Self-Checkout Workflow Management Using Predictive Guidance

Solution Overview

Problem

Self-checkout kiosks face challenges in efficiently processing transactions due to the need for end-user intervention, which can lead to increased computational overhead and reduced transaction processing speed.

Innovation Solution

Implementing a system that uses computer vision and workflow interaction metrics to predict item and payment types, and adjust communication levels based on end-user interaction, thereby reducing the need for end-user intervention and streamlining the transaction process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If the self-checkout kiosk requires end-user intervention for each transaction step, then the system can guide users through the workflow, but the computational overhead increases and transaction processing speed decreases

Engineering Contradiction:

Improveuser guidance through workflowVSAvoidtransaction processing speed

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The system performs preliminary actions by predicting the next workflow step and preparing communication messages in advance. The machine learning model anticipates what information the user needs next based on current transaction state, allowing the system to proactively present guidance rather than reacting to user inputs, thereby reducing processing delays

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system enables self-service by automatically determining the next workflow step and generating appropriate communications without requiring explicit user direction. The machine learning model autonomously navigates the transaction workflow, making decisions about what information to present and when, reducing the need for continuous user intervention while maintaining operational guidance

Inventive Principle:

Principle #25Self-service

2Ease of operation

If the system waits for end-user input at each step, then user control is maintained, but computational overhead increases and processing throughput decreases

Engineering Contradiction:

Improveuser control over transactionVSAvoidtime delay for end-user input

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The system performs preliminary actions by predicting the next workflow step and preparing communication messages in advance. The machine learning model anticipates what information the user needs next based on current transaction state, allowing the system to proactively present guidance rather than reacting to user inputs, thereby reducing processing delays

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms where the machine learning model continuously monitors user interactions and adjusts its predictions accordingly. This feedback loop allows the system to learn from user behavior patterns and improve its predictions, reducing the time needed for user input while maintaining user control through adaptive guidance

Inventive Principle:

Principle #23Feedback

3Ease of manufacture

If the kiosk uses traditional step-by-step workflow, then the transaction process is simple to implement, but the number of steps increases and processing efficiency decreases

Engineering Contradiction:

Improvesimplicity of workflow implementationVSAvoidtransaction processing throughput

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The system performs preliminary actions by predicting the next workflow step and preparing communication messages in advance. The machine learning model anticipates what information the user needs next based on current transaction state, allowing the system to proactively present guidance rather than reacting to user inputs, thereby reducing processing delays

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes parameters by dynamically adjusting the workflow based on predicted user needs and transaction state. The machine learning model modifies the sequence and timing of workflow steps according to real-time conditions, optimizing the number of steps required while maintaining implementation simplicity through automated adaptation

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20250148885A1Self-checkout workflow management system

Publication Date: 2025.05.08 TOSHIBA GLOBAL COMMERCE SOLUTIONS INC

AI summary

Techniques to reconfigure self-checkout kiosks based on workflow interaction metrics. A measure of interaction is determined for at least a portion of a workflow during a transaction at a self-checkout kiosk. The self-checkout kiosk is configured to output one or more communications via one or more graphical user interface (GUI) screens associated with the workflow. A communication level is set for the workflow for the transaction. The communication level is determined based on the measure of interaction and without requiring an intervention. A next communication to output is determined based on the communication level. The transaction is executed at the self-checkout kiosk.