything work. It really takes an army. 

I think it's on us as an industry to go out and train. If we're going to turn to automation, if we're going to need all of these new high-skilled jobs, it’s incumbent on us as an industry to help create opportunities for our workforce to be trained to succeed in those new jobs. We don't want to displace people who are working in less skilled jobs with more skilled people, we want to create that opportunity for our associates to become the skilled workforce themselves. 

At Telsa, we've got automation and controls training programs partnered with educational institutions. In Nevada, we support a K-12 school program for robotics competitions. And a lot of those things are starting to pay dividends for us now, across the country, because we've been able to develop that workforce to some extent. 

So that's one of the ways that we're attempting to address it. But we still just need people, right? It's a huge, huge piece of manufacturing. And that’s where Optimus, our humanoid robot, comes in. But that’s for the most physically taxing and least desirable jobs that we have difficulty filling in the first place, so we can get people out of those roles and into the more skilled jobs.

What skills do you think future supply chain leaders will need? 

You have to be able to think about the whole system. In my role with factory design, we don’t just care about this one production line that people are designing, we care about how that line fits into the overall picture. We care about how it works with construction, how it works with the utility design, and how it works with the supply chain. Questions like: Do you have enough space for your parts? Have you considered the ergonomics piece? The whole system all the way back to as far as you can go. 

Supply chain is the same way. You really have to think about the whole system. It takes a lot of learning and passion to see, understand, and recognize the other opportunities out there for improving the entire system. It’s important to step out and understand the entire system, and not just the one thing that you were intending to optimize for.