Server-Based Diagnostic Model for Information Processing Apparatus
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing diagnostic systems for image forming apparatuses, such as multifunction peripherals, face challenges in accurately identifying malfunctions due to increased complexity and the lack of consideration for dynamic information like changes in settings and sensor values over time, which also fail to predict potential malfunctions before they occur.
Innovation Solution
A system where a server apparatus collects time series information and error data from multiple information processing apparatuses, generates a diagnostic model through learning, and sends it to the apparatuses for self-diagnosis, enabling the prediction of potential errors by analyzing collected data using the model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a diagnostic model is generated using only error information and static operating mode data, then the system complexity remains low, but the diagnosis accuracy is insufficient and cannot capture dynamic changes over time
Solution Approach 1:
The system performs preliminary data collection and model training before actual diagnosis is needed. Time series data from multiple apparatuses is collected and used to train the diagnostic model in advance, so that when diagnosis is performed, the model is already prepared to accurately identify malfunction causes based on historical patterns
Solution Approach 2:
A server apparatus is introduced as an intermediary between multiple information processing apparatuses and the diagnostic model. The server collects time series data from multiple apparatuses, trains the diagnostic model, and then provides it for use, distributing the computational complexity and data management burden away from individual apparatuses
2Measurement precision
If time series information and data from multiple apparatuses are collected and used for model training, then the diagnosis accuracy improves, but the amount of data to be processed and stored increases significantly
Solution Approach 1:
Data from multiple information processing apparatuses is merged and combined into a unified dataset for model training. By combining time series information from multiple apparatuses, the system creates a more comprehensive training dataset that improves model accuracy while sharing the data collection burden across the network
3Reliability
If dynamic information such as changes in setting values and sensor values over time is incorporated into the diagnostic model, then the ability to detect potential malfunctions before they occur improves, but the complexity of data collection and processing increases
Solution Approach 1:
The information processing apparatus automatically collects its own time series data including setting values and sensor readings, and performs self-diagnosis using the trained diagnostic model. The system serves itself by gathering necessary data and applying the model without requiring external intervention for each diagnosis instance
Solution Approach 2:
The diagnostic model analyzes time series data to provide feedback about the likelihood of future malfunctions. By continuously monitoring changes in setting values and sensor readings over time, the system generates feedback that indicates potential issues before they manifest as actual errors, enabling preventive maintenance
Data Source
AI summary
A server apparatus collects, from a plurality of information processing apparatuses, time series information indicating a change over time in settings and a state of each information processing apparatus, and error information indicating an error that has occurred in each information processing apparatus, generates learning data including the collected information, and generates a diagnostic model, which is a trained model for a malfunction diagnosis in the information processing apparatus, through learning based on the generated learning data. The information processing apparatus obtains the diagnostic model from the server apparatus, outputs a diagnosis result indicating a likelihood that an error will occur in the information processing apparatus by collecting information necessary for the malfunction diagnosis within the information processing apparatus and executing the malfunction diagnosis using the collected information and the obtained diagnostic model.