Efficient inventory management in wineries has been a recurring challenge for several years due to the complexity of the supply chain processes, the constant flow of assets and the peculiarities of the bottles themselves. Wineries and other players in the wine supply chain invest a lot of time and money in inventorying their bottles (currently they have the barcode which is common to several lots), causing in many cases errors or insufficient information.
Similarly, systems have not yet been integrated to enable comprehensive demand planning, which would have great potential as a commercial tool for retailers, and would provide valuable information to producers and distributors. In addition, there is still no standard that allows the individual identification of bottles and facilitates their control and mass tracking in wineries, wholesalers' warehouses, distributors, importers, or even retailers that are part of the HoReCa sector such as wine shops, specialized restaurants or hotels with a large sales capacity.

The objective of the project is to develop a comprehensive system for inventory control and monitoring, as well as demand forecasting in the wine sector. Specifically, it is intended to design and implement an intelligent warehouse management system capable of processing data coming from RFID readers, integrating mobile technologies, Internet of Things (IoT), ML and Cloud Computing.

It has been proposed to develop the initiative in two phases. The first, oriented to the identification of bottles and subsequent data processing for demand prediction, and a second in which computer vision systems based on artificial intelligence will be implemented to carry out the refinement of the identification and inventory processes. The aim is to build the winery of the future, and to reuse the model for specialized businesses involved in the distribution, storage, import and sale of wine.

The first phase refers to the present call for proposals, and the aim is to equip the bottles with technology that allows remote or automatic readings to be carried out. This will start by validating and integrating new IoT devices based on UHF RFID that are still in pre-industrialization stages. The second challenge will be to collect the information and implement machine learning models to carry out forecasts for procurement and stock management.
This solution will digitally empower SMEs that are positioned in the final stages of the supply chain, and market the product once it has been manufactured in the factory to the end consumer. The advantages that the solution will offer are the control of bottle stock in close to real time, the significant reduction of the inventory process and the reduction of errors.

Members

Funded with the support of