Model-Based Self-Scan Validation to Reduce Rescan Rate and Shrinkage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Do-it-yourself checkout solutions, such as self-checkout terminals, handheld devices, and mobile shopping, lead to higher shrinkage rates due to customer errors and fraud, which existing retailers struggle to address effectively without increasing labor costs or frustrating customer experiences.
Innovation Solution
Implementing a model-based data validation system that uses machine-learning generated models to determine when a rescan of items is required, considering factors like customer behavior, transaction history, and item combinations, to identify potential un-scanned or mis-scanned items, thereby reducing the need for unnecessary rescans and improving fraud detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If rescanning is performed on all self-scan transactions to detect fraud and errors, then detection accuracy improves, but labor costs and customer frustration increase
Solution Approach 1:
The system changes the parameter of rescan probability from a fixed value to a dynamic value based on multiple factors including customer behavior patterns, transaction characteristics, item risk profiles, and historical fraud data. This allows the system to adjust the likelihood of requiring a rescan for each transaction individually, optimizing between detection accuracy and customer experience
Solution Approach 2:
The patent introduces an intermediary validation system that sits between the self-scan transaction and final processing. This intermediary uses machine learning models to assess risk and selectively route transactions for rescanning, rather than applying uniform rescanning to all transactions. The intermediary acts as a filter that identifies high-risk transactions without disrupting low-risk ones
2Reliability
If rescan rate is increased to catch more fraud attempts, then shrinkage reduction improves, but customer experience deteriorates
Solution Approach 1:
The system applies different levels of validation scrutiny to different transactions based on their specific characteristics. High-risk transactions (unusual item combinations, new customers, suspicious patterns) receive intensive validation with higher rescan probability, while low-risk transactions proceed with minimal intervention. This local differentiation optimizes shrinkage reduction where needed while preserving customer experience where safe
Solution Approach 2:
The system performs preliminary risk assessment and customer behavior analysis before the transaction is completed. By evaluating multiple risk factors in advance and generating a rescan probability score beforehand, the system can make informed decisions about which transactions require validation, rather than applying uniform post-transaction rescanning
3Ease of operation
If model-based validation is implemented to reduce unnecessary rescans, then customer experience improves, but system complexity increases
Solution Approach 1:
The validation system is designed to operate autonomously using machine learning models that automatically analyze transaction data, customer behavior patterns, and risk factors without requiring manual configuration or intervention. The system self-adjusts rescan probabilities based on learned patterns from historical data, reducing the operational burden on staff while maintaining sophisticated validation capabilities
Solution Approach 2:
The patent replaces manual, mechanical rescanning processes with an automated digital validation system. Instead of employees physically rescanning items based on simple rules, the system uses machine learning models to analyze multiple data dimensions and automatically determine validation needs, substituting computational intelligence for manual labor and simple mechanical processes
Data Source
AI summary
Various embodiments herein each include at least one of systems, methods, and software for model-based data validation to identify when self-scan checkout data requires validation. Some embodiments, in the form of a method includes receiving, via a network from a self-scanning device, a self-scan dataset of items for purchase within a purchase data processing transaction and evaluating the self-scan dataset to determine whether to require a rescan of items represented in the self-scan dataset. In such embodiments when a rescan is determined to be required, the method includes transmitting via the network to at least one of the self-scan device and at least one device of a store employee data indicating a rescan is required. However, when a rescan is not determined to be required, the method includes permitting the purchase data processing transaction to proceed.