Machine learning becomes fashionable: retailers use automated inventory management

Clothing and fashion make up the majority of the more than 530 billion euro the European e-commerce sector is worth. The fashion industry is buoyed by online marketplaces and new technology primarily applied to the improvement of the supply chain and the customer’s experience.

“The next generation of consumers are the iBrains,” said GFK Global Head of Fashion, Home, and Lifestyle, Marco Wolters in the E-commerce Europe report. This piece of research revealed that six percent of shoppers in Europe shop online every day, many of them benefiting from the latest technology and science advances.

“They are the first generation that is fully merged with smartphones and social media. These digital natives will be leading edge consumers that drive (mobile) disruption. And they will come fast. In 2020 the i-Brains will represent 40% of all European consumers,” added Wolters.

Deep learning allows retailers’ staff to find out if new items are in stock

For instance, many e-tailers are utilising a machine learning technique – deep learning – to simplify the shopping experience for both consumers and the company’s staff. Case in point is a new visual inventory search system developed by Microsoft that allows retailers’ employees take a photo of an item to see if it’s in stock.

Microsoft recently partnered with a major e-commerce retailer in the fashion industry to design an algorithm capable of identifying whether a newly arrived item was in stock using only a mobile phone image of the new item as a reference, so that retailers would be able to visually search their apparel catalogue based on photos in order to look up matching item.

For each new item that arrives, a retailer staff must determine if the merchandise matches an item already in stock. “One of the primary benefits of working in immediate proximity to our customers is the real-time, face-to-face feedback that we receive,” wrote Dan Behrendt, Technical Communications Director at Microsoft in a recent article on Entrepreneur.

“Inventory management for online retailers drives one of their largest expenses,” the Microsoft Developer Blog authors write. “Ultimately, our garment segmentation tool proved quite capable of segmenting photograph foreground from background and allowing us to eliminate the query image’s busy background,” continue the developers involved with this project.