Smart Inventory Management Shelve Using Neural Network and Google Cloud

Abstract

Stockout is a situation of disquiet in the current inventory for the retail organizations. This situation leads to 4% of the loss of sales. This paper forestalls Out-Of-Stock circumstances by cautioning when the stock level gets low or basic. The smart shelves are equipped with a base grid of light sensors which act like a pixel that detects the light level above them and the top sensor band of the ultrasonic sensor for observing the level of every item on the racks and send these data's persistently for analyzes to the neural system in a Raspberry pi through Arduino mega and then sends the processed data to the server for hosting the web application using the Django framework to monitor and re-stock alerts in real-time using the graphical user interface and further for the distributed storage. It achieves perfect efficiency in product replenishment. Our paper focuses to advance the store activity by limiting rack out of stock occurrences to improve the provider's visibility to stock status and to upgrade the purchaser shopping experience.