Dissertations / Theses on the topic 'Mining engineering'

Normalization of deviance has been thoroughly studied and proven to have a dramatic impact on the medical industry, particularly in the field of anesthesiology, and for the National Aeronautics and Space Administration (NASA). Few such studies have been conducted in the mining industry. This research was designed to show whether normalization of deviance is occurring within the subculture of mining engineers.

This research project was based on a cross-sectional surveillance of a group of mining engineers and consultants belonging to the Society for Mining, Metallurgy and Exploration (SME).

There were three hypotheses for this research: 1) there is a correlation between ethics, compensation, risk tolerance, and normalization of deviance; 2) there are either positive or negative associations between each of the independent variables—ethics, compensation, and risk tolerance—to the dependent variable—normalization of deviance; 3) the data would make it possible to predict normalization of deviance among mining engineers. All three hypotheses were proven true in this study.

This research is important because it shows that normalization of deviance exists among mining engineers.

How do we find patterns and anomalies, on graphs with billions of nodes and edges, which do not fit in memory? How to use parallelism for such Tera- or Peta-scale graphs? In this thesis, we propose PEGASUS, a large scale graph mining system implemented on the top of the HADOOP platform, the open source version of MAPREDUCE. PEGASUS includes algorithms which help us spot patterns and anomalous behaviors in large graphs. PEGASUS enables the structure analysis on large graphs. We unify many different structure analysis algorithms, including the analysis on connected components, PageRank, and radius/diameter, into a general primitive called GIM-V. GIM-V is highly optimized, achieving good scale-up on the number of edges and available machines. We discover surprising patterns using GIM-V, including the 7-degrees of separation in one of the largest publicly available Web graphs, with 7 billion edges. PEGASUS also enables the inference and the spectral analysis on large graphs. We design an efficient distributed belief propagation algorithm which infer the states of unlabeled nodes given a set of labeled nodes. We also develop an eigensolver for computing top k eigenvalues and eigenvectors of the adjacency matrices of very large graphs. We use the eigensolver to discover anomalous adult advertisers in the who-follows-whom Twitter graph with 3 billion edges. In addition, we develop an efficient tensor decomposition algorithm and use it to analyze a large knowledge base tensor. Finally, PEGASUS allows the management of large graphs. We propose efficient graph storage and indexing methods to answer graph mining queries quickly. We also develop an edge layout algorithm for better compressing graphs.

This project describes a novel approach to the development of a multimodal video segmentation system for the analysis of multimedia data. The current practices of multimedia data analysis rely either solely on one of the video and audio components or on the presence of both together. The proposed approach makes use of both the video and audio inputs in parallel, complementing each other during the video processing stage, towards optimising both the accuracy and speed of the method. Unlike in the other commonly established methods, the video analysis here is carried out using both the luminance and the chrominance values of the colour images, instead of relying on either of them. The approach considered in the proposed method of video cut detection primarily uses a modified luminance based histogram analysis algorithm, supported by the additional sub-sampling and median filtering options. They improve the efficiency of the method through enhancing its speed and the accuracy of detection respectively. The algorithm mentioned above uses a progressively varying threshold for indicating a significant variation in the measurement of successive histograms for a window length of 2 image frames. The method worked successfully for the videos with varying rates and sizes of the frames that have been under investigation. Because of the degrading effect of chrominance histogram analysis on the processing speed its use is kept to a minimum. This is restricted only to verify the existence of possible cuts, failed to be identified by the luminance analysis. The indication of such cuts could be obtained through audio classification analysis.

The large-scale wind energy industry is relatively new and is rapidly expanding. The ability of a wind turbine to extract power from the wind is a function of three main factors: the measured wind speed, the power curve of the turbine, and the ability of the machine to handle wind fluctuations. The key parameter determining wind turbine performance is wind speed and it is normally measured with an anemometer placed at the nacelle of a turbine. The dynamic nature of wind speed, however, is a barrier for applying predictive engineering in wind energy. Traditional approaches based on physical science and mathematical modelings have limitations on wind power prediction models. Conventional approach based on dynamic modeling has disadvantage of power generation process modeling due to time-shift nature of the process. Data mining is a promising approach for modeling wind energy, e.g., power prediction and optimization, wind speed forecasting, power curve monitoring and fault diagnosis. It involves a number of steps including data pre-processing, data sampling, feature selection, dimension reduction and, etc. This thesis focus on applying data mining to predictive engineering in wind industry, and ultimately builds wind speed prediction and wind farm power prediction models, develops turbine dynamic control and power optimization strategy, explores methodology for system level fault diagnosis. However the philosophy, methods and frameworks discussed in this research can also be applied to other industrial processes. This thesis proposes a series of predictive models under the framework of data mining. Chapter 2 introduces a methodology for short term wind speed prediction based on wind farm layout information. Chapter 3 and Chapter 4 present prediction models for wind turbine parameters. Chapter 5 proposes strategies for dynamic control of wind turbines. Chapter 6 explores the fault diagnosis and prediction using SCADA data.

