POS Shelf Tracking for Accurate Purchase and Return Attribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In store systems where items are automatically registered based on consumer behavior, inconsistencies arise when multiple consumers interact with a display shelf simultaneously, leading to errors in identifying the correct purchaser or returner, which can result in discrepancies between intended and registered purchases.
Innovation Solution
A tracking-based point-of-sales system utilizing sensors and cameras to track customer movements, authenticate individuals, and register purchases or returns accurately, ensuring correct association of items with the right consumer through a processing unit that manages purchase and return activities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic item registration based on consumer behavior is implemented, then settlement processing efficiency is improved, but identification accuracy deteriorates when multiple consumers interact with the shelf simultaneously
Solution Approach 1:
The system segments the identification process into two distinct operations: (1) detecting product displacement/return actions at the shelf, and (2) tracking consumer movement to determine spatial relationships. By dividing the monitoring function between shelf sensors and tracking sensors, the system can accurately attribute actions to specific consumers even when multiple people are present.
Solution Approach 2:
The system introduces tracking information as an intermediary element that mediates between the product action detection and consumer identification. The tracking sensors provide intermediate data about consumer positions and movements, which the processing device uses to correctly identify which consumer performed which action, resolving the ambiguity when multiple consumers are near the shelf.
2Device complexity
If simple consumer identification near the shelf is used, then system complexity is reduced, but purchase registration accuracy deteriorates due to inability to distinguish between purchasing and returning actions
Solution Approach 1:
The system uses dynamic tracking of consumer movement to distinguish between purchasing and returning actions. By monitoring the temporal sequence and spatial trajectory of consumer movements relative to product actions, the system can dynamically determine whether a consumer is taking or returning items, improving registration accuracy without requiring complex static analysis.
Solution Approach 2:
The system implements feedback by continuously monitoring consumer positions and using this information to correct or confirm purchase registrations. When a product return action is detected, the system uses tracking information to identify the returning consumer and provide feedback to adjust their purchase list, ensuring accurate registration.
Data Source
AI summary
A tracking-based point-of-sales (POS) system includes a first sensing device, a second sensing device, and a processing device. The processing device is configured to perform location tracking with respect to the one or more customers around the product display section based on location information, perform a first operation to determine a purchasing customer who has performed a product displacement activity based on a result of the location tracking, and perform a product registration. The processing device is configured to perform a second operation to determine a product-returning customer who has performed a product placement activity based on a result of the location tracking, and perform a product deregistration. The processing device is configured to, upon the product deregistration causing an anomalous number of product in a purchase list of the product-returning customer, perform an anomaly operation.