Artificial Intelligence is taking over retail and has been used across the entire product and service cycle. Right from pre production to post sale, retail players are leveraging AI in different forms in order to bring automation. The following article sheds light on different examples where AI is successfully integrated into key retail functions. Let’s go through them one by one.
It is a quintessential area where AI can transform efficiency. As per the Capgemini report, retailers have experimented with optimal route plans in supply chain over their many years of operation. With AI, each optimized route plan is saved for an algorithm to learn and improve its suggestions. UK-based Tesco, China’s JD.com and Alibaba are some of the retailers implementing AI-based optimized routing. Indian e-commerce company Flipkart which was acquired by US retail giant Walmart this year is using machine learning (ML) to arrive at a structured address classification system for order deliveries, addressing the challenge posed by India’s unstructured postal address systems. The ML solution classifies and resolves inconsistencies with a 98% accuracy rate.
AI-powered visual aided picking (within warehouses/ distribution centers)
As report indicated, Physical forms of AI are now extending beyond chatbots to robots in the warehouse. The AI robots markets is expected to grow at 28.8% CAGR between 2017 and 2023. In Ocado’s warehouses, robots also collaborate with each other through visual recognition in addition to autonomous picking and packing jobs. This allows them to come together or split up to fulfill a typical 50-item order in minutes. American retailer Kroger has now partnered with Ocado to build AI-powered warehouses and up their grocery delivery capabilities. Grupo Casino, Kroger, and Morrisons are some of the other retailers partnering with Ocado to build smart warehouses.
To predict customers’ purchasing patterns over the next 30 days, German-based ecommerce player Otto analyzes about three billion past transactions and 200 variables, including sales, searches, and weather conditions. The AI system predicts customer purchases at 90% accuracy, thereby reducing product returns by over two million items a year.
AI for stock replenishments
AI-driven insights from varied data sets offer significant scope to automate stock replenishments. UK retailer Morrisons is working with Blue Yonder, a tech firm, on an AI deployment that analyzes different data sets: internal data sets (such as sales) along with external data sets (such as weather patterns or public holidays). This allows the company to predict demand down to the individual store level and then automates the product orders. As a result, shelf gap was found to be reduced by 30% during trial sessions.36
AI for assortment rationalization (rationalizing SKUs in range)
Global fashion retailer H&M faced a significant challenge: $4 billion worth of unsold stock. To address this, it applied machine learning on different data sets, such as returns, purchases, loyalty card, search results and store receipts. This aim was to customize assortments to each individual store, reversing the previous practice of merchandising based on past sales. The number of SKUs reduced by 40% as a result
(The above mentioned article is based on Capgemini report titled “BUILDING THE RETAIL SUPERSTAR”)