The fashion industry’s impact on the environment is significant, and being sustainable is a key priority. There is a growing awareness of the environmental and social issues caused by fast fashion, and retailers are increasingly adopting initiatives and investing in becoming more sustainable. Technology will be a key enabler, supporting retailers in this journey. It can help across the entire fashion value chain, from design to selling, to become more sustainable.
How the Fashion Industry Impacts the Environment
Fast fashion has a huge impact on the environment. As per estimates, the global fashion industry is responsible for ~10 % of human-caused greenhouse gas emissions and ~20 % of global water pollution. 93 billion cubic meters of water, enough to meet the needs of five million people, is used by the fashion industry annually, contributing significantly to water scarcity in some regions. Around 20% of industrial wastewater pollution worldwide originates from the fashion industry.
Reports suggest that more than 50% of fast fashion clothes are discarded within a year of purchase; 30% of global clothing is never sold; and another 30% is only sold at a discount. This has resulted in mounting piles of unsold stock, with several large retailers carrying unsold inventory worth billions of dollars. It is estimated that of the 100 billion garments produced each year, 92 million tons end up in landfills. It is evident that the fashion industry is plagued by the problem of overproduction.
• Overproduction: Traditional fashion retailers operate largely on a push-based model characterized by long product lead times and huge production quantities planned and produced season after season. The conventional approach to sourcing has been to design assortments and leverage a trusted supplier network for production, typically engaging suppliers in the Far East to take advantage of lower costs. This one-size-fits-all conventional approach to fashion supply chain planning, where make, buy, and move take place in a lean and operationally efficient way, has a flip side: Overproduction.
Overproduction occurs when brands produce significantly more garments than they sell to their customers. Fashion overproduction is estimated to range somewhere between 35-40% in any given season. It simply means that to sell 100 garments today, brands end up producing ~140 garments. Overproduction leads to surplus stock and tons of unsold inventory.
• The Impact of Overproduction: Overproduction adversely impacts both the retailer as well as the environment. Here are a few dimensions where there is a marked impact –
• Revenue:
• Lost sales due to the inability to cater to consumer demands in a timely manner.
• Reduced full-price sell-through due to aggressive discounting to deal with stockpiles.
• Lost sales due to reduced space available for new stock.
• Margins:
• Inventory write-offs (deadstock).
• High inventory management costs (storage, transport, handling etc.).
• Markdowns have a significant effect on margins.
• Capital tied up in unsold inventory.
• Customer Experience:
• Excessive markdowns tarnish the brand image.
• Difficulty in acquiring new customers without inspirational offerings.
• Lack of in-demand items likely to push consumers to competition.
• Sustainability:
• Overproduction is leading to a clothing waste crisis, and large fashion retailers are enduring a bad reputation for their inability to act swiftly.
How Technology Can Help Solve the Issue of Overproduction & Improve Sustainability
Technology is a key enabler in helping retailers address the problem of overproduction. For example: Retailers can leverage AI & ML to better interpret demand signals from the market based on customer preferences, emerging trends, and competitor data. The retailer also needs a supply chain strategy to plan and produce these garments, in line with the demand, in a timely manner. This is where a Bimodal supply chain strategy can help, by allowing a differentiated approach based on the characteristics of the assortment of the product, for production and introduction in the market.
So, What is a Bimodal Supply Chain?
A bimodal supply chain helps businesses strike the right balance between operational efficiency and agility in meeting evolving business needs. This model empowers companies to leverage the traditional strengths of their supply chain network to keep costs low while allowing them to embrace emerging technologies to keep pace with new-age competitors.
A supply chain strategy based on the nature and needs of the assortment category determines the approach chosen by the fashion retailer for a given product. Fast-fashion retailers can adopt a bimodal approach as follows:
• Linear supply chains for traditional assortments (Mode 1)
• Agile, responsive supply chains for trendy fashion assortments (Mode 2)
While Mode 1 offers operational efficiency and cost savings, Mode 2 is characterized by a shorter time to market, demand volatility, shorter lifespan, and low to medium order quantities, ideal for assortments that require innovation and high responsiveness.
How Technology Helps In Executing a Bimodal Supply Chain
Digital technologies are key enablers in setting up a bimodal supply chain. In addition to AI, ML, and generative AI, some of the other technologies that can help in this transformation are RFID, smart store technologies, digital twins, and robotics.
