Machine Usage Severity Scoring for Predictive Maintenance Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine maintenance systems fail to account for the varying usage severity of machines, leading to inadequate timing for servicing and system overhauls, as they rely on age and utilization rate without considering the specific tasks and conditions each machine endures.
Innovation Solution
A method and system that calculates a machine usage severity score using predictive features like average fuel rate, ground speed, and gear shift rate, aggregated through models like Euclidean distance, to identify machines needing maintenance, incorporating machine learning for pattern recognition and sensor data analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional maintenance systems use age and utilization rate to determine maintenance timing, then the system is simple to implement, but it fails to account for varying usage severity and application types, leading to inadequate maintenance timing
Solution Approach 1:
The patent transforms the maintenance determination parameters from simple age and utilization rate to a comprehensive severity score based on multiple telematics parameters including fuel rate, ground speed, and operational intensity. This parameter transformation enables more precise maintenance timing assessment while systematically managing the complexity through structured data aggregation
Solution Approach 2:
The patent introduces a severity score as an intermediary metric that synthesizes multiple complex telematics data points into a single comparable value. This intermediary enables precise maintenance timing determination without requiring direct analysis of all individual parameters, thus balancing accuracy with system manageability
2Reliability
If condition based maintenance is used without considering application type, then the maintenance schedule is easier to generate, but it does not accurately reflect actual wear and tear from different usage severities
Solution Approach 1:
The patent applies local quality by tailoring the severity assessment to specific machine applications and usage patterns. Different telematics parameters are weighted and aggregated according to their relevance to specific application types, enabling maintenance schedules that reflect actual local usage conditions rather than generic benchmarks
Solution Approach 2:
The patent performs preliminary aggregation and analysis of telematics data to establish baseline severity patterns for different applications before generating maintenance schedules. This preliminary processing creates reference frameworks that simplify subsequent maintenance determinations while capturing the complexity of various usage scenarios
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method (300) for determining machine usage severity including collecting (302, 304) repair data (104) for multiple machines (20) over a repair time period and telematics data (102) from sensors (22, 24, 26) on the machines (20) over an activity time period. The method (300) can include calculating (306) predictive features from the telematics data (102) for each of the machines (20) and creating (308) a severity model based on the predictive features. The severity model can be validated (310) with the predictive features and corresponding repair data (104) for each of the machines (20). The method (300) can include receiving (312) telematics data (102) from sensors (22, 24, 26) on a deployed machine (20(1)) for a deployed period of time, calculating (314) a plurality of machine predictive features from the telematics data (102), and feeding (316) the machine predictive features into the severity model to calculate a severity score (318) for the deployed machine (20(1)). The method (300) can include displaying (320) a recommendation to perform maintenance on the deployed machine (20(1)) when the machine usage severity score exceeds a selected threshold.