Accurate demand forecasting is a key business need. Demand models typically incorporate many demand factors like product, price, promotion, or season. The local weather, however, is one of the more crucial drivers in many consumer facing sectors such as retail, e-commerce, food & beverages, travel & tourism. Including weather in automated forecasting processes improves forecast accuracy, steadies operations and thereby reduces costly overstocking or revenue losses from out-of-stock situations. Smart weather-driven forecasting needs high-quality weather data as well as lots of experience in weather modelling. Dr. Stefan Bornemann, COO at wetter.com, will explain (i) which data and modelling techniques are used to create daily sales forecasts considering weather effects and (ii) which challenges occur when building an automated cloud-based forecasting solution for hundreds of products or locations.
Dr. Stefan Bornemann (wetter.com GmbH)