At the moment, grocery stores use very different methods for forecasting the sales of their products, and they use data from several different systems to do that. The amount of food waste varies greatly between different stores, but they all have a common goal set by the EU: they should cut their current amount of waste by half by 2030.
Individual data sources and the data they offer can be combined and harmonized in a single data stream to a compatible format. Data can be retrieved from the purchase management system or cash flow estimates of a store, for example, and condition data, such as weather forecasts, can be used to enrich that data. The harmonized data is imported into the store’s system, ideally allowing artificial intelligence to forecast the following day’s sales or the next week’s purchase needs for the purchasers.
A tangible example is the in-store bakery, where the demand for freshly-baked bread rolls is higher on a sunny Saturday morning than on a rainy Wednesday morning. The use of data results in cost benefits and improves food waste management.
To make this use case possible, data is enriched and harmonized in a reliable manner to match the customer’s needs.