Search Documents

By David Humphreys

It is difficult to overstate the importance of China to the future of the world mining and metals industries. In a remarkably short time, China has risen to become the world?s largest producer and consumer of steel and its second largest consumer of non-ferrous metals such as copper and aluminium. It is currently the world?s largest buyer of copper on international markets, the second largest buyer of iron ore, and the third largest buyer of alumina. At the same time, it is the world?s largest or second largest producer and exporter of tin, lead, zinc, magnesium and a host of minor metals and industrial minerals, as well as a fast-growing exporter of coal.

May 1, 2002

By P. Chaddock

The Thompson Mill has a capacity of 15,000 T ID of ore and produces two grades of nickel concentrate, copper concentrate and occasionally graphite concentrate. Up to July of 1985 we processed two distinct ore types: T-1 ore, averaging 2. 8% Ni and . 2% Cu and Pipe Open Pit ore, averaging . 8% Ni associated with peridotite and serpentine. Presently our ores from T-1 mine and Thompson Open Pit are of the same type and are processed together. (Figure 1)! n October, 1982, we decided to install an Outokumpu Courier 300 on-stream analyser on an initial 6 streams. The equipment was delivered to the mill on May 2, 1983 and the first analyses were generated by June 17th. The calibrations were 35% complete by the mid-July summer shutdown and were completed in the fall. 1. 2 Analyser Slurry Feed System:The key to successful on-stream analysis is delivery of the sample streams to the analyser.Our system uses pressure-thief sampling for collection. (Figure 2) Initially we used thief samplers modelled after the Outokumpu design, with turbulence bars upstream of the collection point. These units have been replaced with simple laterals as they have worn out. Our thief samplers are mounted vertically on pump discharges and we believe sufficient turbulence exists to ensure representative sampling.

Jan 1, 1987

By M. M. K. McCrank

"PRACTICAL SAMPLING TECHNIQUESIt is important to achieve reliability when sampling heterogeneous materials where there is a difference in composition, and non random distribution of the constituent particles. Throughout the design, construction and installation of sampling systems it must be borne in mind that the sampling point must be capable of:-allowing each particle in the stream an equal opportunity to be sampled.-accepting and collecting as large a volume or mass of material as necessary to satisfy the requirements of particle counting statistics.-establishing a sampling plan so that the sample is representative of the whole to a defined standard of accuracy.Design ConsiderationsFactors that may influence the design of the sample point include:-the mass, or volume flow rate of the material being sampled-the size and shape of the material or flowing stream-the physical condition of the material being sampled (i.e. temperature, wetness, size consist, density).-the chemical composition of the material being sampled -the location of the sample point and the space available for installation, extraction, reduction where necessary, secondary sampling, and return of the sample rejects downstream.-whether the process to be sampled is in an existing plant or one yet to be designed and built.-ease of access for servicing and maintaining the equipment.-whether the sampled material must be collected continuously or intermittently.-whether an existing sampling method is likely to be adequate or whether a new technique has to be developed."

Jan 1, 1989

By Jörg Fischer-Bühner

Conventional 22ct Au jewellery alloys are comparably soft in the as-cast and soft-annealed state and hardening by cold-working is the only way of giving strength to the jewellery. The paper first focuses on hardening of 22ct jewellery alloys by conventional means, i.e. solid solution hardening with base metals (aiming particularly at an increased as-cast hardness) as well as precipitation hardening by alloying with base metals beyond the solubility limits and subsequent age-hardening heat treatment. Co-additions of ~2% turned out to result in most pronounced solid solution hardening up to ~100 HV1 in an investment cast & quenched state. Subsequent heat treatments, consisting of solution annealing at 850°C for 1h followed by water quenching and ageing at 300°C ? 400°C for ~ 1h lead to pronounced age-hardening with a peak hardness of ~ 260 HV1. In contrast, 22ct Au alloyed with Ni, Fe or Sb display no age-hardening effect within the same range of heat treatment parameters. The 2nd part of the paper reports on some preliminary trials to enhance the hardening effect of the base metals by additions of suitable elements, which possess a strong tendency to phase formation with the added base metal during ageing treatments. It is shown, that the age-hardening behavior of 22ct and 990/995 Au alloyed with Co or Ni can be remarkably influenced by additions of Sb. A total of only 10wt%0 of CoSb or NiSb-additions lead to a peak hardness of ~ 150 HV1 for a 22ct Au and ~ 100-120 HV1 for 990 Au , which is obtained already after very short ageing times (20-40min) at 300-400°C after solution annealing at 700°C / 1h / water quenching. It is suggested that the age-hardening mechanism in these alloys consists in precipitation of intermetallic CoSb- and NiSb-phases.