Each year, organizations in Australian mining industry (asset intensive industry) spend substantial amount of capital (A$86 billion in 2009-10) (Statistics, 2011) in acquiring engineering assets. Engineering assets are put to use in operations to generate value. Different functions (departments) of an organization have different expectations and requirements from each of the engineering asset e.g. return on investment, reliability, efficiency, maintainability, low cost of running the asset, low or nil environmental impact and easy of disposal, potential salvage value etc. Assets are acquired from suppliers or built by service providers and or internally. The process of acquiring assets is supported by procurement function. One of the most costly mistakes that organizations can make is acquiring the inappropriate or non-conforming assets that do not fit the purpose. The root cause of acquiring non confirming assets belongs to incorrect acquisition decision and the process of making decisions. It is very important that an asset acquisition decision is based on inputs and multi-criteria of each function within the organization which has direct or indirect impact on the acquisition, utilization, maintenance and disposal of the asset. Literature review shows that currently there is no comprehensive process framework and tool available to evaluate the inclusiveness and breadth of asset acquisition decisions that are taken in the Mining Organizations. This thesis discusses various such criteria and inputs that need to be considered and evaluated from various functions within the organization while making the asset acquisition decision. Criteria from functions such as finance, production, maintenance, logistics, procurement, asset management, environment health and safety, material management, training and development etc. need to be considered to make an effective and coherent asset acquisition decision. The thesis also discusses a tool that is developed to be used in the multi-criteria and cross functional acquisition decision making. The development of multi-criteria and cross functional inputs based decision framework and tool which utilizes that framework to formulate cross functional and integrated asset acquisition decisions are the contribution of this research.

Mining automation has incrementally progressed from line-of-sight remote operation to teleoperation and automatic control of mobile machines, mainly due to significant advances in underground communication systems. The present trend points towards a robotic mining environment where mobile machinery and stationary equipment will be fully integrated with a mine-wide information system overseeing all aspects of mining via a communication network. The successful design and implementation of the software and hardware components necessary to realize this vision depends on the level of seamless integration achieved. The complexity involved in terms of systems functionality and coherence necessitates systems analysis and computer-aided software engineering tools to actively support this integration effort.
Hence, the primary objective of this thesis is to introduce and relate systems analysis concepts and tools to the business of mining. This investigation begins by setting the industrial context of this work with respect to past initiatives and future trends. It discusses different approaches to the design and implementation of mining information systems. It reviews the fundamentals of software and information engineering as well as structured and object-oriented analysis and design. It presents a survey of computerized tools for systems analysis. It then applies systems analysis concepts and tools to a high-level top-down analysis of a Mine Information System and examines a specific mining process in detail. Finally, it compares the applicability of structured versus object-oriented analysis and design methodologies to the complex problem of mining.

A P&H 2800XP electric mining shovel working in a Western Canadian surface coal mine was instrumented with microprocessor-based monitoring equipment, and its performance monitored as it excavated a series of test blasts. Hoist and crowd motor armature voltages and currents, dipper trajectories, cycle times and load weights were recorded. A diggability index has been established based on the responses of the hoist motor. Values of the diggability index correlated well with digging conditions as observed during monitoring and with muckpile fragmentation size distributions as determined by a photographic survey. This correlation establishes the ability of an instrumented shovel to diagnose the efficiency of ground preparation practices (blasting) by identifying variations in muckpile diggability. The data on dipper trajectory has demonstrated that variations in digging practices do exist amongst an experienced group of shovel operators, and that variation in trajectory significantly influences values of the recorded motor performance parameters. An approach is described to account for variations in digging practice on assessments of diggability. An allied investigation into the ability of time studies to define diggability, revealed dig cycle times to be operator dependant, and unrelated to levels of digging effort.

