Thu, Aug 11, 2022
mostly clear

Fashion's shift toward on-demand, real-time production models

Press Release

Credit: AADT

In today's fast-paced fashion industry, mass production models cannot keep up. That is why it is important that supply chains digitalize and shift toward more on-demand, real-time production models. This was the focus of the third day of the 2021 World Digital Textile Forum, hosted by the Advancement Association for Digital Textile (AADT) on Nov. 18.

Day three featured two fashion technology experts. First up was Taime Koe, CEO and co-founder of Six Atomic, a Singapore-based AI-driven apparel supply chain solution provider that provides automation technology for the fashion supply chain. Next was Mark Russell, co-founder and COO of Infinity Innovation, a provider of comprehensive and customizable digital supply chain solutions for the fashion industry.

Paul Chan, chief amazement officer of Hanin Enterprises, a Hong-Kong-based D2M and on-demand service provider, served as the moderator for day three of the forum.

Be part of real-time fashion or be left behind

Koe began by addressing the problems with the traditional pull model in today's rapidly changing fashion industry. Pointing out that the lead time is too long, "It means that brands would have to make decisions today and hope that months from now their investment would still be on-trend in consumers' minds."

However, in today's social-media-driven world, trends do not work this way and brands know it. Now, new trends can emerge on an hourly basis from social media apps like TikTok. This also means that consumers today expect that brands can keep up with and meet their fast-changing demands.

"Within the last decade, leading brands have been in a race to be the fastest," Koe said. This started with fast fashion brands like Zara in the 1990s to today's real-time fashion era brought on by Shein. Shein has managed to reduce lead times to an astonishing three days, from an already "fast" lead time of three weeks during the fast fashion era.

This is all made possible with the digitization of tons of data. By collecting data at every step, Koe pointed out that Shein is able to make more informed data-driven decisions that help achieve a higher chance of retail success, as well as reduce waste and inefficiencies.

"They analyze consumer data, predict demand, as well as adjust production on the fly. All the vendors and manufacturers in their network provide very transparent feedback to the brand Shein itself so that they can make decisions very quickly," Koe said. This provides Shein with a huge advantage over competitors who are less efficient at this real-time production model.

Koe also believes that end-to-end integration needs to happen at every single step of the supply chain. This is where Six Atomic shines.

Six Atomic can help brands and suppliers achieve real-time fashion through integration and automation. One of their core products uses an algorithm to generate garment patterns automatically within a few seconds.

Six Atomic's solution breaks down every part of the garment such as collars, patterns, cuffs, etc., to create a modular library. The starting point for the library can be pre-existing patterns provided by the brands or pure algorithms. According to Koe, this library is capable of creating millions of unique SKUs. From there, materials can be selected, seam allowances can be customized, and sizes can be selected. When the designer is ready to generate the pattern, all they have to do is click a button and the pattern is generated automatically by the algorithm in the backend and ready to download, reducing this process to a few seconds instead of hours.

Koe highlighted that pattern files come with adjacent file formats that contain rich metadata on the pattern, as well as relationships between pattern pieces. "When the adjacent file is opened in a 3D visualization software like Browzwear, once the file is exported the relationship between pattern pieces is also imported. This means that the 3D visualization of this pattern can be simulated with one click of a button, instead of having a designer or pattern maker manually simulate it," she explained.

So, how can textile suppliers and manufacturers get ready for real-time fashion? The primary thing is to digitize data for integrations, according to Koe.

"As you can see, data is extremely important in every part of the process to make this real-time fashion vision possible," she said. Data for textile suppliers include data for design, such as fabric thickness, color, pattern and quality of materials. It also includes data for 3D simulations, including the accurate scan of the fabric, texture, weight and physical properties that the 3D software would need to simulate an accurate representation. Data for automated pattern generation includes the stretch, shrinkage property, width for pattern nesting, as well as directional design. There is also data for sourcing so that brands know the pricing, inventory and availability, as well as shipment lead times.

Once brands have this data, they have the option to integrate with digital textile databases, like Frontier.cool.

More large brands have already started catching up with this digital transformation and are bringing in more automation. Suppliers that are more open-minded to change are inviting automation and digitalization into their supply chain and are being brought in by brands to lead this journey, Koe added.

"Real-time fashion is already a reality as we can see with Shein. More and more brands have made improvements in their supply chain efficiency to move in this direction. So it's your choice whether you will evolve with the rest of the industry or be left behind," she concluded.

Kickstarting digital transformation in supply chain

Russell began by addressing five basic principles being addressed by digitalization and the digital entities being created in the industry, which include reduced barriers to digital entry, decentralization, decoupled value chains, the importance of data, and technological eco advantage.

Regarding customer expectations, Russell pointed out that there is a massive shift in terms of what the consumer is looking for. "Everybody talks about push and pull models. I think as we talk more about pull models, there are expectations from customers in terms of moving away from mass production into enabling brands to purchase on a micro-season, almost month by month basis," he said. In fact, on-demand solutions and a degree of personalization are also starting to show up with a lot of brands.

As the industry moves into digital processes, it has been able to look at real-time data. Russell believes that the need for "less is more" is working directly with on-demand solutions. Getting data on a real-time basis and month-by-month basis is key to how a brand positions its raw materials, affecting everything from ratio information in terms of what it used to cut traditionally, which was all based on assumption. Now he says there is a lot more information about styles and colors that are selling, as well as size ratios, that are quite impactful.

However, when it comes to engaging and trying to figure out a brand's existing digital footprint and where they want to be, there is a lot of confusion. This confusion is centered on what to do with the type of technologies available and how to make sense of and string them all together to make a real end solution.

"Everybody is looking at 3D for product creation, in terms of taking a lot of time and cost out of the front end and making the decisions a lot easier and faster. I think that becomes more relevant nowadays due to the inability for people to travel," Russell said. "Enabling digital transformation takes a big, deep dive into really understanding the existing digital maturity of the company and where it sees itself..and can be a bit of a challenge depending on the brand."

Russell pointed out that digital adoption is embraced and widely accepted by SMEs and smaller companies. The struggle lies in big retail, possibly because they are not as digitally savvy as the younger brands. A lot of big retail brands have traditionally built processes and systems around push models. Trying to move from a push to a pull model can be very challenging.

The path to digital transformation first requires the digital partner to assess the customer's digital footprint. "It is very important to see where they are, what's in place and how they're using it," Russell said. Next comes the development of a digital maturity model. This focuses on how the business is transformed and operates to increase its competitive advantage through digital initiatives. This is followed by identifying the points of differentiation through digital strategy, defining a clear roadmap to enhance business efficiency and effectiveness, and finally building a culture to support digital maturity.

"There is a very set process in terms of how that's implemented. A lot of people want to run before they can walk. It's very important for brands and companies to understand that there are key processes and strategies involved," Russell explained.

"Within Infinity, we've made it a point that we want to ensure the system and digital assets that we operate facilitates everything from big retailers, SMEs to smaller brands and what we classify as non-apparel," Russell said.

Credit: AADT

The third day of 2021 World Digital Textile Forum focuses on "Fashion’s Digital Transformation in Supply Chain: Now or Never", discussing why it is important that supply chains digitalize and shift toward more on-demand, real-time production models.