Estimation Model Correction via Conversion Table for Production Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for estimating production progress in a production site are prone to errors due to the selection of unreliable statistic values, leading to inaccurate results when using estimation models.
Innovation Solution
An information processing apparatus is designed with an obtaining unit to use a learned model to estimate work processes in unit time, an evaluation unit to assess the accuracy of these estimates, a determination unit to create a conversion table for correcting low-accuracy processes, and a correction unit to refine the estimation results using this table.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If an estimation value is selected from a plurality of statistic values, then the estimation process can be simplified, but the result information includes errors and reduces reliability
Solution Approach 1:
The patent applies feedback by using actual work results to correct estimation values. The correction value is calculated based on the difference between actual work results and estimation values, then fed back to adjust future estimations. This continuous feedback loop improves reliability while maintaining simplified estimation processes.
Solution Approach 2:
The patent changes the parameter approach by transitioning from selecting static statistic values to dynamically adjusting estimation values using correction factors. The correction value is derived from actual work results and applied to modify estimation parameters, enabling the system to adapt to real conditions while keeping the estimation process simple.
2Productivity
If statistic values are calculated from work time data, then production progress can be estimated, but the estimation accuracy decreases when selecting from multiple statistic values
Solution Approach 1:
The system uses actual work results as feedback to correct estimation values. By calculating correction values from the difference between actual and estimated work results, the system continuously refines estimation accuracy while maintaining productive estimation processes.
Solution Approach 2:
The patent applies preliminary action by pre-calculating correction values based on historical data and using them to adjust future estimations. This preliminary correction setup ensures that estimation accuracy is improved before new production progress needs to be measured.
3Measurement precision
If a conversion table is created to correct estimation results, then estimation accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent extracts the correction function into a separate conversion table that maps estimation values to corrected values. This extraction isolates the complexity of correction logic from the main estimation process, allowing the core estimation functionality to remain simple while improving accuracy through the lookup table.
Solution Approach 2:
The system creates a conversion table that copies and stores correction relationships from historical data. Instead of complex real-time calculations, the system uses pre-computed correction mappings stored in the table, simplifying the correction process while maintaining high accuracy.
Data Source
AI summary
An information processing apparatus comprises: an obtaining unit configured to obtain an estimation result estimated using an estimation model; a determination unit configured to determine a conversion table configured to correct the estimation result using the estimation model based on an estimation accuracy in the estimation model; and a correction unit configured to correct the estimation result based on the conversion table.