Currently, WAIR focuses on the inventory process, specifically the replenishment process. By making the replenishment process intelligent, it changes from being a static process to a dynamic process. The replenishment process is often based on minimum to maximum rules or covering periods, which result in a static replenishment process for all stores and SKUs. WAIRs solution – the AI Replenisher creates a sales forecast with >95% forecast accuracy, ensuring that users have the stock in the store where it sells (with customer demand). By that, they increase the full-price sell-through rate, which results in a higher revenue margin and reduced waste, also known as the end-stock.

WAIR’s ForecastGPT-2 (not related to an LLM or Chatgpt), the technology behind the AI Replenisher already has gained popularity in the market for retailers. It combines pooled industry data, advanced deep learning technology, powerhouse computing, domain expertise, and insightful features to build an AI model that’s best in class. This is reflected in the number of parameters their models contain. It indicates its ability to learn, compute, and make accurate predictions. They encapsulate the ‘knowledge’ the model acquired from the training data.

Taking Fashion Inventory Management to A Higher Level

WAIR’s deep learning model is unique because not only do they use the most advanced AI there is but they implement the “pooled data” approach, where the company combines data of all clients, partners, and third-party data into one standardized database resulting in large amount of data gathered in one place. The patterns and insights between these data points (features and parameters) are considered when calculating the likelihood that an SKU will sell at a specific location. Then general task of the model ForecastGPT-2 is to generate forecasts and with that the KPI is “forecast accuracy” and has replenishment forecast accuracy of at least 95% accurate, which is an actual breakthrough in the retail industry field of work. Only 1% of the market uses deep learning models as intelligent as WAIR – companies like SHEIN and Amazon.

The credit for such a revolutionary concept and ideology goes to Mitch van Deursen, the second-generation owner of a large fashion retailer and also a serial entrepreneur in technology companies. Upon assuming the company’s leadership, he quickly realized the potential for more intelligent, more cost-effective operations that could increase revenue and profit margins. Recognizing the ongoing pressures facing the retail sector, Mitch began to explore the use of AI to optimize efficiency and profitability in the early days of AI.

Initially, WAIR was conceived as a collaborative project with BrainCreators, an Amsterdam-based company specializing in machine learning and AI. Since then, it has evolved into an independent entity. Nevertheless, WAIR maintains a strong focus on serving the needs of retailers, putting them at the forefront of its efforts. In addition, they are committed to ongoing research and development to stay at the cutting edge of technology.

Our level of intelligence and amount of data has proven that we have chosen the right path. From the early days we decided to fixate our company on only “the layer of intelligence” and not on the process(the process is handled in the ERP), as we believe the processes are already rightfully in place within the architectures of retail. The processes simply need to be made intelligent and that is where the truth value is, which makes us specialized and focused on creating sales forecasts with the highest forecast accuracy. We have best in class AI evangelists and researchers + 100+ years of experience in retail combined with a unique strategy compared to other vendors.

Addressing Client Needs

For instance, Daka, a popular sports retailer in the Netherlands, operates 18 stores nationwide and has an online shop. The sports retailer’s central warehouse experienced stockouts, unable to replenish empty stores, while other stores had excessive inventory, leading to mid-term stockout in season. This necessitated inventory redistribution, resulting in a costly and time-consuming process.

Daka implemented WAIR’s AI Replenisher, equipped with a fashion retail-specific Deep Learning model, to predict and address the seasonal influences on their sales accurately. It considers external data sources and retailer-specific information, such as sales data, product details, demand, and individual store performance, to make SKU-level sales predictions. This allows Daka to allocate stock based on revenue potential and replenish its physical stores, which are heavily influenced by seasonality.

By continuously learning from customer behavior, the model suggests incremental replenishment adjustments that significantly improve sell-through rates and reduce overstock and enhance revenue. The AI Replenisher seamlessly integrates with a company’s ERP software, allowing teams to harness the full capabilities of WAIR directly from their familiar ERP interface.

Daka is only one of the numerous examples that portray the brilliance of WAIR. For the days to come, Mitch and his team will concentrate on the company’s AI Replenisher, and in September 2023, they will launch the AI Product Describer, followed by the AI Redistributor in December; after that, the AI Channel Reserver is next that can ensure the highest revenue margins by predicting what can increase sales growth, followed by Initial Distribution and Dynamic pricing. “We see that the information of the models output has been really accurate, which makes us now able to fast forward on building solutions on top of that information.

 “We are an Artificial Intelligence and Deep Learning company with significant retail experience. We bet big on AI a couple of years ago because we believed not only SHEIN or Walmart should have access to these technologies. We deliver technology only these tycoons possess. We know deep learning; we know reinforcement learning and we know retail from the inside out. Luckily we are exactly on the right spot and miles ahead” concludes Mitch.