Proposal System Using Learned Model for Dynamic Device Recommendations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing proposal systems for purchasing new devices, such as printers, fail to account for changes in usage situations, leading to inappropriate product recommendations based solely on past usage patterns.
Innovation Solution
A proposal system that includes a first acquisition unit for sales performance information, an acceptance unit for usage situation information, and an inference unit using a learned model to infer suitable purchase targets, allowing for dynamic adjustments based on changing usage scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a new product is presented based only on the usage information of a customer, then the proposal system can operate with simple processing, but it cannot propose appropriate information when the customer's usage situation changes
Solution Approach 1:
The system pre-acquires and stores sales performance information for multiple devices before a purchase proposal is needed. This preliminary data collection enables the system to quickly compare current usage situations against historical sales data without complex real-time processing, while still providing accurate proposals that adapt to changing usage situations.
Solution Approach 2:
The patent introduces sales performance information as an intermediary element between usage information and purchase proposals. This intermediary data layer allows the system to bridge the gap between simple usage tracking and complex purchase recommendations, enabling accurate proposals without requiring complex direct processing between usage patterns and product recommendations.
2Device complexity
If the system uses only past usage information to propose products, then the system structure remains simple, but it fails to account for changes in usage situations
Solution Approach 1:
The system processes multiple types of information (usage information and sales performance information) through a unified proposal generation mechanism. This multi-functional approach allows the same system structure to handle both simple usage-based proposals and more complex scenario-based proposals, maintaining structural simplicity while improving reliability through diverse data inputs.
Solution Approach 2:
By pre-acquiring sales performance information and storing it in advance, the system prepares multiple potential proposal options before a purchase decision is needed. This preliminary preparation ensures that when usage situations change, the system can immediately provide reliable proposals without requiring complex real-time analysis, thus maintaining simple system structure while improving proposal appropriateness.
3Measurement precision
If the system acquires and processes multiple types of information (sales performance and usage situation), then proposal accuracy improves, but information processing complexity increases
Solution Approach 1:
The system performs the complex task of acquiring and organizing sales performance information in advance, storing it in a ready-to-use format. This preliminary processing eliminates the need for complex real-time information gathering and analysis, allowing the system to achieve high proposal accuracy by simply comparing current usage situations against pre-processed sales data.
Solution Approach 2:
The system automatically manages the acquisition, storage, and organization of sales performance information without requiring complex external intervention during the proposal generation process. This self-service approach to data preparation reduces processing complexity at the time of proposal generation, as the system can directly utilize the pre-organized information it has independently prepared.
Data Source
AI summary
In the proposal system, sales performance information that is related to a device is acquired and information that is related to a usage situation of a device is accepted from a customer. A device to be a purchase target is inferred by a learned model based on the information that is related to the usage situation of the device and the sales performance information, and then a proposal information is outputted as a result of that inference.