Investigations on Performance of Load Haul Dumpers in Underground Mines and Improvement of its Availability and Utilization Using Reliability Analysis

Abstract

Every organization in today’s competitive world intends to improve its economy by increasing their production and productivity rates. Unequivocally, the production in Indian underground mines over the years is not satisfactory, due to a variety of reasons. There are manifold of avenues for the betterment of production and one such approach is through enhanced utilization of mechanized equipment such as Load Haul Dumper (LHD). These LHDs are prone to continuous and random occurrence of numerous potential failures during the operation. Understanding of each failure mode will help to take an appropriate maintenance action, which leads to reduce the downtimes of the machinery. One approach of productivity improvement efforts is through an increase in percentage availability and utilization of these machines. The higher availability and utilization of these machines under certain operating constraints, leads to an increase in the production and productivity of these capital intensive equipment. The accomplishment of this goal can be possible only with the improvement of the reliability of equipment by reducing the occurrence of breakdowns. The purpose of the research performed for this thesis is to make an attempt to control the occurrence of uneven breakdowns, by using reliability analysis. The developed methods can be used to identify the problems/causes affecting LHD downtime, to assess suitable maintenance management strategies for repair and replacement action and to identify the economic lifetime of LHD systems. The major objective can be explained as providing quantitative forecasts of diverse performance characteristics of LHDs through reliability computations, evaluations, and forecasts. Such characteristics are Reliability, Availability, and Maintainability (RAM), downtimes, frequency of failures, and Overall Equipment Effectiveness (OEE) of the system. The estimation of these characteristics is significant for optimal decision making. To perform the reliability analysis, a fleet of LHDs deployed at both coal and non-coal mines of M/s The Singareni Coal Collieries Company Limited (SCCL), Telangana, M/s The Hutti Gold Mines Company Limited (HGML), Karnataka, and M/s The Hindustan Zinc Limited (HZL), Rajasthan were selected to carry out the field investigation for collection of failure and repair data of equipment. Before performing the reliability analysis it is necessary to perform the trend and serial correlation tests to determine whether and how the failure patterns are changing with respect to time and to validate the Independent and Identical Distribution (IID) nature of the data sets. As the data have been collected from the field investigations and if such tests are not performed, then there is a possibility of arriving at incorrect conclusions. On analysis it has been found that the time between successive failures (TBFs) of LHDs is free from the presence of trends. Statistical based analysis like statistic U-test (Chi-squared test) has been carried out to determine the method for reliability modeling. Generally, the Renewal Process (RP) method can be utilized for perfect repairs and the Non-Homogeneous Poisson Process (NHPP) method can be utilized for minimal repairs. However, most repair activities may realistically not result in such two extreme situations but are a unique combination in this range. In this study, the RP approach has been utilized to perform the reliability analysis for estimation of percentage of each sub-system (i.e., Engine, Braking, Transmission, Tyre, Hydraulic, Electrical and Mechanical) reliability and failure rate of the LHDs. Best-fit distribution of data sets have been identified by the utilization of Kolmogorov-Smirnov (K-S) test. Maximum Likelihood Estimate (MLE) method has been used to estimate the theoretical probability distribution parameters (η, β, and γ) of best-fit curves. Reliability of each individual sub-system has been computed according to the best fit distribution. In addition to the operational procedures and technical expertise, maintenance efficiency is also a significant factor that needs to be considered for assessment of system performance or effectiveness. Improvement of performance of the equipment mainly depends upon the adoption of suitable maintenance management actions (repair/replacement). Keeping this in view, in this analysis maintainability percentages has been estimated for the LHDs after successful identification of best-fit probability distribution function. Further, the reliability-based Preventive Maintenance (PM) time intervals were forecasted for the enhancement of equipment reliability. vi vii vii vii In this research work, the most appropriate methods of reliability modeling and optimization such as Failure Mode Effect Analysis (FMEA), Reliability Block Diagram (RBD) and Fault Tree Analysis (FTA) are presented to estimate the percentage of reliability, availability, and maintainability (RAM). The FMEA approach has been utilized in this research study to investigate the failure behavior of the LHDs. FMEA identifies the reasons for occurrence of potential failure modes and provides the necessary recommendations or corrective actions to reduce the uneven occurrence of failures. The risk-based numerical assessment was made with prioritization of critical failure modes through the Risk Priority Number (RPN) calculation. A new risk management approach known as “MATLAB Fuzzy rule based interface system” was utilized to validate the calculated RPN values. The series and the parallel configuration systems are the two important approaches in RBD approach and are widely used to estimate the overall system reliability of the LHDs. In this study, all the connections of the components in a machine have been identified in series configuration system. Hence, the overall system reliability of the LHDs were estimated using series configuration system calculation. In addition to that, the reliable life (TR) of the equipments were also calculated to forecast the duration (threshold time) for occurenance of next subsequent failure. In this study, FTA was carried out to identify the percentage of unavailability of the system and to know the influence of each component/ cut set/ critical part on the system failure. Further, the computed values of reliability, availability and maintainability were validated with MATLAB based Artificial Neural Network (ANN) results for identification of percentage error value. In addition to this, an attempt has been made for the improvement of performance of the equipment through evaluation of Overall Equipment Effectiveness (OEE). The OEE percentage of each LHD has been computed in terms of percentage availability, performance and quality. This thesis provides a scientific base for evaluating the reasons for performance drop of the equipment and suggests the necessary remedial measures and recommendations to the mining industry for the improvement of performance on the basis of RAM analysis and OEE calculations.