Search Documents

By I Canbulat, J Emery, L Yao and C Zhang

This paper documents a process of data analytics and machine learning applied to coalmine roof convergence data. It follows a publication documenting the formation of the database. For each set of analyses Python scripts were utilised to extract data from an SQL database, with subsequent analyses also executed in Python, utilising open-source tools such as scikit-learn. Preliminary analyses with classic machine learning algorithms included multiple logistic regression, K-nearest neighbours, decision tree, random forest, support vector machine, kernel support vector machine, artificial neural network (ANN), naïve Bayes and multiple linear regression. Results indicated that relationships were complex and non-linear, with the artificial neural network most suitable. An ANN model was then refined through a standard process of hyperparameter optimisation and data augmentation to arrive at a final model. Due to the black box nature of the ANN further insight into how the parameters interacted was sought and found with Shapley additive explanations (SHAP values). SHAP values utilise Game Theory, where each input parameter or feature is a ‘player’ while the data set is the ‘team’. The SHAP value is the impact of each player on the target value, essentially establishing the contribution of each input parameter. The results from all analyses are delivered in the paper, together with insights, applications, strengths, and limitations of the methodology.

Nov 29, 2022

By Greg Sheppard, Eduardo Nebot

Australian researchers have developed new technologies aimed at improving safety and productivity at openpit mines. A data acquisition and feedback system for electric rope shovels has been shown to substantially reduce cycle times and decrease excessive machine duty caused by stalls. This data acquisition system was developed by the Cooperative Research Center for Mining Technology and Equipment (CMTE). It has been tested on a P&H 4100A shovel at Coal & Allied’s Hunter Valley operations and a P&H 4100XPB at BHP Billiton Mitsubishi Alliance’s Goonyella Riverside Mine in Central Queensland. In addition, CMTE member P&H Mine Pro Services has used the technology to acquire field data from a P&H 4100XPBrope shovel at RAG’s Belle Ayr Mine in Wyoming’s Powder River Basin.

Jan 1, 2002

By Y Tang, G Si

The analysis of rock fragment sizes plays a pivotal role in various stages of geotechnical and mining engineering. Traditional case-based methods often falter in addressing the variability in rock fragment images due to fluctuating geological conditions and imaging environments. To counter this challenge, AI-based models necessitate training on expansive and diverse data sets, encompassing a wide range of conditions and fragment types. However, the creation of such data sets is often hampered by financial and environmental constraints. This paper introduces a novel approach for rock fragment recognition, which leverages the power of AI for data augmentation, automated annotation and prediction. The methodology integrates three key components: Style-Based Generator Architecture for Generative Adversarial Networks (StyleGAN) for data augmentation, the Segment Anything Model (SAM) for automated annotation, and the multiscale version of you only look once (YOLO-MS) real-time instance segmentation model. The proposed method begins by training a new StyleGAN model using the original data set. It then delves into exploring the latent vectors of different rock fragment images to generate specific ‘style representations’. These latent vectors form the foundational elements of each image in a high-dimensional space, encapsulating critical features such as size, shape, texture, and colour variations. The next step involves conducting quality checks on the generated images to ensure their realism and suitability as training data. Subsequently, the augmented image set is annotated using modified SAM module, followed by training the YOLO-MS real-time instance segmentation model. The performance of this model, trained on the augmented data set, is compared with a model trained on the original data set, particularly in terms of generalisation abilities in new scenarios. The results exhibit a significant enhancement in accuracy, robustness, and generalisability of the model trained on the augmented data set, underscoring the potential of StyleGAN as a powerful tool for data augmentation in geotechnical engineering image analysis.