Oct 1, 2003

By M. Y. Solar, W. G. Davenport

Most of the recent growth in nickel demand has been for the production of stainless steels and virtually all of it has been satisfied by new ferronickel and nickel pig iron projects. But "good" laterite resources are more and more difficult to find: Ni grades are decreasing while Fe contents are increasing and SiO2/MgO ratios vary from 1 to 3 and even higher if limonites are included. Predicting the smelting behaviour of such ores requires a more fundamental understanding of the thermodynamics of these systems in commercial operations. One valid question is thus whether these furnaces are at thermodynamic equilibrium or not. The present paper investigates this question, compares calculated and measured O2 partial pressures and mechanical/chemical slag loss ratios and concludes that ferronickel furnaces are not at thermodynamic equilibrium. This departure from expectation is actually to the operators’ advantage: the targeted ferronickel grades can be obtained at lower reduction potentials than calculated from pure thermodynamics.

Jan 1, 2014

By Brad Hewitt, Jeff Belke

"Murrin Murrin Operations has not had a successful beginning. Numerous reliability and operability issues many of which are the result of design decisions have been addressed through mechanical and operational changes and the CCD circuit is no exception. Four design aspects of the CCD circuit that have caused and continue to cause operational problems are:1. the inclusion of solution neutralisation gypsum in the CCD circuit feed which was designed solely for HPAL residue without proper investigation of its impact on the achievable underflow density and resulting wash efficiency;2. the decision to construct the thickener base from thickened solids;3. the lack of redundancy in the overflow, underflow and tailings pipe work; and4. the lack of thickener operational flexibility in the central rake-lift and feedwell dilution system.In November 2002, the lack of operational flexibility in the rake-lift and the feedwell dilution system through mechanical modification as follows:•, CCD1 has a modified rake mechanism that now allows central vertical lift of 600 mm;•, CCD1 has forced dilution using the Outokumpu Turbodil® technology and replacing the E-Duc® technology of Eimco;The decision to install, the installation and the results of the CCD1 rake-lift mechanism and feedwell dilution system are discussed in this paper. The installation was justified on improved operational robustness and an increase in the in the underflow density of 1 per cent (per cent w/w). At approximately the same time process modifications in the solution neutralisation circuit were implemented and these changes altered the rheology of the slurry reporting to CCD1. Since the modifications to CCD1 and the solution neutralisation circuit, the CCD circuit has averaged a 2 per cent improvement in the circuit underflow density.The CCD1 modification was completed on time and mechanically has performed well. From a risk point of view, the central rake-lift and Turbodil®dilution system add a degree of freedom and robustness to CCD1 operation that has extended the range of operational confidence. In this respect, the modification has been a success and is complementary solution neutralisation improvements."

Jan 1, 2005

By Bassam Kattaa

"Abstract - A heavy mineral survey of beach and dune sands including geochemical analysis for trace elements was undertaken to evaluate the economic potential of the sands and to understand the nature and distribution of the heavy minerals relative to potential source rocks. Results indicate that heavy minerals are generally derived from outcropping volcanic rocks that comprise a proportion of the hinterland in the south part of our survey area. The most abundant minerals are magnetite and pyroxene. The morphologies of the heavy mineral grains are angular to subangular, suggesting that they were transported short distances from the source rocks. Factors controlling heavy mineral distributions include wave and wind action, proximity to provenance, the fluvial drainage network, and longshore currrents.The most economically important mineral is ilmenite. Zircon and rutile concentrations are relatively minor and thus have lower economic potential."