Compact and succinct data structures use space that approaches the information-theoretic lower bound on the space that is required to represent the data. In practice, their memory footprint is orders of magnitude smaller than normal data structures and at the same time they are competitive in speed. A main drawback with many of these data structures is that they do not support dynamic operations efficiently. It can be exceedingly expensive to rebuild a static data structure each time an update occurs. In this thesis, we propose a number of novel compact dynamic data structures including m-Bonsai, which is a compact tree representation, compact dynamic rewritable (CDRW) arrays which is a compact representation of variable-length bit-strings. These data structures can answer queries efficiently, perform updates fast while they maintain their small memory footprint. In addition to the designing of these data structures, we analyze them theoretically, we implement them and finally test them to show their good practical performance. Many data mining algorithms require data structures that can query and dynamically update data in memory. One such algorithm is FP-growth. It is one of the fastest algorithms for the solution of Frequent Itemset Mining, which is one of the most fundamental problems in data mining. FP-growth reads the entire data in memory, updates the data structures in memory and performs a series of queries on the given data. We propose a compact implementation for the FP-growth algorithm, the PFP-growth. Based on our experimental evaluation, our implementation is one order of magnitude more space efficient compared to the classic implementation of FP-growth and 2 - 3 times compared to a more recent carefully engineered implementation. At the same time it is competitive in terms of speed.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2006.
Includes bibliographical references (p. 81-89).
Data of massive size are now available in a wide variety of fields and come with great promise. In theory, these massive data sets allow data mining and exploration on a scale previously unimaginable. However, in practice, it can be difficult to apply classic data mining techniques to such massive data sets due to their sheer size. In this thesis, we study three algorithmic problems in data mining with consideration to the analysis of massive data sets. Our work is both theoretical and experimental - we design algorithms and prove guarantees for their performance and also give experimental results on real data sets. The three problems we study are: 1) finding a matrix of low rank that approximates a given matrix, 2) clustering high-dimensional points into subsets whose points lie in the same subspace, and 3) clustering objects by pairwise similarities/distances.
by Grant J. Wang.
Ph.D.

"Industrial restructuring without parallel in recent British industrial history" is how the current Chairman of British Coal, Sir Robert Haslam, has described events in that industry. Since 1960 upwards of three quarters of a million jobs have gone in the deep coal mining industry alone. Numerous studies have analysed the underlying mechanisms behind the rapid decline of the nationalised coal industry, but hitherto little attention has been paid to the national linkage effects of that decline. This thesis is an attempt to analyse the consequences of industrial restructuring in coal mining on its UK engineering suppliers. In so doing, the thesis develops into much more than an empirical case study of industrial linkage and becomes a critical analysis of state capital-private capital relations. In particular, it focusses on the shifting boundaries of state ownership in the energy sector of the 'eighties. It considers what are the main processes involved and some of the consequences for those people and places most dependent on mining related jobs for their livelihoods.

The electric mining shovel is the main rock loading machine at the majority of surface mining operations. The performance and utilization of these multi-million dollar machines can govern an entire mining operation. Despite their significance, they remain largely unexplored in terms of their potential. The thesis discusses the development of the forward and inverse kinematic models, the Newton-Euler dynamics, and the actuator dynamics for an electric mining shovel. The thesis relates to work undertaken towards the development of a mining shovel simulator, the purpose of which is to explore productivity and machine performance issues as a function of the interaction of the shovel with the ground. Both the parametric and nonparametric system identification results are presented, based on field test data collected from an operating mining shovel. Simulator results describing analytical and empirical models, as well as continuous-time controller and disturbance rejection are also presented. The thesis concludes with a discussion of planned continuing work.

Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2006.
Earthquakes and other seismic events are known to have catastrophic effects on people and property. These large-scale events are almost always preceded by smallerscale seismic events called precursors, such as tremors or other vibrations. The use of precursor data to predict the realization of seismic hazards has been a long-standing technical problem in different disciplines. For example, blasting or other mining activities have the potential to induce the collapse of rock surfaces, or the occurrence of other dangerous seismic events in large volumes of rock. In this study, seismic data (T4) obtained from a mining concern in South Africa were considered using a nonlinear time series approach. In particular, the method of surrogate analysis was used to characterize the deterministic structure in the data, prior to fitting a predictive model. The seismic data set (T4) is a set of seismic events for a small volume of rock in a mine observed over a period of 12 days. The surrogate data were generated to have structure similar to that of T4 according to some basic seismic laws. In particular, the surrogate data sets were generated to have the same autocorrelation structure and amplitude distributions of the underlying data set T4. The surrogate data derived from T4 allow for the assessment of some basic hypotheses regarding both types of data sets. The structure in both types of data (i.e. the relationship between the past behavior and the future realization of components) was investigated by means of three test statistics, each of which provided partial information on the structure in the data. The first is the average mutual information between the reconstructed past and futures states of T4. The second is a correlation dimension estimate, Dc which gives an indication of the deterministic structure (predictability) of the reconstructed states of T4. The final statistic is the correlation coefficients which gives an indication of the predictability of the future behavior of T4 based on the past states of T4. The past states of T4 was reconstructed by reducing the dimension of a delay coordinate embedding of the components of T4. The map from past states to future realization of T4 values was estimated using Long Short-Term Recurrent Memory (LSTM) neural networks. The application of LSTM Recurrent Neural Networks on point processes has not been reported before in literature. Comparison of the stochastic surrogate data with the measured structure in the T4 data set showed that the structure in T4 differed significantly from that of the surrogate data sets. However, the relationship between the past states and the future realization of components for both T4 and surrogate data did not appear to be deterministic. The application of LSTM in the modeling of T4 shows that the approach could model point processes at least as well or even better than previously reported applications on time series data.

In order to develop a philosophy and set of methods which would allow mines to be operated more effectively, a study was initiated to determine how time-based methods used in manufacturing could be adapted to mining. The study was done in collaboration with Inco Ltd. A review was made of both lean manufacturing and time-based competition methods. In addition, continuous programs such as Total Quality Management and Total Preventive Maintenance were studied. This study showed that time based principles, cellular mining, and total preventive maintenance had the most promise for adoption in mining. Simulations of three different mining methods under different scenarios and cost analyses were done to demonstrate the viability of these methods. The conclusions were the following. (1) Time based principles lead to greater profitability, and thus, should be used to drive mine planning and production. (2) Cellular mining leads to more efficient mining production. (3) The adoption of a comprehensive Total Preventive Maintenance program is needed to improve equipment availability and to complement cellular mining.

Thesis (M.Eng.)--Massachusetts Institute of Technology, Dept. of Civil and Environmental Engineering, 2003.
Includes bibliographical references (leaves 65-67).
In large decentralized institutions such as MIT, finding information about events and activities on a campus-wide basis can be a strenuous task. This is mainly due to the ephemeral nature of events and the inability to impose a centralized information system to all event organizers and target audiences. For the purpose of advertising events, Email is the communication medium of choice. In particular, there is a wide-spread use of electronic mailing lists to publicize events and activities. These can be used as a valuable source for information mining. This dissertation will propose two mining architectures to find category-specific event announcements broadcasted on public MIT mailing lists. At the center of these mining systems is a text classifier that groups Emails based on their textual content. Classification is followed by information extraction where labeled data, such as the event date, is identified and stored along with the Email content in a searchable database. The first architecture is based on a probabilistic classification method, namely naive-Bayes while the second uses a rules-based classifier. A case implementation, FreeFood@MIT, was implemented to expose the results of these classification schemes and is used as a benchmark for recommendations.
by Mario A. Harik.
M.Eng.