Sep 1, 2024

By Márcio Aurélio dos Santos

A Automação Industrial utiliza de forma intensiva os mesmos equipamentos e softwares de TI como servidores, historiadores, switches, firewall, dentre outros. No entanto, são poucas as áreas de automação que aplicam de forma eficaz todos os controles e cuidados que esses ativos requerem. Neste cenário, a garantia de altos níveis de disponibilidade dos sistemas de automação inicia-se pelo ambiente físico no qual tais sistemas estão instalados. É comum a improvisação de ambientes inadequados dentro das áreas industriais ou a instalação em subestações elétricas que não foram construídas com o rigor necessário para receber os ativos de TI Industrial utilizados nos sistemas automação. A solução para manter elevados os índices de disponibilidade destes ativos é, sem dúvida, instalá-los em Data Centers, comumente utilizados nas áreas de TI, o que é facilmente justificável pelo fato de que a indisponibilidade dos ativos de automação representa um impacto maior aos resultados das indústrias ao se comparar com a indisponibilidade dos ativos de “TI Corporativa”. Assim, este trabalho conceitua um ambiente de Data Center, discute as particularidades relacionadas à Automação Industrial (projetos de cabeamento estruturado e as interligações entre os níveis 1 e 2 de Automação) e apresenta as estratégias que a Samarco Mineração está adotando com relação ao tema.

Jul 30, 2018

By Reinaldo J. Villa, Joshua C. Adams

The I-395/SR-836/I-95 reconstruction project is an $818-million design-build construction effort that includes construction of viaduct, segmental and signature bridges in Miami leading into Miami Beach in Florida. Based on the load testing performed during the design phase, the design-build team determined that conventional driven pile and drilled shaft foundations were not feasible for support of segmental and signature bridges. Due to the weak limestone formation and successful use of auger cast-in-place (ACIP) piles in the Downtown Miami area on commercial developments, ACIP piles were selected as a foundation system to support bridges. This is the first time in history where the bridges will be supported by ACIP pile foundations in an interstate highway in the State of Florida. The design build-team worked closely with the Florida Department of Transportation to develop requirements for the load test program. The paper will provide an overview of the geotechnical data collected, results of dynamic/bi-directional load testing performed on test piles, data collected, installation of piles, and lessons learned. The load test program included a total of fifteen test piles with dynamic, bi-directional, and lateral load testing on 30 and 36-inch diameter ACIP piles and instrumented with thermal integrity profiling wires.

Sep 8, 2021

By Thomas V. Falkie, R. Venkataramani

The empirical nature of this investigation required measurements of variables under operating conditions in underground coal mines. The variables to be measured were based on the postulated character¬istics of the loading function, as described in the previous chapter. One class of variables presumed to influence the loading function are geological variables. Accordingly, the sites for data collection were selected in different bituminous coal seams at dispersed locations in order that variation in geological conditions would exist. Seven mines in central and western Pennsylvania and one mine in northern West Virginia were selected as data collection sites. Five different coal seams were represented by these mines. The remaining classes of variables - geometric, mechanical, and procedural - were measured as they occurred in the mines. A description of the measurements, and the conversion of these measurements to forms which could be-analyzed for effect on the loading function is given. Mining Dimensions At the underground working face, the following dimensions were measured: a. Thickness of coal excavated. b. Thickness of shale excavated. c. Total height excavated. d. Width excavated. e. Depth of penetration of each machine cut. With the exception of the last of these dimensions, all were measured

Jan 1, 1972

By S Versace, D Arnott, D Bojcevs, A Shaw

The George Fisher deposit is a significant sediment hosted silver-leadzinc resource located 2 km north of the Hilton Mine in NW Queensland. The George Fisher feasibility project involved the completion of underground and surface drilling programs to delineate Measured and Indicated Resource which could be converted to an Ore Reserve through mining and commercial studies. The geological information which was collected and used for the feasibility study included a major program of diamond drilling, predominantly from underground sites, but with some drilling also completed from surface. All drillcore was geologically and geotechnically logged, sampled and assayed. Data was also collected from underground cross-cut and drive development which was mapped for rocktype, ore distribution, structure and geotechnical condition. This new geological data, generated during the Pre-Feasibility and Feasibility stages, was combined with results of all previous drilling campaigns in the George Fisher area to produce a comprehensive database on which geological modelling and Resource estimation could be based. To allow fast and accurate collection and utilisation of data from the drilling program, digital collection processes were used where possible. This resulted in rapid data entry into the project database, significantly enhanced data quality, 100 per cent data ownership within the project team and increased productivity levels. Processes used during the project data collection program are described, lessons learned are related and suggestions for ænext timeÆ are listed using the benefit of hindsight.