Jan 1, 2002

By Rita Awwad

Building Information Modeling (BIM) has been gaining a significant edge in the construction industry over the last decade. BIM allows gathering all building information in one shared database that can help all construction entities better understand, integrate and visualize all work progress from inception to operation of the building. The main purpose behind BIM is to bring all project participants together (Client, Architect, Contractor, Consultant) since the initial stages of a project allowing them to cooperate and work as a team in the promise of an increased productivity, reduced cost, enhanced quality and faster delivery. However, a full-fledged implementation of BIM tools and benefits is not yet achieved in the construction industry and remains a debatable issue for researchers and practitioners in the construction field. This paper aims at assessing BIM awareness and usage in the Lebanese construction industry through conducting interviews with contractors, architects and consultants that are key players in the Lebanese market. A comprehensive literature review about BIM adoption in some developed countries and in the surrounding region is also provided in order to better evaluate the Lebanese industry status in comparison with foreign industries. This research sheds the light on BIM awareness of the different construction parties in the Lebanese market, assesses the extent of its current use, identifies the pertaining implementation challenges, and provides recommendations to enhance BIM role in the Lebanese construction industry.

Aug 1, 2013

By H. C. Counsell, R. Webber

"The mill at Denison Mines' Elliot Lake Operations started up in 1957, with a capacity of 6,000 tons per day. Over the years various plant modifications expanded th.is to 7 500 tons per day. In 1976, management decided to carry out a major expansion of the milling capacity, and during 1980 the grinding capacity was increased to 15,000 tons per day.The original comminution system consisted of a conventional crushing plant feeding minus S/s-inch material to rod and pebble mills, in closed circuit with mechanical classifiers. The key decision that had to be made was whether the expansion should include conventional or autogenous grinding flow sheets.During 1977, extensive pilot plant studies at Lakefield Research showed that two-stage semi-autogenous grinding, with some steel media in the primary mill and with pebble media in the second stage, produced stable grinding conditions. This type of circuit was selected by Denison for the expansion, and went into production in 1980. Primary millThe flow sheet of the semi-autogenous grinding circuit is shown in Figure 1. The primary grinding mill is a Koppers 28 foot by 10 foot mill, driven by a 4,000 h.p. synchronous motor. It is fed with a minus 8-inch run-of-mine ore, and 4-inch grinding balls are added as supplementary grinding media. There are two Koppers 15.5 foot by 22 foot secondary pebble mills, driven with 1500 h.p. synchronous motors. Each grinding stage operates in closed circuit with cyclones, and the primary cyclone overflow is the feed to the secondary mills. The circulating load in the primary stage can be controlled by adjusting the cyclone apexes."

Jan 1, 1984

By CIM Environmental, Social Responsibility Society

Within the mining industry, the pathway from exploration, to planning, construction, operation and closure has become more complex over the last few decades. The current financial, environmental and social influences present the mining industry with unprecedented challenges and opportunities. In addition, international environmental and social reporting standards and codes are becoming increasingly robust. The reporting requirements provide an added impetus for issuers to effectively manage environmental and social impacts and risks. Reporting requirements also provide the opportunity to be transparent. The intended audience for these Environmental, Social and Governance (ESG) Guidelines for Mineral Resource and Mineral Reserve Estimation includes practitioners, industry participants at all levels of skill and experience, corporate management and board members, government agencies, academic institutions and interested parties. For practitioners, this document provides underlying key principles of environmental, social and governance practices. The ESG Guidelines provide guidance for those using environmental, social, governmental and governance information when preparing and implementing mineral exploration programs and contributing to the preparation of mineral resource and mineral reserve estimates for mineral properties that are either at the study stage (i.e., scoping / preliminary economic assessment, pre-feasibility, or feasibility studies) or in operation. The guidelines can also be used to help mineral resource and mining companies apply good practice to manage their environmental and social risks and opportunities. Companies can identify and retain appropriate professionals to prepare environmental and social materials to support company studies. The guidelines are also a resource for companies developing effective environmental and social management strategies at any project stage. This document is not intended to be prescriptive and is not intended to provide exhaustive guidance on what to consider in terms of environmental, social, legal and governmental modifying factors in the preparation of MRMR estimates. Rather, it is intended as general guidance to assist companies and practitioners in considering and applying these modifying factors for MRMR estimates. This document is also not intended to guide environmental and social impact assessment (ESIA); however, the measures needed to manage impacts help inform the constraints and costs for defining mineral resources and mineral reserves and inform the costs and schedules of mineral project studies. All documentation of MRMR estimates made by an issuer and intended to be, or reasonably likely to be, made available to the public in a jurisdiction of Canada must comply with the requirements of National Instrument 43-101 Standards of Disclosure for Mineral Projects (NI 43-101), as amended from time to time (Canadian Securities Administrators, 2016a and 2016b). Disclosure is not addressed in these ESG Guidelines. Determination of what to include in disclosures is defined in the applicable Securities’ legal frameworks. Appendix A presents a summary table showing the linkage of the principles and topics in these ESG Guidelines to the potential effects on project costs, schedules, government permits and approvals, social licence to operate and financing.