Most of the nitrogen load from a mine originates from nitrogen-containing explosives used in the mining that do not explode perfectly. Some of the chemicals in the enrichment process may also contain nitrogen and these compounds may dissolve in the process water. A certain amount of the dissolved explosive residues and process chemicals will be discharged in runoff and excess water into downstream water bodies, where the resulting increase in nutrients may cause eutrophication. The thesis first presents the Suhanko mine expansion project and its current situation. The literature review that follows then takes up several central themes. It begins with a description of the explosives generally used in rock blasting and their dissolving properties, noting that the water resistance of different explosives varies significantly. Next, the review presents the principles of open pit mining, the reasons of nitrogen emissions in blasting, and methods for reducing emissions. In this connection, the thesis provides an account of the techniques currently used for nitrogen removal in operating mines and the range of possible techniques. One objective of the research was to determine the nitrogen mass balance of the Suhanko mine when it is extended as planned. The estimates presented draw on calculations of the water balance for the project’s environmental impact assessment (EIA) as well as on the values in the literature for the proportion of undetonated explosives in mining. The consumption of explosives at Suhanko estimated in the thesis uses data from a pre-feasibility study of the project and from the EIA report. The calculations of nitrogen loads are based on monitoring data from an operating metal ore mine similar to that planned in Suhanko; for example, the reference mine uses the same types of explosives. The results indicate that most of the explosive residues at Suhanko will migrate to the waste rock stockpiling areas; the second largest proportion will end up in the concentrator plant in the extracted ore and recirculated drainage water; and the third most significant percentage will migrate to the stockpiling areas for low-grade ore. The results of the nitrogen balance calculations show that runoff waters from the waste rock stockpiling areas will contain a maximum of 11 mg N/l. The annual nitrogen load from the different waste rock areas will vary between 0.43 and 8.02 t. Excess mine water will be discharged from the flotation tailings pond into Takalampi Pond, and the annual nitrogen load from this operation under different water balance scenarios will vary from 17.6 to 74.3 t. The experimental part of the research studied the extent to which the peatland areas near the proposed waste rock areas might be suitable for wetland treatment of mine runoff water. The related fieldwork involved measuring in situ hydraulic conductivity and taking peat samples. The samples were analyzed to determine the degree of humification, water content and ash content of the peat, and a chemical characterization was performed as well. The slopes of the peatland areas with potential as treatment wetlands were estimated using laser scanning data from the Suhanko area and their surface areas were calculated based on the aerial photographs. In situ measurements indicated higher hydraulic conductivity in the uppers layer of the peatlands; the values decreased significantly below depths of 50 cm. The degree of humification varied at different depths: in the upper layers, the peat was generally weakly decomposed or undecomposed; in the deeper layers, it was more decomposed. For the most part of peatlands, the slopes are less than 10‰, but in some areas they exceed 25 ‰. The field studies identified peatlands near the Ahmavaara and Konttijärvi South waste rock areas that meet the technical requirements for conversion into treatment wetlands