Jan 1, 1998

By Jeonghwan Kim

Intelligent Excavation System (IES) has been considered to be one of the desirable solutions to improve the productivity, quality and safety of the current earthwork processes. IES is composed of various hardware and software modules including task planner (work command generator), local and global work environment sensors, GPS and other sensors to acquire the excavator?s status, robotic controller and electronic valves, and the manipulator (excavator). A data communication manager was required to coordinate the signal or data-flow in between these modules considering the operational status of IES such as; excavation start/completion, truck detection, terrain model update, emergency stop, local sensing, GPS/motion sensor data handling, and so on. This paper presents the role and functions of this data communication manager. After every possible operational status of IES operation was identified, a protocol of each communication process in the form of string-type was made. The components of IES are connected with either wired or wireless network. Therefore, they can communicate with the manager using socket (TCP/IP) communication. The data communication manager is also capable of accessing a web-based project management information system, which constantly updates the performance of excavation robot to online database.

Aug 1, 2013

By G. S. Hill

A large amount of the stereological error associated with liberation studies can he eliminated by assessing areas instead of linear intercepts. The presentation of a simple geometrical model as a stereological 'worst case' allows calculation of the worst possible grade/recovery curve for a given assemblage of areas. The true results are expected to lie in the window between this worst case and the hest possible case, represented by uncorrected data. The width of this window typically corresponds to an error margin of about 5% grade at a given recovery. An indication of where the data lies within the range of possible grade/recovery results may he obtained from a qualitative analysis of the locked section complexity. Assemblages consisting of texturally complex locked sections (ie. containing many mineral nuclei) are likely to exhibit minimal stereological bias and therefore will tend to lie close to the uncorrected side of the grade/recovery window.

Jan 11, 2011

By D W. Evans

Research into the dynamic properties of ground support elements continues to grow at a global level, given the increasing importance of this field of application within underground hard rock mining. Measuring the mechanical response of such elements under the rapid onset of simulated rock burst loads forms the basis for this experimental work, typically utilising gravitational drop testing methods. This field of research is fundamentally driven by an increasing need for dynamic performance data, given the constant evolution of rock bolting product designs, materials selection, combination testing and specific application needs based on ground types and loading scenarios. A smaller number of globally recognised dynamic test facilities have predominantly lead research within this arena, typically in association with academic institutes, larger industry groups and government research institutes. However, due to the increasing requirement for dynamic performance data, additional facilities have now emerged in order to satisfy the overall demand for test work. These newer facilities are more typically associated with suppliers to the underground hard rock industry, with experimental designs looking to replicate, at least in part, the work of the primary facilities. For standardisation, repeatability and calibration purposes, the cross validation of test data between primary and secondary facilities becomes an obvious and important consideration across this field of research. This paper outlines the development of a new dynamic drop test facility for testing rock bolts, including the structural design and associated high speed data capture system. Based on the utilisation of identical bolt types, material grades, anchoring systems and decoupled ‘free-lengths’, comparative data has been generated to provide cross validation of the new test facility against data generated by a globally recognised primary facility. The cross validation of data provides increasing confidence in the repeatability and accuracy of the new facility as activity continues towards the provision of further experimental test work.

Nov 29, 2022

By Aaron Samuel Young, Hiago Antunes Amador de Oliveira, William Pratt Rogers

Low commodity prices and lack of new high-grade deposits are pushing the mining industry for innovation. Lean mining has been vastly discussed in the past decades as a way to continuously improve mining efficiency. Although very theoretical, lean mining presents a set of new solutions enabled by the digital transformation that the mining industry is going through. This paper proposes a framework for a short interval control system with the purpose to allow timely responses by front line supervisors upon operational deviations. In the first part, the author briefly reviews concepts of short-range planning and lean mining. Then, the materials and methods used for the development of the short interval control system are discussed. Finally, alpha trial results are presented, and next steps discussed.

Oct 3, 2019

By Michael Arnold, Jignyash Manyam, Yoav Rappaport

The USACE required construction of over 13 miles of cutoff wall at an average depth of 65 feet for the Herbert Hoover Dike (HHD). To meet project requirements, to provide quality control and make sound operational decisions, automated data capture, retention and visualization is crucial. Accurate and complete data management allows the authors to deliver critical business insights and to improve operational efficiencies. The author’s understood the necessity of capturing all their data, and efficiently processing and displaying it for multiple stake holders, for both internal and external purposes. The authors realized early in the project life cycle that data ownership and management is one of the fundamental features of the construction process. The authors developed their own software and use two in house applications; WEB-BGM online portal and, b-project which store rig production information combined with field quality control. b-project is used to create and deliver the Enterprise Database (EDB) to the USACE. Another biproduct of capturing the production data is the generation of the Geographic Information Systems (GIS) and also business analytics visual interactive display, developed through Microsoft Power BI. The paper focuses on the business process workflow and various technologies employed and lessons learned from the field operator perspective to achieve higher returns on business software operations.

