Maintenance Threshold Classification for Failure Probability Decisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing maintenance systems for large numbers of equipment face high costs and potential safety risks due to limited equipment maintenance and the transition from time-based to condition-based maintenance, which can compromise safe operation.
Innovation Solution
A computer system that classifies equipment into states requiring maintenance, candidate states for maintenance determination, or non-maintenance states using threshold values and failure probabilities, considering operation risk and cost, utilizing Weibull distribution and Sigmoid functions for predictive maintenance modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If condition-based maintenance is implemented to reduce maintenance costs, then maintenance cost decreases, but safety risk increases due to limited equipment maintenance
Solution Approach 1:
The patent changes the parameter of maintenance decision-making from binary (maintenance/not maintenance) to a continuous spectrum using three threshold values (first threshold, second threshold, third threshold) that divide equipment states into multiple regions. This allows for nuanced maintenance decisions based on precise failure probability assessments, reducing costs while maintaining safety by targeting maintenance only when truly necessary.
Solution Approach 2:
The patent replaces traditional mechanical/time-based maintenance systems with an information-based decision support system that uses failure probability calculations, threshold value comparisons, and automated classification to determine maintenance needs. This substitution enables more precise and cost-effective maintenance planning while ensuring safety through systematic risk assessment.
2Reliability
If maintenance is performed on all equipment to ensure safe operation, then safety risk decreases, but maintenance cost increases significantly
Solution Approach 1:
The patent applies local quality by treating different equipment items differently based on their individual failure probabilities and operational characteristics. Instead of uniform maintenance across all equipment, the system classifies each equipment item into specific regions (first region, second region, third region) using threshold comparisons, allowing maintenance resources to be concentrated where they are most needed while reducing unnecessary maintenance elsewhere.
Solution Approach 2:
The patent implements partial action by performing maintenance only on equipment that falls into the first region (high failure probability), rather than maintaining all equipment. The threshold-based classification system identifies and targets only the critical subset of equipment requiring maintenance, avoiding excessive maintenance actions on low-risk equipment while ensuring safety for high-risk items.
3Reliability
If threshold values are set to be conservative to ensure safety, then safe operation is maintained, but maintenance cost increases
Solution Approach 1:
The patent introduces dynamics by establishing multiple threshold values (first, second, third thresholds) that create a dynamic classification system with multiple regions. This allows the maintenance decision boundary to be flexible and adaptive rather than fixed, enabling the system to balance safety and cost by adjusting which equipment falls into which region based on their specific failure probabilities and operational contexts.
Data Source
AI summary
A computer system is coupled to a facility having at least one piece of equipment, and holds threshold value information for managing threshold values for classifying the piece of equipment, based on a failure probability of the piece of equipment, into any one of a first state indicating a state in which maintenance is required, a second state indicating a state in which a determination as to whether maintenance is to be executed is required, or a third state in which the maintenance is not required. The computer system is configured to determine the maintenance threshold value and the failure probability of the at least one piece of equipment, and classify the piece of equipment into one of the states based on the failure probability of and the threshold value information on the piece of equipment, and outputting determination assistance information including a result of the classification.