Suurin osa kaivoksen typpikuormituksesta on yleensä peräisin louhinnassa käytettävistä typpeä sisältävistä räjähdysaineista, jotka eivät räjähdä täydellisesti. Jotkin rikastusprosessissa käytettävät kemikaalit saattavat myös sisältää typpeä, ja typpiyhdisteitä voi liueta rikastusprosessin vesiin. Osa räjähdysainejäämien ja kemikaalien sisältämästä typestä kulkeutuu ympäristöön kaivosalueelta ulos johdettavien vesijakeiden mukana ja ravinteiden määrän kasvaminen voi aiheuttaa rehevöitymistä vastaanottavassa vesistössä. Aluksi tässä työssä esitellään Suhangon kaivoshankkeen laajennusprojekti ja sen nykytilanne. Kirjallisuuskatsauksessa käsitellään louhinnassa yleisesti käytettäviä räjähdysaineita ja niiden liukenemisominaisuuksia ottaen huomioon, että eri räjähdysaineiden vedenkesto-ominaisuudet poikkeavat merkittävästi. Työssä esitellään avolouhintamenetelmän periaatteet ja louhinnan aiheuttamien typpipäästöjen syitä sekä niiden vähentämiskeinoja. Lisäksi työssä käsitellään kaivostoiminnassa nykyisin käytössä olevia typenpoistotekniikoita ja käytettävissä olevia mahdollisia tekniikoita. Tämän työn tavoitteena oli muodostaa typpitase Suhangon kaivoshankkeen laajennukselle. Typpitase perustuu kaivoksen YVA-selostuksessa esitettyihin vesitaselaskelmiin ja kirjallisuuden pohjalta tehtyyn arvioon louhinnassa räjähtämättä jäävän räjähdysaineen määrästä. Räjähdysaineen käyttömäärää Suhangon kaivoksella on arvioitu esikannattavuusselvityksessä esitettyjen louhintasuunnitelmien ja YVA-selostuksessa arvioitujen tuotantomäärien pohjalta. Kuormituslaskelmissa on hyödynnetty louhintamenetelmältään vastaavan metallimalmikaivoksen tarkkailutuloksia, jossa louhinnassa käytettävä räjähdysaine on vastaavantyyppinen kuin mitä Suhangon kaivoksella on suunniteltu käytettävän. Tehtyjen laskelmien perusteella suurin osa räjähdysainejäämistä kulkeutuu sivukivialueille, toiseksi suurin osa rikastamolle sekä malmin että kuivanapitoveden mukana ja osa kulkeutuu myös marginaalimalmin varastointialueelle. Typpitaselaskelmien perusteella sivukivialueilta purettavien aluevesien typpipitoisuus on enimmillään noin 11 mg/l. Aluevesien vuosittainen typpikuormitus eri sivukivialueilta toimintavaiheen aikana vaihtelee 0,43 tonnin ja 8,02 tonnin välillä. Vaahdotuksen rikastushiekka-altaalta typpeä poistuu vesistöön ylitevesien mukana Takalammen suuntaan ja ylitevesien vuosittainen typpikuormitus eri vesitaseen tarkastelutilanteissa vaihtelee 17,6 ja 74,3 tonnin välillä. Työn kokeellisessa osiossa tutkittiin suunniteltujen sivukivialueiden läheisyydessä sijaitsevien turvealueiden soveltuvuutta sivukivialueiden valumavesien kosteikkokäsittelyyn. Maastokäynnin aikana turvealueilta mitattiin in situ hydraulinen johtavuus ja alueilta otettiin turvenäytteitä. Turvenäytteistä määritettiin turpeen maatuneisuus, vesipitoisuus ja tuhkapitoisuus. Potentiaalisten turvealueiden kaltevuuksia arvioitiin laserkeilausaineiston pohjalta ja niiden pinta-alat määritettiin ilmakuvien perusteella. In situ mittausten perusteella hydraulinen johtavuus on korkeampi turpeen yläkerroksissa ja laskee huomattavasti alle 50 cm syvyyksissä. Turpeen maatuneisuus tutkituilla turvealueilla vaihteli eri syvyyksillä. Pintakerroksissa turve oli pääsääntöisesti heikosti maatunutta ja syvemmissä kerroksissa voimakkaammin maatunutta. Laserkeilausaineiston pohjalta määritetyt kaltevuudet osoittavat, että turvealueiden kaltevuudet ovat pääosin alle 1 ‰, mutta joissain osissa turvealueita kaltevuus on yli 25 ‰. Kenttätutkimusten perusteella Ahmavaaran ja Konttijärven eteläisen sivukiven läjitysalueen lähellä on kosteikkokäyttöön soveltuvia turvealueita, koska ne täyttävät kosteikoille annetut tekniset vaatimukset

