A computer vision based approach to reduce system downtimes in an automated high-rack logistics warehouse

EUR-pallets circulate worldwide by the millions and play a vital role in enabling short and long-distance logistics. Due to increasing automation in logistics and warehousing, monitoring and assessment of the pallet quality is a key aspect for ensuring high warehouse availability. Automated warehouse equipment requires high system availability and reliability, since manual intervention in case of outages is cost intensive and laborious. In this paper a data-driven approach relying on computer vision for assessing the pallet quality will be introduced, with the objective of reducing maintenance effort and system downtimes caused by defective pallets in an automated high-rack logistics warehouse.