Sep 8, 2023

By M. Lindgren, M. Latva-Kokko

Process performance can be significantly influenced by proper design of hydrometallurgical reactors. Optimal reactor configuration is case specific and depends on process requirements such as reaction rate limiting factor and particle size distribution. Choice of materials and manufacturing methods also play important roles in highly corrosive and abrasive environment often involved in hydrometallurgical industry. In this paper performance of different reactor designs has been compared in some typical hydrometallurgical applications. Results show that significant improvement in energy consumption, gas utilization efficiency and equipment lifetime can be achieved with process specific reactor design.

Jan 1, 2014

By S. K. Mitterecker, T. Hofbauer, F. Baerhold

"Acid regeneration is a very important process step in each acid leach circuit. It is required to recover dissolved metals and HCl from spent leaching acid. Andritz Metals presents its recently commissioned spray roast pilot plant. The facility is now available for testing and demonstrating acid recovery or metal oxide production which is a requirement in several flow sheets under development for the mining and specialized oxides industries. Results from pilot test campaigns are reported here. We are currently focusing on a nickel ore leaching process, which requires both a feasible iron control and acid regeneration step. In addition, we disclose the results of a magnesium ore and aluminium recycling test campaign conducted in our pilot plant facility.INTRODUCTION Hydrochloric acid regeneration is widely used in the steel industry to regenerate spent pickle liquors into hydrochloric acid at a typical concentration of 18 wt.%. Andritz Metals (formerly known as Ruthner) has built more than 200 acid regeneration plants worldwide (Andritz Metals brochure). As technology leaders Andritz also has vast experience with specialty metal oxide plants, producing metal oxides such as aluminium oxide (Al2O3), magnesium oxide (MgO), nickel oxide (NiO), cobalt oxide (Co3O4) and mixed oxides from metal chloride solutions (Baerhold F., Lebl A., Starcevic J., 2006), (Jedlicka, H., n.d.), (Ruthner M.J., Krischner H., 1973), (Ruthner Ind. Planungs AG, 1967). Hydrochloric acid leach circuits have been developed and proven in the industry for many years. Examples of industrial applications are: synthetic rutile production (TiO2) (Walpole E.A., Winter J.D., 2002), Quebec Iron and Titanium’s UGS process (Patoine M.-C., De Mori A., Borowiec K., Coscia C., 2002), magnesium and potassium production (Priemer H., 1973) and Goro Nickel (Vahed A., Colton D.F., Duterque J.P., Karner W., Baerhold F., 2004), (Taylor A., 2013). One of the critical process steps in developing a hydrometallurgical process is acid regeneration, as it is an economic requirement to recover both the dissolved metals and acid (commonly HCl) from the spent leaching solution."

Jan 1, 2016

By G. Hausen, H. Micka

"Modern signal and power electronics made adjustable-speed AC drives technically feasible and reliable and economically attractive. Cycloconverters are discussed as a proven technology for mine hoists. \IntroductionUntil rather recent technical and commercial advances in power and signal electronics, DC motors were used almost exclusively for applications requiring continuously adjustable operating speeds and torques. More recently, the significantly higher efficiency and the accordingly lower operating costs of AC motors gained importance. An effective and efficient speed control of AC motors requires the control of AC voltage and frequency. Table 1 shows typical applications of adjustable-speed AC drives.CycloconvertersOf particular interest to the mining industry are cycloconverters for hoist drives. The cycloconverter represents by far the simplest, most efficient and most economical means of converting a constant-voltage, constant-frequency supply into an adjustable-voltage, adjustable-frequency supply as required for the speed control of AC drives operating in the frequency range to half of the utility frequency.Cycloconverters feeding synchronous motors or induction motors are ideally suited for slow-speed direct-coupled drives."