Overbreak and damage to rock walls is one of the most serious problems encountered in blasting operations. Several techniques have been developed to control the undesirable effects of rock blasting. These techniques are collectively known as wall-control blasting methods.
The stress distribution around pressurized holes has been numerically evaluated, in order to analyze the mechanism of wall-control blasting methods. The effect of blast geometry and the role of discontinuity on this stress field has also been studied in detail. The results obtained by numerical modelling have been verified by controlled blasting experiments, and further supported by analysis of existing roadcuts on a large scale.
It was found that the mechanism of wall-control blast can be explained by the collision and superposition of the stresses between the holes. A narrow fracture zone between the holes was produced by tensile stresses on the centreline. It is neither necessary nor realistic to assume onset of fractures at the midpoint between holes by reinforcement of the stresses from each hole.
The analysis shows that a burden can be defined as being infinite when the ratio of that to the spacing is greater than unity. For pre-split blasting (infinite burden) in an isotropic and homogeneous material the hole separation could range up to 15 borehole diameters. The decoupling ratio between the explosive charge and the borehole diameter should be smaller than 0.5. This ratio would generally be between 0.2 and 0.3 for pre-splitting (infinite burden), and between 0.3 and 0.4 in the presence of a free face.
A discontinuity parallel to the free face and located at the back of the holes causes high stress levels between the discontinuity and the boreholes, resulting is a shattered one in this region. The presence of a similar discontinuity at the front of the holes leads to considerable overbreak and development of an undamaged "hump" of rock between holes. The effect of a discontinuity oriented normal to the centreline at the midpoint between holes has minimal effect on the blast results. As the angle of the discontinuity with the free face decreases from 90$ sp circ,$ the damage zone between the holes and the discontinuity increases, and the shape of the final wall changes from a smooth face to a corrugated shape. A closed-discontinuity or an open discontinuity cemented with strong filling materials has little effect on the results of the blast. However, as the width of the discontinuity increases, the size of the damage zone also increases. An open discontinuity, 50 mm wide or more, plays a role similar to a free face.
In roadcut blast design, hole deviation is a key parameter in determining the quality of the face. However, consistent hole deviation in the same direction has minimal effect in the result of the blast. This type of deviation is usually associated with bedded rocks, with alternating bands of soft and hard rock on the face. The degree of deviation is dependent, amongst other factors, on orientation, thickness, frequency and the position of these bands.

