Maintenance Parts Prediction for Fewer Return Trips
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing maintenance systems face inefficiencies due to unpredictable component needs and limited carrying capacity, leading to increased travel and resource management challenges for maintenance personnel.
Innovation Solution
An information processing system predicts component requirements for multiple devices and notifies maintenance personnel of necessary components and quantities, allowing for optimized component delivery based on predicted needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a person in charge brings components from a storage place for maintenance work, then the necessary components can be obtained, but the person in charge must return to the storage place multiple times when components are insufficient or surplus components are carried
Solution Approach 1:
The system predicts component requirements in advance before maintenance work begins. By predicting which components will be needed and in what quantities, the system allows maintenance personnel to prepare appropriate component lists beforehand, reducing the need for multiple trips to the storage place during actual maintenance work.
Solution Approach 2:
The system collects actual maintenance data and component usage information, then uses this feedback to improve future predictions. By analyzing historical maintenance records and component consumption patterns, the system continuously refines its prediction accuracy, enabling better advance preparation and reducing unnecessary travel.
2Loss of time
If a person in charge brings surplus components based on prediction, then the number of returns to storage place can be reduced, but the carrying capacity of the person in charge is limited
Solution Approach 1:
The system predicts not only the necessary components but also anticipates potential additional needs by predicting component requirements for multiple possible maintenance scenarios. This allows maintenance personnel to carry slightly more components than the absolute minimum required, ensuring that unexpected additional needs can be met without returning to the storage place.
Solution Approach 2:
The system dynamically adjusts prediction parameters such as prediction range, confidence levels, and component quantities based on various factors including maintenance history, device complexity, and environmental conditions. This enables flexible optimization of component carrying quantities to match actual needs while respecting carrying capacity constraints.
3Productivity
If component prediction is implemented for multiple devices, then overall maintenance efficiency can be improved, but the complexity of predicting components for multiple devices increases
Solution Approach 1:
The system uses a unified prediction model that can handle multiple devices simultaneously by identifying common component patterns and relationships. Rather than creating separate prediction systems for each device, the system analyzes data across multiple devices to find universal patterns, reducing overall system complexity while improving predictive accuracy through broader data analysis.
Solution Approach 2:
The system divides the prediction task into manageable segments by analyzing devices in groups or categories based on similarity. This segmentation allows the complex prediction problem to be broken down into smaller, more tractable sub-problems that can be solved more efficiently while still providing comprehensive predictions for all devices.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing system includes one or more processors configured to: predict component names having a possibility of being used for maintenance of plural devices determined in advance and quantities of the component names; and notify, in a case where a maintenance request is received for a device among the plural devices, a person in charge, who is in charge of maintenance work for the device and who is determined from among plural workers, of information indicating a component name and a quantity of the component name for the maintenance work and information indicating a component name and a quantity of the component name based on a result of the prediction, as details of components to be brought.