Mine Productivity Monitoring Using Cycle-Based Factor Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mine management systems fail to accurately detect productivity deviations from planned levels due to insufficient consideration of factors like route length, slope, road surface conditions, and operator performance, especially in mines with multiple or unknown routes, making it difficult to improve productivity effectively.
Innovation Solution
A mine management system that includes a server device for accumulating and processing working data from mining machines, using machine learning to identify out-of-plan factors through cycle data analysis, and a determination device to determine productivity deviations, enabling real-time detection and visualization of productivity issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional productivity monitoring systems are used, then basic fuel consumption and load transportation rates can be detected, but the systems fail to accurately detect productivity deviations caused by multiple factors such as route characteristics, road surface conditions, and operator performance
Solution Approach 1:
The system segments productivity analysis into discrete work cycles, extracting cycle data including start/end times, routes, payloads, and operational parameters. This segmentation allows detailed analysis of multiple factors (route characteristics, road conditions, operator performance) without overwhelming system complexity, as each factor can be evaluated within its specific cycle context.
Solution Approach 2:
The system transforms raw working data into multiple derived parameters including cycle data, route characteristics, operational metrics, and environmental factors. By changing parameters from simple fuel consumption rates to comprehensive cycle-based metrics that incorporate route length, slope, road surface conditions, and operator behavior, the system achieves accurate multi-factor productivity deviation detection.
2Measurement precision
If the system considers multiple factors affecting productivity, then detection accuracy improves, but the system becomes difficult to apply to mines with multiple or unknown routes
Solution Approach 1:
The system achieves universality by designing a route-independent cycle data extraction framework that works across multiple route configurations. The cycle data structure and analysis methods are applicable to any mine layout whether routes are known or unknown, fixed or variable. The system processes working data through standardized cycles that can accommodate diverse operational scenarios including different numbers of routes, route changes, and unknown terrain.
Solution Approach 2:
The system dynamically adapts to different mine configurations by extracting cycle data based on actual operational patterns rather than predefined route structures. The cycle extraction automatically adjusts to varying route numbers, lengths, and characteristics, allowing the same system to accurately detect productivity deviations across static and dynamic route configurations without requiring reconfiguration.
3Measurement precision
If comprehensive working data is collected and analyzed, then accurate detection of productivity issues is achieved, but real-time detection and response capability is reduced
Solution Approach 1:
The system performs preliminary action by pre-defining the cycle data extraction framework and analysis parameters before actual productivity monitoring begins. The structure for detecting multiple factors (route characteristics, operational parameters, environmental conditions) is established in advance, enabling rapid real-time analysis without requiring complex on-the-fly computations. This preliminary setup allows comprehensive data analysis while maintaining fast response times for productivity deviation detection.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An object is to provide a mine management system that can accurately detect that productivity falls out of a plan in connection with a plurality of factors in various mines. A mine management system 1 includes a server device 200 that accumulates and processes working data of mining machines 101 to 103 and a determination device 250 that determines whether or not a working condition of the mining machines 101 to 103 in which the productivity falls out of a plan has occurred. The server device 200 has an extracting section 211 that extracts cycle data as the working data for each work cycle, from the accumulated working data, an indicator computing section 212 that computes a plurality of out-of-plan factor indicators representing occurrence factors of the working condition in which the productivity falls out of a plan in the cycle data, and a learning section 214 that allows the determination device 250 to learn a relationship between the plurality of out-of-plan factor indicators that identify cycle data in which the productivity falls within a plan.