Amazon Inventory Reconciliation Using AI

Abstract

Inventory management is critical to Amazon’s success. Thus, the need arises to apply artificial intelligence to assure the correctness of deliveries. The paper evaluates Convolutional Neural Networks for inventory reconciliation. The network detects the the number of items being carried, relying only on a photo of the bin. Convolutional performance is evaluated against Support Vector Machines, both non-linear and linear. Our Convolutional method performed over 3x better than random, and over 75% better than our best Support Vector classifier.

Keywords: Convolutional Neural Network, Deep Learning, Support Vector Machine, Radial Basis Function, Amazon Bin Image Dataset