Purchase-Purpose Customer Segmentation for Retail Demand Forecasting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Small-scale retail stores face significant waste loss due to difficulty in predicting demand for each product accurately, as they have smaller sales volumes and complex customer purchase patterns that existing methods struggle to capture.
Innovation Solution
A product group and customer group extraction device that estimates purchase purposes based on customer history information, generates learning data, and clusters this data to extract product and customer groups, enabling more accurate demand prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If demand prediction is performed for each product in small-scale stores, then prediction accuracy is required to be high, but prediction becomes more difficult due to smaller sales volumes
Solution Approach 1:
The patent segments customers into distinct groups based on their purchase behavior patterns, product preferences, and shopping characteristics. By dividing the customer base into segments with similar behaviors, the system can make more accurate demand predictions for each segment, which then aggregate to improve overall prediction accuracy despite limited total sales volume in small-scale stores
Solution Approach 2:
The patent introduces a new dimension of analysis by clustering customers based on multiple behavioral attributes simultaneously (purchase frequency, product categories, spending patterns, time of purchase). This multi-dimensional customer segmentation transforms the prediction problem from individual product forecasting to group-based demand patterns, making predictions more reliable even with smaller sales volumes
2Loss of substance
If order quantity is set to be small with respect to actual demand for fear of leftover, then waste loss is reduced, but products may run out leading to loss of sales opportunities
Solution Approach 1:
The system continuously monitors actual purchase behavior, compares it with predicted demand, and uses this feedback to refine customer segment profiles and improve future predictions. This closed-loop feedback mechanism allows the system to learn from past performance and adjust order quantities dynamically, reducing both waste and stockouts over time
Solution Approach 2:
The patent performs demand prediction and customer segmentation in advance before the sales period begins. By identifying high-demand product groups and customer segments beforehand, the system enables proactive ordering decisions that ensure product availability while minimizing excessive inventory, thus preventing both waste and stockouts
Data Source
AI summary
An estimation unit of a product group and customer group extraction device estimates a purchase purpose for which a customer has purchased a product, on the basis of history information including customer information for identifying the customer, product information on the product purchased by the customer, and date and time information of purchase of the product. A generation unit of the product group and customer group extraction device generates learning data from the history information added with the purchase purpose. An extraction unit of the product group and customer group extraction device clusters the learning data for each customer on the basis of the purchase purpose to extract product groups, and extracts the customer group on the basis of similarity of the product groups extracted for each of a plurality of the customers.