Industry Technologies
By Yvonne Heinen-Foudeh, Senior International Correspondent
Technology Predicts - Humans Make the Decisions.
In a market that’s constantly defying people’s predictions, and where the backdrop of retail changes almost daily, the ability to plan based on accurate, timely data is key to securing the future — always provided that the data is available in digital format, of course. The complex and mathematical AI technology enables a prediction quality far superior to human possibilities – even if it is 'only' regarding the time required for processing and analyzing gigantic amounts of data.
The fashion market with its short product life cycles is the prime example of the need for completely new planning approaches. More and more brands and retailers are juggling a complex mix of channels across the digital, physical, franchise, and marketplace outlets. With that said, the shape of demand is constantly changing. Current disruptions along the supply chain reinforce the effect.
The previous retail business model was turned on its head well before COVID exposed its weakest links, which means that the traditional approach to forecasting and planning – intuition based, channel-focused and inflexible – has been out of lockstep with the realities of retail for far longer than many organizations realize.
Traditional planning had been oriented around one particular objective or channel and has been based on analysis of historical information from prior seasons. “A contemporary planning strategy may leverage same historic sales performance, but by contrast also places comparable importance on eCommerce behavior patterns, footfall data, material wastage, returns figures and product data housed in PLM and other systems, sourcing, and supply chain indicators” explained Gavin Fallon, General Manager for Board International, to the author.
As one of the leading providers of intelligent forecasting and decision-making software, Board International has established itself as well with retail and brand customers in the fashion industry - including Burberry, H&M, and Puma. For features such as complete reporting on business performance in real-time view and dynamic, location-independent financial planning and analysis reports, a total of around 2,000 users worldwide already use the predictive, AI-driven forecasting and functions by the developers from Boston, Massachusetts (USA) with European headquarters in Chiasso, Switzerland for their related decision-making processes.
Artificial boosts natural intelligence And this is precisely where AI comes into play. Where large or even gigantic amounts of data are processed and analyzed, artificial intelligence can add considerable value. What basically applies to artificial intelligence today (and according to mathematicians and IT experts for the next 40 to 50 years) also applies here: The high efficiency - assuming appropriate computing capacities - cannot be topped. But when it comes to cognitive thinking, and when emotions and creativity are required, natural intelligence is superior.
Shelf-checking AI to improve product availabilityRecognizing the market and opportunities, the Alphabet subsidiary from Mountain View, California has jumped on the bandwagon of AI-based software tools as well for their Cloud-managed Database Services: Google Cloud just introduced an AI-powered shelf-checking solution to enhance e-commerce sites for better customer shopping experiences and to help stationary retail formats to transform their systems concerning product availability, visibility of what shelves actually look like, and to understand where restocks are needed.
.jpg)
Puma uses Board International's AI-based forecasting and data analytics platform in the fashion market. In the picture: view into the flagship store of the activewear brand at the company headquarters in Herzogenaurach, Germany.
Built on their Vertex AI Vision and powered by two machine learning models — a product recognizer and tag recognizer — Google Cloud claims the shelf-checking AI enables retailers to solve a very difficult problem: how to identify products of all types, at scale, based solely on the visual and text features of a product, and then translate that data into actionable insights? Right now the project finds itself in the pilot stage and is expected to be rolled out worldwide in the coming months.
“With the add-on feature Google Cloud users will not have to expend time, effort, and investment into data collection and training their own AI models”, says the accompanying statement from the U.S. tech company on the launch of the prototype. “Leveraging Google’s database of billions of unique entities, our shelf-checking AI can identify products from a variety of image types taken at different angles and vantage points.” Furthermore, Google emphasizes the high degree of flexibility in the types of imagery that get supplied for the shelf-checking. Imagery can be used from a ceiling-mounted camera or an associate’s mobile phone, we convince ourselves in a test drive. Equally good they could get generated from a store-roaming robot on shelf-checking duty.
“With the add-on feature Google Cloud users will not have to expend time, effort, and investment into data collection and training their own AI models”, says the accompanying statement from the U.S. tech company on the launch of the prototype. “Leveraging Google’s database of billions of unique entities, our shelf-checking AI can identify products from a variety of image types taken at different angles and vantage points.” Furthermore, Google emphasizes the high degree of flexibility in the types of imagery that get supplied for the shelf-checking. Imagery can be used from a ceiling-mounted camera or an associate’s mobile phone, we convince ourselves in a test drive. Equally good they could get generated from a store-roaming robot on shelf-checking duty.
"Google's cloud-managed database services support Macy's in creating a compelling shopping experience for our customers. Instead of local operations, it's also helping us accelerate digitalization at an uncertain time for our industry and our company."
