A Digital Twin for Decision Making on Livestock Feeding

This work is part of the IoFEED project, which aims at monitoring approximately 325 farm bins and investigates business processes carried out between farmers and animal feed producers. We propose a computer-aided system to control and optimize the supply chain to deliver animal feed to livestock farms. Orders can be of multiple types of feed and shipped from multiple depots by using a fleet of heterogeneous vehicles with multiple compartments. Additionally, this case considers some business-specific constraints, such as product compatibility, facility accessibility restrictions, prioritized locations, or biosecurity constraints. A digital twin–based approach is implemented at the farm level by installing sensors to remotely measure the inventories. Our approach combines biased-randomization techniques with a simheuristic framework to make use of data provided by the sensors. The analysis of results is based on these two real pilots and showcases the insights obtained during the IoFEED project. The results of this work show how the internet of things and simulation-based optimization methods combine successfully to optimize the feeding operations of livestock farms.