To keep up with rapidly changing fashion trends, apparel companies produce clothes at a breakneck pace. Since 2000, the number of garments produced worldwide has doubled. Most of what’s produced never gets sold, though, and instead of offloading this surplus at a discount, or donating it, many companies discard it. Each year they send billions of dollars worth of unsold inventory to a warehouse or an incinerator.
The fashion industry is a large contributor to global overconsumption and pollution, as well — enough to drive protesters to fiery demonstrations.
It’s hard for a leopard to change its spots, especially when manufacturing processes and supply chains are long-solidified.
That’s why some tech startups are experimenting with new approaches to apparel manufacturing, with an eye fixed toward a more sustainable future.
The promise of 3D printed clothes
Imagine walking into a store, having your measurements taken in less than a minute and receiving a one-of-a-kind, perfectly fitting outfit in just a couple of hours. Custom clothes. On demand. Minimal waste.
That’s the future that the team at Unspun is on the verge of realizing.
The idea began in 2015. Beth Esponnette, then an assistant professor at the University of Oregon, was eager to remedy the rampant waste she observed while working in the apparel industry. She teamed up with fellow Stanford alum Walden Lam, who was leading strategy for lululemon’s Asia-Pacific operations. They soon brought on Kevin Martin, a mechanical engineer, to round out their founding team. They settled on their flagship product — a pair of denim jeans.
Denim jeans are woven. In a basic weave, there’s one set of yarns that is stationary and another set of yarns that is inserted in between. This makes long rectangular sheets of denim fabric, which are then cut out into individual panels and stitched together. A lot of excess fabric gets thrown out at this point, Martin told Built In.
Overall, this is a fine process if you want to stamp out thousands of identical, same-sized jeans at a speedy rate. It’s of little use, though, when you want to manufacture individually-sized jeans on demand.
“What would we need to do to flip the process from ‘Make a bunch of stuff and hope it gets sold’ to ‘Sell something, and then produce it’?”
According to Martin, Unspun was guided by one main question: “What would we need to do to flip the process from ‘Make a bunch of stuff and hope it gets sold’ to ‘Sell something, and then produce it’?” Figuring that out — and scaling it — would solve one of the industry’s glaring problems.
The team looked to three-dimensional printing, or additive printing, which had just started to enjoy a boom in popularity in industries ranging from medical to automotive before trickling into the fashion world.
The business-wear brand, Ministry of Supply, for example, has experimented with using 3D printing for some of its knit garments. And Eileen Fisher teamed up with Intel to show off its ability to print out knit sweaters in a 2017 demo.
The basic technology existed. But Unspun wanted to take it a step further — applying it to woven garments.
The team joined the HAX hardware accelerator program and booked a flight for Shenzhen, China. They toiled day and night for four months, working on their initial prototype.
“I think four months of Shenzhen time was probably 12 to 15 months of U.S. time. Just fast. You can do everything,” Martin said.
Unspun’s technology allows the paneling and cutting stages of the traditional weaving process to be skipped altogether. Instead, the yarns are woven directly into a 3D structure, as a single piece. To accomplish that, though, the printer needs to be able to manipulate every yarn individually.
“You can imagine then the ... challenge of [now having] a machine with thousands of individual actuations on it. You need to control it at that fine level,” Martin said.
To put that challenge in perspective, imagine going from playing a piano in which every key automatically travels down at the same time with one touch to a piano that has hundreds of independently moving keys.
The Unspun team returned to San Francisco, where they are still working to get their hardware up and running. According to Martin, once their printer is operational, the entire process — from selecting a custom pair of jeans to wearing it home — will only take two and a half hours.
In the meantime, they’re running a storefront where they use traditional cut-and-sew methods of production. To deliver on their promise of making perfectly fitting jeans, Unspun relies on another, less traditional, bit of technology too: 3D body scanning software.
The potential of 3D body scanning
More than half of online clothing purchases are returned due to wrong sizing, and over one-third of consumers purchase multiple sizes of the same item to make sure they get one that fits. All these returns cost retailers tens of billions of dollars a year. They’re incentivized to help customers choose the right fit — which is why they’re investing in 3D body scanning technology.
In a typical body scan, like the one Unspun uses for its customers, a person stands on a rotating platform while infrared sensors capture 100,000 data points on their body in less than a minute. The data is then used to identify a person’s sizes the same way a tailor would.
CALA, a tech startup that helps designers and fashion brands overcome production challenges, saw the potential of 3D body scans early on. The company leveraged the scan’s ability to allow customers to find their exact size before placing an order — a boon for bootstrapped brands trying to drive down return costs and avoid amassing an inventory.
Initially, customers loved the novelty, and brands appreciated the reduced return rate. In the end, though, the scan didn’t provide enough value for customers, especially those who didn’t want form-fitting clothes. “The big question from a consumer point is the friction involved in having to do the 3D body scans,” CALA cofounder and chief technology officer, Dylan Pyle, told Built In. “You have a pretty hard time convincing customers that this was a step they need to take before they buy this item.”
CALA ultimately pivoted away from 3D body scanning to focus on its other platform tools. However, several boutique apparel companies are still integrating the technology into their brand offerings, adorning their showrooms with a futuristic-looking contraption positioned where the tailor used to stand, or using smartphone apps that enable customers to upload their body scan data from home.
“The big question from a consumer point is the friction involved in having to do the 3D body scans.”
Big tech is betting on 3D body scanning technology to solve the fit-related returns problem.
In 2017, Amazon acquired Body Labs, a New York City-based startup specializing in 3D body scanning and modeling. The following year, the company invited people to its NYC offices to measure their bodies across the span of several months and in some cases participate in a 3D photo scan in exchange for Amazon gift cards, according to the Wall Street Journal.
BodyBlock AI, a subsidiary of Fit3D (the company whose scanners Uspun uses), looks to leverage a massive existing database of 3D body scans to help online shoppers find the right size — without requiring them to get scanned themselves.
Online retailers can feature BodyBlock’s plugin, which prompts customers to input a few basic data points (height, weight, age, etc.) before checkout. BodyBlock then compares these inputs with its database and predicts the accurate size for the customer.
The possibility of mass customization
“The opportunity for mass customization is there,” Unspun’s Martin said. “[But] it necessitates the hardware and the automation side.”
He envisions a day when we can shop online without ever guessing what size to order. We’ll seamlessly upload our at-home body scan data or Instagram photos into an Amazon Prime account — set it and forget it.
“I totally believe that the future of apparel will be on-demand and will be automated,” Martin said.
At which point, he hopes, the days of wasteful fashion are numbered.
