system
The system addresses inefficient inventory management by using historical and environmental data with machine learning to forecast demand, optimizing order quantities and timings, and improving forecasting accuracy through a feedback loop.
JP2026096672APending Publication Date: 2026-06-15SOFTBANK GROUP CORP
View PDF 1 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- SOFTBANK GROUP CORP
- Filing Date
- 2024-12-03
- Publication Date
- 2026-06-15
AI Technical Summary
Technical Problem
Inventory management in retail and food industries is inefficient, leading to opportunity losses and excess inventory due to inaccurate demand forecasting, which is costly and time-consuming for small and medium-sized operators.
Method used
A system that collects and preprocesses historical sales data, weather information, and promotional event data, using machine learning to forecast demand and determine optimal order quantities and timings, with a feedback loop for model improvement.
Benefits of technology
Enables efficient inventory management by reducing ordering errors and costs, allowing for accurate demand forecasting and quick response to environmental and promotional changes.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure 2026096672000001_ABST
Abstract
We provide the system. [Solution] A means for collecting past sales data, weather information, seasonal information, and promotional event information, and for pre-processing this data, A means of generating forecast data by using a machine learning model to perform demand forecasting based on preprocessed data, A means for determining the optimal order quantity and timing, taking into account the characteristics of each store and product, based on demand forecasts. A means to automatically execute the generated order proposal, A means to evaluate the difference between actual sales data and forecast data, and to improve the forecasting model, A system that includes this.
Need to check novelty before this filing date? Find Prior Art