Image Form Apparatus Management Device Abnormal Occurrence Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing management systems for image forming apparatuses face limitations in predicting abnormal occurrences due to high calculation workloads, increased network workload, and system configuration costs, leading to potential delays in image processing and mechanical control.
Innovation Solution
A management device that collects status data from image forming apparatuses, processes it using a first stage determination unit with simple data processing, and employs a second stage determination unit with weighted logic to predict abnormal occurrences, thereby reducing calculation workload and improving prediction reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If neural networks or Bayesian inference are used for abnormal occurrence determination, then prediction accuracy is improved, but calculation workload increases causing delays in image processing and mechanical control
Solution Approach 1:
The patent segments the abnormal occurrence determination process into multiple stages: a first stage using simple threshold comparisons for quick filtering, and a second stage using more complex neural networks or Bayesian inference only for cases that require deeper analysis. This segmentation allows the system to maintain high prediction accuracy while minimizing calculation workload and avoiding delays in image processing and mechanical control.
2Reliability
If status information is collected and sent to a server via network, then abnormal occurrence prediction capability is improved, but network workload and system configuration cost increase
Solution Approach 1:
The patent implements self-service by enabling the image forming apparatus to perform abnormal occurrence determination locally using embedded neural networks or Bayesian inference algorithms. This eliminates the need to send status information to external servers, reducing network workload and system configuration costs while maintaining prediction capability. The apparatus serves itself by having the determination function integrated within the device rather than relying on external infrastructure.
3Productivity
If limited status data (only anomaly count) is used, then calculation workload is reduced, but types of predictible abnormal occurrences are limited
Solution Approach 1:
The patent segments the data processing approach by using simple threshold-based filtering for common abnormal occurrences with limited data, while reserving complex analysis methods for specific cases where multiple status parameters are available. This allows the system to maintain high calculation efficiency for frequent issues while expanding prediction coverage for less common abnormalities when sufficient data exists.
Solution Approach 2:
The patent applies partial action by using simplified determination methods for the majority of cases where basic anomaly count data suffices, and only applying comprehensive multi-parameter analysis when necessary. This partial use of complex methods maintains calculation efficiency while expanding the range of predictible abnormal occurrences when data availability warrants it.
Data Source
AI summary
A management device for an image forming apparatus including a status data collection unit where multiple types of status data are received from the image forming apparatus and stored in a status database, a target data creation unit where multiple types of target data are created based upon the multiple types of status data, a first stage determination unit where the multiple types of target data are identified as being above or below reference values set for each type, and a second stage determination unit where a weight value set for each status data type is attached to the determination results of the multiple types of status data of the first stage determination unit and as a whole of the multiple types of status data determined with majority logic for abnormal occurrence prediction.