Jan 1, 1988

By F. Baerhold, S. Mitterecker

"Acid recovery is an important process step in pickling and leaching processes. Pyrohydrolysis in spray roasters and fluidized beds is a well-established technology to recover hydrochloric acid, but the hydrolysis in the liquid phase is more and more considered as an alternative new process route. The characteristic features of these three processes will be discussed along with select industry trends. For example, hydrochloric acid based leaching processes for ilmenite and nickel laterite ores produce far larger volumes of waste acid streams than the steel industry. Scaling-up of the spray roasting technology to the highest possible capacity is necessary in order to become more cost effective. In addition, process routes based on ferric chloride pyrohydrolysis allow the production of super-azeotropic HCl (up to 30 wt.%). Finally, the new ECOmode concept optimizes the operation of both pickling lines and the spray roast acid recovery plant. Long term results from industrial applications show that fuel savings in the spray roaster of 25% are easily achieved without additional process steps. An attractive feature is also to upgrade existing installations to higher plant capacities.INTRODUCTION Pyrohydrolysis is the conversion of metal halides into metal oxides at elevated temperatures in the presence of water vapor. Reaction products are a hydrogen halide gas and a metal oxide. The process was originally developed and introduced into the market by ANDRITZ Ruthner in the 1960’s. It has found wide technical application in the field of acid recovery and oxide production. The introduction of this process has revolutionized pickling technology. The use of hydrochloric acid has strongly increased over the last decades, since pyrohydrolysis is available for the total recovery of waste pickle liquors from mild steel pickling, as illustrated in Figure 1.Chloride media are also used increasingly in the metallurgical industry. Oxide production processes for Mg, Co, Ni, and Al have been installed. Processes using hydrochloric acid are known for the leaching of ores as well as for the re-extraction of metals from organic solvents. In both cases the recycling of HCl is of substantial importance to achieve highly economic processes and to reduce waste."

Jan 1, 2016

By M. C. Pietrobon

"A common problem in assessing the applicability of laboratory flotation data is determining how closely conditions during laboratory testing of the ore matches those that occur in a plant. Typical laboratory based investigations do not consider factors such as water quality, pulp temperature and the pulp chemical environment during both grinding and conditioning on laboratory flotation performance. Disregarding these important factors may seriously compromise the validity of conclusions based solely on laboratory experimental results. Furthermore, considering these factors during the flowsheet design stage in the development of new ore resources may reduce financial risk associated with building a new plant. A methodology is described here which embraces these factors in an attempt to bridge the gap between laboratory and plant. The methodology should be considered a vital first step in all laboratory based experimental programs and has implications for fundamental studies on the flotation behaviour of single minerals. This methodology increases the confidence with which findings of laboratory based studies can be transferred to the plant and highlights the effect of variables which may be accessible to manipulation in the plant.Pulp chemical parameters, such as pulp dissolved oxygen demand, pH, Eh, mineral oxidation and solution composition were manipulated in the laboratory by adjusting the duration of gas purging after grinding, the process water input and pulp temperature. A key finding was that matching of the laboratory pulp chemistry to the plant pulp chemistry was dependent upon oxidation after grinding. In turn, mineral oxidation was dependent upon the pH in grinding, water composition and the duration of oxygen purging after grinding. The methodology closely matched the dissolved oxygen demand and surface chemistry of laboratory and plant pulps."

Jan 1, 2002

By D. Duff, R. A. Cameron, R. E. S. Whitehead

"Although CO2 content indicates the quantity of carbonate in altered basaltic rocks, it does not measure the intensity of carbonate alteration. CO/CaO molar ratios, which correlate with the relative proportions of individual carbonate species, measure the intensity of carbonatization. The Timmins gold area is characterized by pervasive carbonatization, with large parts of the area having CO2/Ca0 ratios greater than 1, representing C02 supersaturation relative to CaO. The mean CO/CaO in the Timmins area is0.67, compared to 0.13 in two unmineralized control areas. Regional arsenic abundance in the gold-bearing area averages 4 ppm, compared to about 0.35 ppm for the non-mineralized areas. The average K20 and Rb abundances are similar in both mineralized and non-mineralized areas (about 0.30% K20 and 7 ppm Rb). Although there appears to have been no wholesale addition of K20 or Rb to the mineralized area from external sources, these alkalies have been redistributed internally, resulting in patterns of local enrichments and depletions which are characteristic of the mineralized region.Local target anomalies associated with known mineralization consist of CO2/Ca0 > 1.5, As > 10 ppm and K2O > 0. 75%. Not all three values are required to be above these thresholds in order to identify local targets. Combinations such as C02/ Ca0 > 1.5 and As < 10 ppm, CO/CaO 1-1.5 and K20 > 0. 75 or K20 > 0. 75 and As > 10 ppm are equally indicative of proximity to gold occurrences. Gold-bearing veins and associated anomalies occur throughout the entire volcanic succession and are not restricted to any particular stratigraphic horizons. The levels and patterns of alteration and occurrences of gold appear to be structurally controlled."

