Prediction Model for Detecting Abnormal Commodity Handling Histories
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traceability systems fail to detect falsifications in the handling history of articles within a business operator, particularly in the alteration of product values such as weight and qualitative data during processing treatments.
Innovation Solution
An information processing device and method that utilizes a prediction model trained on production and handling information to calculate a target product prediction amount, comparing it with actual amounts to detect abnormalities in the handling history.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a traceability system records handling history of articles, then the handling history can be tracked and displayed, but the system cannot detect falsifications in the handling history within a business operator
Solution Approach 1:
The system performs preliminary actions by training a prediction model on historical handling data before actual falsification detection occurs. The model learns normal handling patterns in advance, enabling it to automatically identify deviations without adding complex real-time monitoring infrastructure.
Solution Approach 2:
A prediction model serves as an intermediary between the recorded handling history and the detection of falsifications. The model acts as a mediator that compares actual handling data against predicted normal patterns, identifying abnormalities without requiring direct complex analysis of all handling records.
2Measurement precision
If the traceability system monitors all handling processes, then falsification can be detected, but the measurement precision and detection capability are insufficient for subtle falsifications
Solution Approach 1:
The system implements feedback by continuously comparing actual handling amounts against prediction model outputs. When deviations are detected, the system can flag potential falsifications and use this information to refine future detection, improving precision over time without increasing measurement complexity.
Solution Approach 2:
The prediction model analyzes multiple parameters simultaneously (handling amounts, product types, time intervals, business operator patterns) rather than relying on single threshold values. This multi-parameter approach enables detection of subtle falsifications that would be invisible to simple monitoring systems.
3Measurement precision
If a prediction model is trained on handling history data, then abnormality detection accuracy is improved, but data processing time and computational resources increase
Solution Approach 1:
The prediction model is trained in advance on historical handling data, performing the computationally intensive learning process before actual detection is needed. This preliminary training allows rapid real-time detection without processing delays, as the model has already learned normal patterns during the off-line training phase.
Solution Approach 2:
The system focuses on predicting and detecting only the most critical abnormalities in handling amounts rather than analyzing every possible aspect of handling history. This partial action approach maintains high detection accuracy for falsifications while reducing overall computational burden and processing time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The purpose of the present disclosure is to provide an information processing device with which it is possible to appropriately detect abnormality included in commodity handling histories. An information processing device according to the present disclosure comprises: an acquisition unit that acquires the history of handling of a commodity of interest that indicates the history of handling of the commodity of interest that is to be monitored; a prediction unit that predicts a quantity after handling in relation to the handling included in the history of handling of the commodity of interest and calculates a predicted quantity of the commodity of interest on the basis of the history of handling of the commodity of interest, by using a prediction model having been trained using production information regarding production of a commodity, handing content information that indicates the handling content of the commodity, and quantity information that indicates quantities before and after handling of the commodity, as training data; and an abnormality determination unit that determines, on the basis of the difference between the quantity of the commodity of interest after handling in the history of handling of the commodity of interest and the predicted quantity of the commodity of interest calculated by the prediction unit, whether or not there is abnormality in the quantity change indicated by the history of handling of the commodity of interest.