Sales Volume Prediction Using Store Visits and Dynamic Menus
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sales volume prediction technologies struggle to accurately account for daily changing categories of products, such as dishes in a restaurant, leading to food loss due to discrepancies between predicted and actual sales volumes.
Innovation Solution
A prediction device and method that utilizes trained models to predict store visits and product popularity, considering menu combinations and external factors, allowing for precise sales volume forecasting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional sales volume prediction technology is used, then prediction can be performed using external data, but prediction accuracy deteriorates when daily changing categories exist
Solution Approach 1:
The prediction system is segmented into multiple specialized models: a store visit number prediction model that handles foot traffic patterns, a sales rate prediction model that handles product-specific sales behavior, and a menu management component that tracks daily changing categories. This segmentation allows each model to specialize in specific aspects, improving overall prediction accuracy for dynamic menus
Solution Approach 2:
The system dynamically adapts to changing menu categories by using the sales rate prediction model that can handle product popularity variations. The menu feature data and popular product information are continuously updated to reflect current daily categories, allowing the prediction system to remain accurate despite menu changes
2Loss of substance
If prediction does not consider daily changing categories, then prediction process is simpler, but food loss increases due to prediction discrepancies
Solution Approach 1:
The system performs preliminary prediction of sales volume by combining store visit number predictions with sales rate predictions before actual sales occur. This advance prediction allows the restaurant to prepare appropriate ingredient quantities, reducing food loss from over-preparation while the complexity is managed through automated modeling
Solution Approach 2:
The system uses feedback from actual sales data to continuously improve prediction accuracy. By comparing predicted sales volume with actual sales and updating the sales rate prediction model with real performance data, the system reduces food loss over time while maintaining manageable complexity through iterative learning
3Adaptability or versatility
If the number of menu kinds and combinations is enormous, then product variety is higher, but prediction difficulty increases
Solution Approach 1:
The system extracts and separates the prediction task into distinct components: store visit number prediction handles the customer flow aspect, while sales rate prediction handles the product selection aspect. This extraction simplifies the overall prediction difficulty by breaking down the complex menu combination problem into manageable separate models
Solution Approach 2:
The sales rate prediction model serves multiple functions: it predicts popularity for individual products, accounts for menu feature effects, and handles daily changing categories universally. This multi-functionality allows the system to manage enormous menu variety through a single versatile model rather than requiring separate models for each menu combination
Data Source
AI summary
A prediction device includes: a store visit number prediction unit that acquires data regarding the number of past visits to a store, inputs the data to a trained store visit number prediction model, and predicts the number of visits to a store on a prediction target day by using an output from the store visit number prediction model, a rate prediction unit that acquires data regarding the number of past visits to a store, a past sales volume, a menu feature, and a popular product to predict popularity of each product on the prediction target day, inputs the popularity to a trained sales rate prediction model, and predicts a sales rate of each product by using an output from the sales rate prediction model; and a sales volume prediction unit that predicts a sales volume of each product on the prediction target day by using the number of visits to a store predicted by the store visit number prediction unit and the sales rate of each product predicted by the rate prediction unit.