Jan 1, 1982

By Å. Kruukka, H. F. Broicher

The iron ores of the Kiruna mine in Sweden are of exceptional quality except for the phosphorus content in certain zones of the orebody. The prevention of mixing ores high in phosphorus with those low in phosphorus has always been of significant economic importance. The ongoing introduction of tele-mining and automatic haulage by LHDs requires new means in the field of quality control underground, as the experienced miner as well as the underground laboratory assistant no longer are involved in the process. As of 1998, LKAB has been testing prototypes of LIF (laser-induced fluorescence) analyzers for non-contact and real-time measurement of the phosphorus content of ore being transported by LHDs from the stopes to the respective ore pass assigned to a certain ore quality. The principles of the analysis of laser-induced fluorescence, the different test phases and results are described and conclusions made. With bulk sorting of ore underground, the mining engineer supports the efforts of the metallurgist in the processing plant.

Jan 1, 2002

By A. J. Neale, B. C. Flintoff

"In most mineral processing operations the term audit is usually reserved for the exercise of checking financial records and/ or safety practices. While these activities are time consuming, no one disputes their necessity or value. If we subject these areas to such careful scrutiny, why is it that we don't normally apply the same principles to critically examine process operation. After all, the operating efficiency of the plant impacts directly on the measures of profitability.A mineral processing plant is a complex system that has a number of distinct, but mutually dependent elements: process equipment, instrumentation, operating policies, maintenance policies, control strategies and people. To ensure maximum economic performance, in the dynamic environment that characterizes a concentrator, requires a periodic review of each of these areas. However, to be useful, the study must be based on sound metallurgical principles and be conducted in a highly focused manner. Experience teaches us that the exercise is very revealing and potentially very rewarding.This paper will discuss the process audit process and illustrate its potential value through illustration taken from plant studies.IntroductionThe last twenty years have seen tremendous technological and organizational changes sweep the minerals industry. Process control systems, on-stream analyzers, and gold recovery techniques barely conceived twenty years ago are now incorporated in virtually all new installations in an effort to maximize operating efficiencies. In conjunction with these developments, changes in the way plants are managed, down-sizing of technical staff, reduction in the size of operating crews, and the flattening out of the management structure have significantly reduced operating costs. As a result, plants are making more efficient use of our natural resources, minimizing the costs of production, and in turn maximizing operating profits. Or are they?"

Jan 1, 1991

By B. Arlyuk

On the basis of research of relations among the market participants the analytical model has been created which connects the price of industrial good at stock exchange with volume of sales and the stock of the sellers and buyers, Also a model was developed for assessment of these parameters of the market status by the actually fixed stock exchange and general market indicators. The composition of these models made it possible to create reliable system of short-term price forecasting for commodities or base metals up to one quarter ahead. Imperfectability of creation a model for reliable price forecasting by purely statistic methods (by the "black box" principle) was shown. A model of price behavior qualitative forecasting till one quarter ahead has been developed, which determines the probability of reaching in future of the given level as well as of the price maximums and minimums, Optimal values of threshold probability were determined for forecasting the future price maximums and minimums securing the maximum profit from the forecasting system application for commercial activity. The developed system of forecasting was identified for primary aluminium at the period of 1990-2003. The system was tested daily during 20 months period for speculations on 3M primary aluminium forwards at London Metal Exchange with high commercial result.

Jan 1, 2004

By Jeffrey L. Pagnutti

Train derailment can lead to loss of capital, damage to the environment, and loss of life. Machinery condition monitoring techniques can be used to develop a structural health monitoring system for railway track to detect faults at the incipient stage, before catastrophic failure occurs. This study uses vibration measurements from a novel material haulage unit to develop a decision support system for railway track maintenance. A variety of feature sets and classifiers were compared to determine the optimal configuration for fault detection. The results highlight the shortcomings and advantages of certain methods and describe the process in which the optimal result was obtained. The Parzen data descriptor paired with a refined statistical feature set performed the task of fault detection with less than five percent false positive and false negative classification error. Overall, the outcome of these experiments indicated that condition monitoring of railway systems is a promising approach for fault detection and has the potential to avoid catastrophic failure.

May 1, 2013