Payment Device Demand Prediction Using Historical Store Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in efficiently predicting the demand for operating payment devices while maintaining a low arithmetic processing load, particularly in retail environments where congestion and staff allocation are critical.
Innovation Solution
An information processing apparatus that stores data on unpaid customers, operating devices, and waiting customers, uses sensor data to calculate current unpaid individuals, specifies similar past data to determine required devices, and adjusts staff accordingly, thereby predicting demand without excessive computational burden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If simulation or machine learning is used to predict congestion status and calculate recommended number of payment devices, then prediction accuracy is improved, but arithmetic processing load increases
Solution Approach 1:
The patent uses historical store data as a simplified copy of current conditions to predict future device requirements. Instead of performing complex simulations or machine learning, the system retrieves and compares historical data records that match current conditions, providing accurate predictions with minimal computational overhead.
Solution Approach 2:
The system pre-processes and stores historical store data in advance, organizing it by relevant conditions (number of unpaid people, operating devices, waiting people). When prediction is needed, the system simply queries this pre-organized data rather than performing real-time complex calculations, reducing arithmetic processing load while maintaining accuracy.
2Loss of time
If the number of payment devices is increased to reduce customer waiting time, then service quality is improved, but device cost and operational complexity increase
Solution Approach 1:
The system dynamically adjusts the recommended number of payment devices based on real-time store conditions by comparing current data with historical patterns. This allows the store to optimize device allocation flexibly according to actual demand, reducing waiting time during peak periods without permanently increasing device count or operational complexity.
Data Source
AI summary
An information processing apparatus includes a memory storing instructions, and storing store data including the number of unpaid customers who are unpaid in a store, the number of operating payment devices, and the number of waiting customers who are waiting for payment in a payment devices, at each of a plurality of time points, and one or more processors configured to execute the instructions to acquire the number of currently operating payment devices, calculate the number of currently unpaid customers based on the sensor data of the store, specify past store data in which the number of unpaid customers and the number of operating payment devices similar to the number of currently unpaid customers and the number of currently operating payment devices are associated with each other, and determine the number of required payment devices based on the specified store data.