The importance of big data in the fashion industry

Riddhi Doshi

Advanced big data and artificial intelligence technology help luxury businesses track and study their consumers’ preferences and shopping behaviour, and, in turn, enable the creation of a smart retail environment, wherein every consumer is offered a different and relevant shopping experience.

Abdullah Abo Mihim, Postgraduate Program Leader, Istituto Marangoni School of Fashion, London, deconstructs the current trends, usage and perils of big data and AI in an e-mail interview. The teacher of Economics, Finance, Strategy, Luxury Management, Project Management, Strategic Fashion and Business of Fashion also held a talk and presentation on the same at Mumbai’s Istituto Marangoni recently.

How is big data advancing AI for luxury businesses? The availability of data provides various opportunities to use more data in the context of machine learning. Big data can take AI to the next level, providing the latter with powerful tools to analyse large sums of data on the prescriptive and predictive sides of forecasting and analysis.

India has been doing a great job in endorsing various innovations in the fields of AI and Analytics, powered by the availability of new technology, cheap labour and start-up supported projects in such fields. AI is being used by various fashion designers, not only in trend forecasting but in producing collections embedded in cognitive technologies.

How is the integration of big data and AI helping luxury brands?
As data is collected from multiple sources, AI provides ample opportunities to understand the behaviour of many people covered through different data sets. This provides brands with a strong tool that helps them forecast and predict aspects of consumption and trends of behaviour. Luxury brands are using AI to gain high-quality information on their customers, which potentially helps them create personalised products and customise their offerings online.

Enabled by big data and AI, what does the future of retail spaces look like? What are smart environments and how can these be created?
Thanks to machine learning, data can now generate new tools for business analytics, which can provide a better understanding of behavioural trends within a smart store environment and in the contexts of relationships with vendors or suppliers. In smart store environments that are powered with AI and Analytics, there is a flow of real-time information that provides the brands with valuable information on conversion rates, heat maps, engagement with the store design and live monitoring of stock levels. This has advantages across the whole supply chain, specifically in areas dealing with a shorter cloth range and in situations where points of demand and supply are closer to each other.

What is cognitive couture?
Fashion collections that rely on cognitive computing in various creative and design processes, using deep learning algorithms and neural networks to compare various sets of data such as colours, images and even blogs. Such innovation could enable designers to analyse large sets of data in seconds, adding a predictive edge to the analysis.

How can AI and big data help brands become sustainable? Data can improve the quality of information shared between different stakeholders. Eventually, suppliers can be encouraged to be more responsible and aware while engaging in various business transactions. AI, especially when combined with Analytics, can assist brands in conducting an in-depth analysis of their cost structure, improve margins and reduce over-production waste. This lets brands better understand their consumers’ behaviour, estimate production levels, manage inventories efficiently and source accordingly.

Do big data and AI play a role in packaging as well? The opportunities arise from the ability of Predictive Analytics and AI to better match the powers of supply and demand, and then responding to a supply chain that is getting shorter, in an efficient manner. We already have smart warehousing; there is more potential for it with the developments in the area of machine learning.

Is the fear of robots taking away people’s jobs valid? Humans will always be superior to machines in a lot of decision making processes. Nevertheless, the latter is more powerful in crunching large sums of data and analysing multiple sums of variables simultaneously and in real-time. Robots could destroy jobs in the short term but create some in the long term. It all depends on the extent of interaction between humans and machines.

What are the perils of big data and AI? Big data has a lot of problems, mainly in areas of heterogeneity, complexity and privacy. We are and will continue to face problems of data overload, which, in many cases, could be correlated with some aspects of mental health problems.

It is said that AI contributes to gender disparity. How can this problem be resolved? It is important to invest in programmes that increase women jobs in the tech industry. The UK’s governing bodies, working closely with universities, are implementing new initiatives targeted at growing the number of females working in AI and Data Science. Education is key and must begin at the school level moving up to higher education levels.

Have consumers accepted AI?
Consumers are excited about new technologies, but also equally confused, and worry about its impact. The lack of knowledge about how AI works and what it represents generates negative feelings, which is consequently mixed with brand awareness. Businesses should build a culture around AI that starts with an in-depth study of its effects on various stakeholders including consumers. The next step would be looking at various models and strategies to inform the consumer about the added value of AI in the context of their daily consumption, brand experience and personalisation.