By taking an AI-based machine learning approach the system reduces the risk of any picking/errors and streamlines the entire flow. Featuring a database that already has a million images, the system is designed to evolve and improve during in-the-field use.
 
Italiansped presents a new AI-powered machine learning system that takes spare parts warehouse management to the next level. At present configured on the SACMI spare parts warehouse - using an approach that can be extended to the Italiansped general warehouse - this project provides workers with cutting-edge support to eliminate potential errors in the spare part picking and shipping process.

Machine learning algorithms have, in fact, the task of accurately identifying materials and tracking them inside the warehouse. Key system features include a comparison of the code of the item being picked with a photograph of it taken on site. The result: unequivocal identification and immediate warnings in the event of any discrepancies.

In keeping with the machine learning approach, the system is designed to improve its own performance over time thanks to a vast database that already contains over 1 million photographs. The operator, in fact, takes a photo each time he/she picks an item: the image is compared to the one in the database and, if it matches, is filed for subsequent comparisons.

Tests carried out so far demonstrate that the system achieves optimal performance with just 15 ‘valid’ database photos of a specific warehouse item. Already, with the current dataset, which includes 80,000 pick photos and 850,000 packaging photos, the system is ready for in-the-field use and further learning-based improvement.

Italiansped currently manages a spare parts inventory of 30,000 different item codes and picks over 250,000 items for sale and shipment to customers. With this latest system, Italiansped aims not only to ensure an error-free picking process: it also intends to improve warehouse management in keeping with SACMI's general strategy on research and the implementation of new enabling technologies to boost process efficiency and streamline business flows.