Maintenance Data Prioritization for Fault Detection Under Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing maintenance management systems face challenges in managing multiple business operation apparatuses that transmit various types of data, leading to potential delays in fault detection and system overload due to uneven data processing loads, as they struggle to prioritize and adjust the frequency of data acquisition based on data importance and cycle requirements.
Innovation Solution
A maintenance management system that includes a data processing unit and a load level management unit, which adjusts the number and frequency of data items processed based on priority, importance, and cycle requirements, dynamically optimizing data processing to prevent delays and overload by prioritizing critical data and extending acquisition cycles for less critical items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data transmission is repeated by each business operation apparatus at short time cycles, then real-time monitoring capability is improved, but the management system becomes overloaded and processing ability is exceeded
Solution Approach 1:
The patent implements dynamic adjustment of data acquisition cycles based on apparatus load conditions. The management system monitors its own processing load and dynamically modifies the data acquisition frequency from multiple business operation apparatuses, increasing frequency when load is low and decreasing when load is high, thereby maintaining real-time monitoring capability while preventing system overload
Solution Approach 2:
The system changes the time cycle parameter of data acquisition based on current system conditions. By adjusting the acquisition cycle parameter dynamically rather than using a fixed interval, the system optimizes the balance between real-time monitoring performance and processing capacity utilization
2Productivity
If data is accumulated for a specific period on the side of monitoring targets, then the load of the management system is reduced, but the timing of data transmission is delayed and real-time properties are compromised
Solution Approach 1:
The patent applies dynamic control where the data accumulation period is not fixed but adjusts based on system load conditions. When the management system has sufficient processing capacity, the accumulation period is shortened to maintain real-time properties. When load is high, the accumulation period is extended to reduce transmission frequency and management system load
3Adaptability or versatility
If multiple types of data are transmitted from each business operation apparatus, then comprehensive monitoring coverage is improved, but the amount of data and processing load significantly increase
Solution Approach 1:
The patent applies local quality by treating different data types differently based on their characteristics and importance. Each data type from business operation apparatuses is assigned different acquisition cycles and priority levels according to its specific requirements, rather than applying a uniform acquisition strategy to all data types
Solution Approach 2:
The system segments the data acquisition process by data type and apparatus, allowing independent control of acquisition parameters for different data streams. This segmentation enables selective optimization where critical data is acquired frequently while less critical data is acquired less frequently, reducing overall processing load while maintaining comprehensive monitoring
Data Source
AI summary
To prevent a delay in detection of a fault and the like and to cut down a load on a management side while making transmission of data appropriate in consideration of importance and properties of each type of the data in a case in which a large amount of data is transmitted from management target apparatuses. Items with high priority are processed first, and items with a low priority are allowed to be thinned out in accordance with priority and a weight that a management system 10 defines for each data item, thereby reducing a load. The thinning out is performed by limiting the number of data items to be processed once in accordance with a load level. Data acquisition intervals and priority for each item are dynamically changed by reflecting the importance of the data of each item and a change thereof. Trends of data and the like distributed using telemetry are observed and are fed back to load control through control using an AI or a rule base. In a case in which distribution intervals of each data item are changed, the distribution intervals are changed to a multiple of a basic cycle to curb influences on correlations among the items.