Oct 1, 2022

By S. Möllerherm

"The evaluation of data and information provided by the existing traditional indicators in sustainability assessments is most important. Therefore the Institute of Mining Engineering I from the RWTH Aachen University developed indicators and an assessment system for the mineral raw materials industry on operation, company and sector level on behalf of the Federal Ministry of Economics and Labour (Bundesministerium für Wirtschaft und Arbeit, BMWA) and the Federal Institute for Geosciences and Natural Resources (Bundesanstalt für Geowissenschaften und Rohstoffe, BGR). In the course of the development of indicators for the raw materials industry quality indicators on different levels of the sustainability assessment procedure have been identified and developed. The following paper describes some of the results focussing on the development and integration of quality indicators into sustainability assessment systems."

Jan 1, 2005

By Richard P. Sheldon

In the last several decades, techniques of data collection have been rapidly evolving. Automated spectroscopic techniques in the chemical analysis of rocks for example have advanced to the point that we can analyze samples faster than we can grind them. Remote sensors of LANDSAT give us intensity measurements in four color bands of every acre of the U.S. every 19 days. Common depth point seismic instrumentation systems used at sea for surveying and networks of seismometers arrayed around the country for earthquake studies likewise are generating vast amounts of data. These along with many other earth science data generating systems have caused a veritable data explosion. How do we deal with these enormous amounts of geologic data presently being produced as well as those produced in the past, and to be produced in the future? There is an immediate need to acquire and utilize past and present basic data in order to address today's national earth science problems. Also data being collected and stored now must be compatible with future data in order for us to be able to undertake the wide spectrum of analysis that must be made in the future. This is a big order! In order to organize and to be able to analyze these data, the USGS has embarked on a wide ranging program of compatible data base construction and the development of computer technology, based on interactive computing, to facilitate resource assessments and analyses. These activities are analogous to two sides of a coin. One side deals with the logic, philosophy, and mechanics of information storage and retrieval; and the other side deals with the sensible application of computer technology to the treatment of data. These

Jan 1, 1977

By Christoph Muller, Geoffrey Liggins, Mohan Hensman

Modern commercial blasting is as much a technical process as it is a business one. As with most processes both aspects of blasting are driven by and generate new data. The time-critical nature of commercial blasting in mining and construction projects, dictate the timely exchange of information so as not to impede production or commerce. Up-to-date information about the blast to be undertaken can help suppliers: meet customers’ expectations to address challenging ground conditions; react to a market where quickly changing orders occur regularly; and address as-drilled blast patterns as opposed to baseline plans. “Real-time” data exchange into and out of the blasting process supports the whole-of-mine data management that the industry is migrating to.

Jan 1, 2010

By S. Dessureault, J. Porter, M. Woodhall

Information and communication technology (ICT) is seen as a key source of future productivity improvements in mines. The information technology infrastructure (ITI) necessary to achieve these productivity improvements will come in the form of data integrated from multiple source systems throughout the mining value chain. Two pure approaches are available: supplier-controlled or operations controlled data structures; each strategy has its own advantages and disadvantages. Case studies involving the South African mining industry are used to demonstrate the advantages of using hybrid-controlled data structures. Depending on the level of operational control, new skills are also necessary to be learnt by mining professionals including: information engineering; creating consistent and quality data; and process improvement techniques that use the integrated data for improving the operation.

