An example of AI and machine learning reducing demand forecasting errors by 15%

There’s a lot of hype around Artificial Intelligence and its emerging role in retail supply chain management. But hype can’t compare to real-world testing with actual data. Learn how one of Europe’s largest retailers tested AI-enabled demand forecasting using its own data and dramatically improved accuracy.

  • Read Video Transcript

    We were in contact with a famous Western European retailer who, like all of them, has an issue with finding people with the right skills – finding the forecasting specialists.

    They are already using our forecasting system, but they were looking for gains in productivity and some automation in decision-making within the tool.  So, they shared their data with us, starting with good quality results they already had, but that required a lot of effort to handle and tune the data. We ran the Demand ForecastingAI model on their data and benchmarked our system against theirs by simulating five months of working with their system. During those five months, those fake five months, the AI platform delivered not only the same results as they expected, without any human experts handling the data or tuning the system, but also delivered an improvement in the forecast accuracy beyond expectations as well as a decrease in forecast error by something like 15%.

    They were really excited by the possibilities this type of system will bring. This new approach changes the habit from trying to find the best way to tune the system to using a system that can improve results by injecting into it business and contextual data. It completely changed their approach to forecasting.

Speak with an expert

Just provide us with a few details and we’ll be in touch to discuss your needs