Equipment Performance Classification for Condition-Based Maintenance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current equipment monitoring practices in manufacturing, such as in pharmaceutical production, rely heavily on human expertise, leading to inconsistent and time-consuming identification of performance issues, often resulting in unnecessary resource expenditure or unacceptably high frequencies of equipment failures due to inadequate maintenance scheduling.
Innovation Solution
An automated system that uses a classification model trained with historical sensor data to diagnose equipment performance issues and determine appropriate actions, reducing the need for continuous human monitoring and optimizing maintenance activities based on real-time data analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual monitoring by subject matter experts is used, then equipment performance issues can be identified using personal knowledge and experience, but the process is time-consuming and expertise is inconsistently applied across locations and over time
Solution Approach 1:
The patent replaces the mechanical system of manual expert monitoring with an automated computational system. Machine learning models analyze sensor data to identify equipment performance issues, substituting human experts' manual analysis with automated algorithms that process data consistently and rapidly without requiring human time investment or subjective expertise application.
Solution Approach 2:
The system enables equipment to monitor and diagnose its own performance issues autonomously. Through embedded sensors and machine learning models, the equipment self-identifies performance deficiencies without requiring external expert intervention, thereby eliminating the time loss associated with manual monitoring while maintaining high identification accuracy through automated diagnostic capabilities.
2Reliability
If manual monitoring by subject matter experts is used, then equipment problems can be diagnosed and appropriate actions determined, but costs are high due to the number of man-hours required from highly skilled individuals
Solution Approach 1:
The patent replaces the expensive mechanical resource of highly skilled human monitors with an automated computational system. Machine learning models provide reliable problem identification at a fraction of the labor cost, as the system processes sensor data and diagnoses equipment issues without requiring human expertise, thereby maintaining reliability while dramatically reducing the energy loss associated with human labor costs.
3Reliability
If equipment is maintained on a regular calendar basis or usage basis, then the likelihood of problems is lowered, but this results in an unnecessarily high expenditure of resources if maintenance is performed more often than needed or an unacceptably high number or frequency of performance issues if maintenance is performed less often than needed
Solution Approach 1:
The patent transitions from static, predetermined maintenance schedules to dynamic, condition-based maintenance. Machine learning models continuously analyze sensor data to assess actual equipment condition, enabling maintenance activities to be scheduled dynamically based on real-time equipment state rather than fixed calendar or usage intervals, thereby optimizing both reliability and resource efficiency.
Solution Approach 2:
The system performs preliminary diagnosis of equipment performance issues by analyzing sensor data trends before actual failures occur. Machine learning models detect early signs of equipment deterioration and predict potential problems, enabling maintenance to be performed proactively at the optimal moment - neither too early (wasting resources) nor too late (causing failures) - thereby improving both equipment reliability and resource efficiency.
4Loss of information
If subject matter experts monitor equipment, then personal knowledge can be applied to theorize root causes, but expertise is specific to each individual or team and therefore can be inconsistently applied across locations and over time
Solution Approach 1:
The patent replaces the variable human expert knowledge system with a standardized computational system. Machine learning models encode diagnostic expertise into algorithms that apply consistent analysis rules across all locations and time periods, eliminating the inconsistency inherent in individual expert judgment while preserving and standardizing the value of expert knowledge through reproducible computational methods.
Data Source
AI summary
A method of diagnosing or predicting performance of equipment includes determining values of one or more parameters associated with the equipment by monitoring the one or more parameters over a time period in which the equipment is in use. The method also includes determining, by processing the values of the one or more parameters using a classification model, a performance classification of the equipment, mapping the performance classification to a mitigating or preventative action, and generating an output indicative of the mitigating or preventative action.