A MACHINE LEARNING BASED FRAMEWORK FOR OPTIMIZING DRONE USE IN ADVANCED WAREHOUSECYCLE COUNTING PROCESS SOLUTIONS

Abstract

In each warehouse, managing inventory is an ongoing cycle that is a part of the supply chain. Warehouse inventory management is a process that involves receiving, storing and tracking inventory inside a warehouse followed by optimizing storage spaceand costs. Optimized inventory management provides improvements in fulfillment, shipping and the customer experience. In large warehouses, an inventory check generally means that staff members are lifted to high shelves where they physically reach and examine each standardized pallet, which is a tedious, expensive, risky and energy ineffective process. To increase performance, optimize work and reduce labor costs in contemporary Warehouse4.0, the use of autonomous Unmanned Aerial Vehicles (UAV) for the cycle counting process and others is considered in this paper. Using UAVs for cycle counting provides a revolutionary solution to scanning pallets in a warehouse, using the latest drone platforms, hardware, software, scanning and communications technology. To use drones with integrated camera systems in the Warehouse 4.0 concept, efficient machine learning algorithms are needed in tasks such as scanning barcodes on each pallet, recording the location of each item, drone route optimization and many others. The framework proposed in the paper provides faster and more reliable operation but opens novel problems and challenges that need to be solved.