Jan 1, 2004

By S. Goodman

Successful exploration for and exploitation of a mineral deposit requires an accurate understanding of the deposit architecture in 3D, and of the development of the mineralizing system through time, which requires a thorough comprehension of the structural framework of the deposit. The integration of geological, geophysical and geochemical datasets to arrive at a 4D structural framework is well-established and has been successfully applied by SRK Consulting in projects from deposit to regional scale. In some cases, we can go beyond this, applying the SEEBASETM technique, which utilizes a ?bottom-up? approach to basin analysis. SRK?s approach primarily relies on the interpretation of magnetic and gravity data, calibrated with many other datasets including mapped geology, topography, event histories, wells/drillholes and seismic data. The SEEBASETM model is a culmination of this data integration and is produced through an iterative process of structural interpretation and geophysical modeling, where an interpretation of basement composition and 4D structure is used to constrain depth-to-basement geophysical modeling, and the modeled depths are then used in conjunction with the structural interpretation to constrain contouring of the basement topography. The result is a rigorously calibrated, 3D view of the basement surface, which can be used to understand the basement and basin architecture, and thus serves as a predictive exploration model. This technique, which was developed for use in the petroleum industry, is now finding success as a predictive tool in mineral exploration.

May 1, 2003

Today?s Topics ? Data to decision value chain ? Practical Considerations ? Example ? Before / After DSS ? Quick Decision Support Demo Practical Considerations ? More data is not necessarily better ? Organization and process silos ? Good vs. ?optimal?, but impractical Solutions

May 1, 2008

By Michael G. Nelson, Amy Richins, Ashley Hodgson, W. Pratt Rogers

"Sociotechnical systems are everywhere in the modern industrial world. Sociotechnical systems describe how humans interact with technology throughout a variety of processes. These systems are prevalent in modern industrial settings, such as operating mines. Safety and health management systems (SHMSs) are complex sociotechnical structures designed with the aim of managing and reducing occupational risk. U.S. mining companies are voluntarily adopting SHMSs to supplement regulatory safety compliance efforts. Like all complex sociotechnical systems there are a wide variety of approaches and adoption philosophies around a SHMS. Many SHMSs suffer from high levels of redundancies, which manifest themselves in diverse ways: duplicate processes, forms, permits, and tracking mechanisms. These redundancies have severe impacts on the overall system and absorb critical culture and leadership “energy”. A recent study initiated by the University of Utah and sponsored by the Alpha Foundation is investigating the effectiveness of SHMSs across 15 different U.S. mining sites. A large amount of data has been captured to help characterize elements of SHMSs and correlate their impacts on safety outcomes. Many variables were assessed in the study. Those of importance to this paper are the variables describing the robustness of the sociotechnical aspects of the SHMS. Two such variables are the overall integration of the SHMS within the whole system and how data is managed. Data management in this context deals with the approach to capturing, analyzing, and acting on data within the system. As well, systems are integrated around controlled, data-sharing mechanisms. It has been found that wide variety of approaches has been taken to integrate the SHMS within the governing management philosophy at a given site. A series of case studies is presented that develop some specific best practices for proper data management and integration of SHMSs.Introduction and Project Background Managing occupational health and safety can be done in many different ways, and with many different approaches, but is often based on the host country’s regulatory and industrial culture. However, many companies have extended their effort beyond the regulatory requirements of their host nation and have implemented Safety and Health Management Systems. The goal of these efforts is to reduce and eliminate occupational risk. Many health and safety practitioners have a hard time justifying various elements of health and safety management. There is a lack of institutional knowledge that measures and assesses the effectiveness of these Safety and Health Management System (SHMS) elements. To better understand the effectiveness of SHMSs the Alpha Foundation has funded a research project with the University of Utah. The title of the project is “Characterization and Effectiveness of Safety and Health Management Systems in the U.S. Mining Industry.” The research team will assess the SHMSs of 15 United States mine sites. The outcome of the project will be a method to characterize the strength of the SHMS elements, assess a culture and leadership index, and finally establish correlations between elemental risk and safety outcomes. The study is currently in progress. However, many important outcomes have already been observed, particularly on the data management practices within an SHMS."

Jan 1, 2017

By Laurel E. Blackman

The DFI Project Information Management Systems (PIMS) Committee cooperated with the US Army Corps of Engineers (USACE) to develop a specification for PIMS used on USACE projects. Once completed the Committee developed a more generic version of this specification for general use, which is contained in this document. This first edition has simply removed the USACE-related requirements from the document. The committee will continue to work on this specification to enhance the content and also incorporate recent technological development for these systems. It is the Committee's intention that a second edition will be issued in 2022.

Jan 1, 2021