Areas where technology helps include:
• Inspire design through early and continuous detection of trends based on social signals.
• Enable in-season assortment refresh by moving away from traditional seasonal planning.
• Lower the lead time for product development by recommending countries and suppliers for specific products and by automating order execution.
• Rapid prototyping and quality control during the design & development phase.
• Lower shipping and logistics lead time by near-shoring suppliers and evaluating new modes of transport for suppliers in distant locations.
• Help meet sustainability goals with lower production lead time, wastage, and water consumption enabled through automated, smart factories.
• Maximize sales by increasing space allocation for trending items based on sales and demand patterns.
Challenges in Transitioning to a Bimodal Supply Chain
Adopting a bimodal approach to the supply chain will require retailers to implement changes across the entire value chain spectrum. Retailers will need to evaluate their supply chain to determine where to apply Mode 1 and Mode 2, and transform the way they plan, produce, and distribute assortments by driving digital to the core of their business and harnessing AI in the supply chain at scale.
Some of the key challenges and considerations are:
• Shift in Mindset: Pivoting to a bimodal strategy requires enterprise-wide change, including a change in organizational culture and an agile approach to adopting new ways of working with people, processes, and technology.
• Robust Data Foundation: Build a high-performance data and AI-ML organization to make more informed business decisions and deliver best-in-class customer experiences.
• Right Pricing Strategy: A bimodal supply chain will increase the price points due to higher operating costs, and retailers need to get their pricing strategies right so that margins are not negatively impacted.
• Strong Supplier Collaboration: A more integrated and tighter collaboration with suppliers is required to ensure that the bimodal strategy is a success. Retailers need to build a strong ecosystem partner hub driven by well-established processes and systems.
• Supply Chain Transparency: Supply chain control towers or digital twins offer end-to-end, real-time visibility across the entire network, and help mitigate disruptions and risks before they escalate.
Future Technologies for the Fashion Industry
There are several exciting new technologies that will be transforming the industry in the coming years:
• Gen.AI: This has really captured the imagination of businesses around the globe with the potential it offers. With widespread application across the spectrum, this technology trend will be one to watch out for.
• RFID: Although this technology has been around for a significant period, retailers are now beginning to truly leverage this to transform the landscape by enabling use cases like end-to-end item level traceability and in-store SKU location.
• AR, VR and Metaverse: Brands are now looking to provide immersive, experiential environments for customers to explore and engage with them. This area will see significant cant investments from retailers in the coming years.
• Digital Twin Technology: This will transform the product development and sampling area allowing rapid prototyping and quality control.
• Robotics: This will help fashion retailers significantly improve their efficiency as robots can perform tasks like – inventory management, automated picking and packing, etc.
• In-Store Technologies: The role of the store is evolving and in the coming decade, stores will emerge as the hub of omnichannel retail offering personalized and immersive experiences to the customers as they engage with the brand.
Top 3 Areas Where Gen AI Can Help Fashion Retailers
• Design Inspiration & Product Development: GenAI has ample use cases in the product creation area where it can assist designers in the product development area not only by providing them with design inspirations but also in creating patterns & product sketches.
• Supply Chain Operations: GenAI can help bring in efficiency in the supply chain operations through predictive operations and risk mitigation. Models can also be used in supplier selection and negotiation processes.
• Fashion Brand Storytelling & Customer Experience: GenAI can help retailers in inspired storytelling around their product and their brand, helping them connect better with their customers. Models will also accelerate content development.
Legal & Ethical Issues To Consider While Using GenAI
The use of GenAI raises several legal and ethical issues and businesses need to be wary of these challenges while charting their GenAI journey. Some of the challenges that need to be addressed are:
• Intellectual Property & Liabilities: There might be concerns around who owns the output generated by the LLM models if they are used in the product design and creation. AI-generated creative works might be subject to intense scrutiny of copyright and IP rights.
• Bias & Fairness: There are possibilities where the GenAI models can be impacted by the biases present in the training data leading to the generation of harmful content. Fashion designs and content produced by these GenAI models will require careful curation to make sure that they adhere to brand guidelines.
• Data Privacy & Security: LLM models require vast amounts of data for training purposes. Retailers need to be cautious while handling the personal data of consumers while they look to personalize their fashion offerings and communication.
