Demand Prediction Error Interpretation Using Intra- and Extra-Company Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing demand prediction technologies fail to accurately analyze errors in demand prediction for specific product segments, lacking comprehensive consideration of intra-company and extra-company information.
Innovation Solution
An information processing system and method that utilizes a large language model to generate interpretation examples based on intra-company and extra-company information, including factors contributing to demand prediction errors, thereby enhancing the analysis of demand prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If demand prediction is performed using only past operation data and delivery results, then the prediction process is simple, but the analysis of prediction errors lacks comprehensiveness
Solution Approach 1:
The patent segments the information processing into distinct modules: index acquisition, intra-company information acquisition, extra-company information acquisition, and interpretation example generation. This segmentation allows comprehensive error analysis while maintaining manageable system complexity through modular architecture.
Solution Approach 2:
The patent adds new dimensions to demand prediction by incorporating both intra-company information (internal operational data) and extra-company information (external market data). This dimensional expansion enables comprehensive error analysis without overwhelming complexity, as each dimension can be processed independently through the structured framework.
2Loss of information
If comprehensive intra-company and extra-company information is collected for error analysis, then the error analysis becomes comprehensive, but the information processing time increases
Solution Approach 1:
The patent performs preliminary actions by pre-acquiring and structuring both intra-company and extra-company information before error analysis is needed. The index acquisition and information collection occur in advance, allowing rapid error analysis when required without time-consuming on-the-fly data gathering.
Solution Approach 2:
The patent introduces an intermediary processing layer that systematically organizes and relates intra-company and extra-company information to prediction errors. This intermediary structure enables comprehensive information integration while reducing processing time through pre-established relationships and structured data formats.
3Loss of information
If detailed interpretation examples are generated using large language models, then the understanding of prediction errors improves, but the computational resources required increase
Solution Approach 1:
The patent applies local quality by generating interpretation examples selectively and locally rather than globally. The large language model processes information focused on specific error cases and their relevant contextual factors, optimizing computational resources by avoiding unnecessary processing of unrelated data while maintaining deep error analysis where needed.
Data Source
AI summary
An information processing system includes an index acquisition unit that acquires an index for managing accuracy of demand prediction related to an object segment including a plurality of products handled by an object company, an intra-company information acquisition unit that acquires intra-company information related to the object segment inside the object company, an extra-company information acquisition unit that acquires extra-company information related to the object segment outside the object company, and an interpretation example generation unit that generates, as an interpretation example of the index, a sentence including a factor based on the intra-company information and the extra-company information related to an error of the demand prediction, by using a large language model.