Improvement of Shovel and Dumper Availability in Indian Surface Mines using Reliability Analysis

Abstract

Reliability, Availability and Maintainability (RAM) analysis of mining equipment is essential to reduce breakdown hours, operational cost and capital cost and enhance mineral production. The expenditure on mining equipment and expertise increases with the size and complexity of the equipment. A survey of the literature showed that no studies are available on reliability approaches based on subsystems using Markov and RBD model and studies that included mining factors, although a lot of studies exist on shovel and dumper. Consequently, the literature throws little light on the comparison of different reliability approaches between different mines, which is necessary to understand different mine factors. Therefore, the present study attempted to develop mathematical models based on reliability for shoveldumper system for different surface mines (surface coal mine, surface iron ore mine and surface limestone mine) and determine the life of subsystem. The study aims to overcome the major challenges in the mining engineering equipment of breakdown, complexity, size, competition, cost and safety of equipment as well as increased use of mechanization, automation and amalgamation. The research was carried out using quantitative approach as the present study determines the reliability in mining sector. The failure data were collected over a period of one years from daily downtime reports and maintenance records from SCCL, Telengana (for coal mine), M/s. Subbarayanahalli Iron Ore mines and M/s. Mysore Minerals, Sandure (for iron ore) and Thummalapenta Limestone Mine, Telangana (for limestone). To ensure that the data are independent and identically distributed (IID), the trend and serial correlation test were conducted. The analysis of the data was performed using Isograph Reliability Workbench software and characterised by Reliability Block Modelling (RBD) and Morkov modelling. To model the failure and repair processes of subsystems, RAM analysis using time between the failure (TBF) and time to repair (TTR) were carried out. The match factor for coal mine was 1:4, while it was 1:3 for iron ore and limestone mine, which led to the selection of two shovels and eight dumpers for coal mine and two shovels and six dumpers each for iron ore and limestone mine. A total of ten subsystems were formed for the shovel and dumper system. The highest failure percentage for coal mine was 42.1% viii and 48.3% for iron ore mine and limestone mine. Further, in trend and serial correlation analysis, no trend was observed between TBF and TTR was observed and the IID assumption was proved. The Kolmogorov-Smirnov test (K-S test) yielded the probability distribution functions for different subsystems of shovel-dumper system for coal mine as shovels as 0.373 (KS1) and 0.285 (KS2) and dumpers have 0.356 (BD3), 0.269 (BD4), 0.347 (BD5), 0.334 (BD6), 0.332 (KD7), 0.275 (KD8) 0.268 (KD9) and 0.291 (KD10); for iron ore mine as shovels have 0.312 (KS11) and 0.275 (KS12) reliability and dumpers have 0.359 (BD13), 0.342 (BD14), 0.332 (BD15), 0.409 (KD16), 0.393 (KD17) and 0.394 (KD18) and for limestone mine as shovels, 0.343 (KS19) and 0.348 (KS20) reliability and dumpers, 0.325 (BD21), 0.292 (BD22), 0.329 (BD23), 0.362 (KD24), 0.334 (KD25) and 0.304 (KD26). To improve obtained reliability of each system, the preventive maintenance is needed for each system i.e., shovel and dumper. The time interval for each subsystem of shovel and dumper were determined for the 90%, 80% and 70% of reliability of system. The different shovel-dumper systems considered in the study showed ‘infant mortality’ failure indicating manufacturing flaws during the early usage of equipment; however, it improved over time. RBD was used to develop mathematical models for series and series-parallel connection of the systems; whereas, Markov modelling was used for simultaneously-active and continuous-time. The mathematical model for series-parallel connection of the shovel and dumpers were developed based on their match factor and best fit distribution to improve the reliability. As well as Markov model also developed for same shovel and dumper for different mines by considering both failure rate and repair rate. The findings of the study revealed that the method of RAM analysis adopted in the study was highly effective in determining the time-to-failure of the shovel-dumper system that indicated the need for better maintenance plan and design modifications to improve the reliability of the system. The findings showed the necessity of standardizing the maintenance records to include causes and results of failure and delay time condition and providing training to staff on standardization. Creation of a web-based maintenance platform, real-time policies for maintenance, optimization of spare parts, crew members and inspection periods and long-term maintenance plans were also suggested.