The fashion industry is one of the largest polluting industries in the world — second only to oil and gas. It consumes over 2,700 litres of water to produce just one cotton t-shirt, and along with the wider textile industry, emits roughly 5% of global carbon emissions.
Fashion is incredibly wasteful too. The global industry is responsible for creating 92 million tons of waste each year, which is expected to rise to 148 million tons by 2030. Whether the garments are made from natural or synthetic-based fibres, the dyeing and manufacturing process makes them difficult to compost. As a result, these clothes just end up clogging our landfills for hundreds of years.
At the heart of this problem is the short life cycle of garments. In only 10 years, the life cycles of consumer products have halved, with retailers now pushing new products every 3 weeks. Unfortunately, if these products are not in line with customer needs, they end up being returned, stuck in your closet and/or eventually thrown into landfill. This is not just toxic for the environment but also for the retailer’s’ bottom line! According to IHL, global fashion retailers are losing $100bn in revenue every year due to overstocks alone.
In our current digital age, billions of events are being processed online every second. With every search, product sku, social media follow, image view, we are leaving large datasets that can be analysed and improved. Through AI, “deep learning” algorithms learn from these data feeds to identify patterns, make predictions and perform tasks previously at the mercy of human error. As such, AI offers an opportunity to improve efficiency in fashion’s supply chain and consumer experience, that will dramatically improve fashion’s environmental impact. Here are a few ways in which AI can fix fashion’s sustainability problem.
From production to shipping to consumption, there can be multiple actors emitting various chemical or toxic gases, especially where you have very vast and distributed supply chains. By tracking data such as water consumption levels or fabric waste accumulated accross different stages of the supply chain, AI can alert retailers of which processes need to be improved. For example, EcoVadis helps companies create scorecards for their clients to improve their sustainability. Although they work with companies such as Nestle and Heinekan, consumer product companies are tapping into the EcoVadis database to identify more sustainable materials — fashion retailers can do the same. If you collect data on the consumption of water and oils for various fabrics used in a specific garment, you will be able to predict the fabric blends that would be both more sustainable and cheaper for future designs.
2. Removing inventory guesswork through data
The amount of guesswork that still exists in the fashion industry is astounding. Fashion brands and retailers are still using a limited amount of data to predict how much stock to produce. Sometimes it’s because they are not collecting the data, but most of the time they just don’t have enough data to make reliable decisions. According to IHL, the global fashion industry loses $210bn every year to overstocks and out of stocks but as much as 60% of these losses can be recovered if retailers know how to track, analyse and action the right data. For example, by analysing the browsing and shopping history of your customers, and benchmarking that against your competitors — AI can help retailers anticipate current shopper needs but also identify what they will want in 6/12/18 months time. If you are able to paint an accurate picture of the amount of clothes to stock, it means less products wasted.
3. Automating the creative process
Inspiration can come from anywhere but imagine if you could digitally capture that source of inspiration — a piece of art, a fabric print or a specific garment — and automatically recreate the design from scratch? This is another tested use case for AI. AI can help streamline the design process by allowing you to analyse multiple images and develop test designs based on those images before taking it into manufacture. In September 2017, Amazon developed an algorithm that can design clothing by processing and copying images from Instagram and Facebook. This gives them the potential to identify and recreate online trends for their own private labels automatically. One step further is to make the design process completely virtual with the help of AR/VR. Brands can build a database of different types of fabrics, colours and styles for the AI to generate various combinations that would work best for a particular style, all visualised using AR/VR. If you also add customer data into the mix — then retailers could recreate the optimal outfit for every season without having to put needle to thread until mass manufacture. Of course, I am not suggesting that we get rid of designers. AI will never be able to match the God-given creativity of the Karl Lagerfields and Alexandra Wangs of this world. However it can improve accuracy in the production phase which means less fabric and end garments are wasted.
4. Improving communications between offline and online stores
Earlier we talked about how AI could help retailers identify how much of each style to manufacture, but it can also help retailers understand how to distribute that stock. This is incredibly important as fashion retail becomes increasingly omnichannel. The majority of consumers will use both physical stores and online channels just to make a single purchase. However, retailers still silo ecommerce, mcommerce and brick and morter. AI can monitor purchase behaviour online and on mobile to identify and anticipate stock levels in stores within specific areas, so that if there is an increase in online searches for yellow jackets in London than Kent, the store inventory between the two locations can be adjusted to reflect the different demand levels.Add paragraph text here.
Short trend cycles, water consumption and waste plague the fashion industry but its not unfixable. What ties fashion’s environmental sins together is the amount of guesswork and lack of data being tracked or shared within the industry. Through AI, retailers can reduce the uncertainty in each stage of the cycle and anticipate customer needs/desires better so that their stock ends up in consumers’ wardrobes not in landfill. More importantly, AI will help retailers make better decisions about the sustainability of their fabrics, as well as the production and distribution of their garments, which will both reduce their costs and environmental footprint.
We just sent you an email. Please click the link in the email to confirm your subscription!
OKSubscriptions powered by Strikingly