AI Store Guidance Using Congestion-Aware Customer Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for guiding customers in retail stores do not effectively address congestion issues, as they are typically designed for facilities like theme parks and do not consider the unique challenges of retail environments.
Innovation Solution
An in-store customer guidance system comprising a store server, camera, and mobile terminal that identifies congestion levels, determines customer status, and generates prompts to guide customers to less congested areas using a generative AI.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If a guidance system is implemented to direct customers to less congested areas, then customer waiting time is reduced and shopping experience is improved, but system complexity increases due to the need for real-time congestion monitoring and AI-based routing
Solution Approach 1:
The patent introduces a store server as an intermediary component that centralizes congestion data collection, analysis, and guidance generation. The server receives congestion information from multiple cameras, processes this data through AI models, and generates optimal routing guidance for customers. This intermediary architecture manages system complexity by consolidating complex functions in a single centralized component rather than distributing complexity across multiple customer-facing devices.
Solution Approach 2:
The system implements real-time feedback loops where congestion data from cameras continuously updates the store server, which then dynamically adjusts guidance recommendations. The server monitors current congestion levels, compares them against historical data, and provides updated routing advice to customers. This feedback mechanism enables the system to adapt to changing store conditions automatically, reducing waiting time without requiring manual intervention.
2Measurement precision
If real-time congestion monitoring is implemented using cameras and AI processing, then guidance accuracy is improved, but energy consumption increases due to continuous image processing and data analysis
Solution Approach 1:
The patent implements periodic sampling of congestion data rather than continuous real-time processing. The store server collects congestion information from cameras at predetermined time intervals (e.g., every few seconds or minutes), processes this batched data through AI models, and updates guidance recommendations accordingly. This periodic approach maintains sufficient measurement precision for effective customer guidance while significantly reducing energy consumption compared to continuous processing.
Solution Approach 2:
The system performs preliminary processing of congestion data by pre-segmenting the store into multiple regions and pre-establishing analysis models for each zone. When congestion data is collected, the AI processing focuses on analyzing changes in pre-defined regions rather than processing entire store images from scratch. This preliminary structuring of data and analysis frameworks improves detection accuracy while reducing the computational energy required for real-time processing.
3Adaptability or versatility
If personalized guidance is provided based on customer shopping status and preferences, then customer satisfaction is improved, but data processing complexity increases due to the need to analyze individual customer behavior patterns
Solution Approach 1:
The patent applies local quality by providing different levels of guidance personalization based on customer location and shopping status. Rather than applying complex personalized analysis uniformly to all customers, the system adjusts the depth and type of personalization according to local conditions - for example, providing more detailed personalized routing for customers in high-value product zones while using simpler guidance in low-traffic areas. This approach improves customer satisfaction where it matters most while reducing overall data processing complexity.
Data Source
AI summary
According to one embodiment, a customer guidance system includes a camera, a checkout apparatus, and a store server. The store server has a storage unit, a communication unit, and a control unit configured to identify and store a degree of congestion for different regions of the store, determine a current status of a customer in the store, select a region of the store as a suggested destination for the customer based on congestion degrees and the customer's current status. The control unit generates a prompt including the suggested destination and inputs the prompt to a generative AI trained to generate a customer guidance response for guiding the customer to the suggested destination. The customer guidance response is then supplied to the customer.