This thesis is based on a case study of the Kiirunavaara sublevel cave (SLC) mine. It focuses on footwall stability and damage development in the mining infrastructure on mine scale. Damage to the infrastructure is mappable for the full height of the footwall by access through decommissioned infrastructure associated with earlier mining stages. Damages range from pure structurally controlled failures (wedge failures) in the upper part of the footwall to fracture growth through intact rock combined with micro‑seismic emissions at the active mining depth. The thesis addresses four distinct research questions; (i) What are the predominant failure mechanisms for the Kiirunavaara footwall? (ii) What is the role of confinement on the damage development in the footwall? (iii) How does the SLC relate to the footwall damage development? (iv) How can infrastructure damage associated to the future mining be estimated using currently available data? Two sets of calibrated numerical models were used to study the damage evolution processes using damage mapping data as the main calibration parameter. Validation of the models was achieved by correlation of model output to micro-seismic locations. The modelling and damage mapping results were used as the basis for the development of a simple prognosis tool for estimating the ultimate extent of infrastructure damage associated to the mining advance for future mining steps. A literature review on slope failure modes, large scale failures in cave mining and failure tracking using micro-seismic locations is included to provide background and definitions. The literature describes principal failure modes as well as mechanism combinations such as structurally controlled failures initiated by deep seated rock mass failures or relaxation. Cases are presented where previously stable structures become destabilised by cave advance and examples where micro‑seismic recordings were used to track deformations and the initiation and growth of newly formed fractures. The Kiirunavaara SLC mine is presented in detail as the main case study of the work. The mine has been in operation since the early 20th century with a transition to underground operation over 50 years ago. The extent of the orebody is 4 km in length with an average width of 80-90 m, the termination at depth has yet to be determined. The ore has an average dip of 60˚ east and a dip-along-strike to the north. Both the footwall and hangingwall rock masses are considered hard and competent with UCS values for the footwall ranging from ca. 130 MPa to extreme cases of 600 MPa. The ore is mined in production blocks about 400 m wide (along strike), Mining of the northernmost blocks, situated in the Lake ore, did not start as open pit operations but has been accessed from the underground via SLC only. The instabilities in the footwall has been addressed by several research studies in the past, with the predominant failure mechanisms in different studies being suggested as large scale tensile failure, complex wedge failure, or rotational shear failure, i.e., some type of principal slope failure. In this work, conceptual numerical models in UDEC were calibrated to fit underground damage mapping data by tracking numerical shear strain concentrations. The conceptual models suggested rock mass damage without the indications of development of large scale slope failure mechanisms such as shear bands. Mine scale PFC models were calibrated with respect to the rock mass strength parameters derived by the conceptual UDEC models and used to study rock mass fracturing in the absence of large scale failure. It is shown that damage to the rock mass occurs mainly close to the active mining in a seismically active zone. This is suggested to weaken and soften the rock mass to allow the development of infrastructure damage in this volume to occur as the rock mass relaxes when entering the stress shadow of the SLC as mining progresses. The damage to the rock mass at the production depth is argued, based on seismic records and a parametric study in UDEC, to constitute of large quantities of local shear failures coalescing to appear as a large scale step-path or rotational shear failure in mapping records. The extent of the associated infrastructure damage is predicated to be limited by the extent of the damaged rock mass zone. A simple bi-linear equation is suggested using ore-width and mining depth as input to estimate the ultimate extent of the damaged zone for each mining stage and thus the limit of later infrastructure damage development. The thesis is concluded with recommendations for future work and potential for continued research.
Denna avhandling baseras på en fallstudie av skivrasgruvan Kiirunavaara. Fokus ligger på liggväggstabilitet och skadeutveckling på gruvans infrastruktur i gruvskala. Infrastrukturskador kan karteras längs med hela liggväggens höjd där tillgång till bergmassan ges via urdrifttagna ortar och ramper drivna i samband med tidigare brytningssteg. Dokumenterade skador varierar med djupet – från strukturstyrda brott i den övre delen av liggväggen till ny sprickbildning genom intakt berg kombinerat med mikro-seismik vid nuvarande brytningsdjup. Fyra distinkta forskningsfrågor avhandlas; (i) Vilka är de dominerande brottsmekanismerna i Kiirunavaaras liggvägg? (ii) Vilken roll spelar inspänning för brottsutvecklingen i liggväggen? (iii) På vilket sätt relateras skivrasbrytningen till brottsutvecklingen? (iv) Hur kan skador på gruvans infrastruktur kopplat till framtida brytningssteg uppskattas med data tillgängliga idag? Två omgångar med kalibrerade numeriska modeller togs fram för att studera skadeutvecklingen i liggväggen med skadekarteringsdata som primär kalibreringsparameter. Validering av modellerna uppnåddes genom att studera samstämmigheten mellan modellresultaten och lokaliseringen av mikro-seismiska händelser. Modellresultaten och skadekarteringsdatabasen användes som grund för att utveckla ett enkelt prognosverktyg för att uppskatta den slutgiltiga utbredningen av infrastrukturskador direkt associerade med gruvbrytningen för framtida brytningssteg. En litteraturstudie av släntbrott, storkskaliga brott i samband med rasbrytning samt brottsövervakning med mikro-seismik är inkluderad som bakgrund och för att definiera terminologier som används genom avhandlingen. Literaturstudien beskriver principiella brottsmekanismer samt kombinationer av mekanismer såsom strukturstyrda brott pådrivna av djupt belägna bergmassebrott eller minskad inspänning. Fallstudier presenteras där tidigare stabila strukturer destabiliseras av rasbrytningens framskridande och exempel där mikro-seismikdata använts för att följa deformationer samt initiering och tillväxt av nya sprickor i intakt berg och bergmassa. Kiirunavaaragruvan presenteras i detalj som den huvudsakliga fallstudien för arbetet. Gruvan har varit aktiv sedan tidigt 1900-tal med övergång till underjordsbrytning för över 50 år sedan. Malmkroppens utbredning är 4 km längs strykningen med en genomsnittlig vidd av 80-90 m, och malmkroppens fortsättning mot djupet är öppen. Malmen har en genomsnittlig stupning av 60 grader öst med en fältstupning mot norr. Bergmassan i både liggvägg och hängvägg anses vara hård och kompetent med UCS värden för liggväggen mellan ca. 130 MPa till extrema fall av 600 MPa. Malmen bryts i produktionsblock med ca 400 m bredd (längs malmens strykning). Brytning av de nordligaste blocken, belägna i Sjömalmen, har inte skett i dagbrott utan har utförts enbart via skivrasbrytning. Instabiliteten i liggväggen har avhandlats i ett flertal tidigare studier. De dominerande brottsmekanismerna har föreslagits i tidigare arbeten som storskaligt dragbrott, komplext kilbrott eller cirkulärt skjuvbrott d.v.s. någon typ av principiellt släntbrott. I arbetet för denna avhandling kalibrerades konceptuella numeriska modeller i UDEC mot skadekarteringsdata från liggvägens underjord, med avseende på koncentrationer av skjuvtöjningar. De konceptuella modellerna visade på bergmasseskador utan indikationer på storskaligt släntbrott, exempelvis koncentrationer av numeriska skjuvband. PFC-modeller i gruvskala kalibrerades gentemot bergmasseparametrarna från de konceptuella studierna i UDEC för att direkt studera upprickningen av bergmassan i frånvaro av storskaliga brottsindikationer. Modellerna visade på att skador i bergmassan främst uppkommer nära brytningsområdet i en seismiskt aktiv zon. Detta föreslås försvaga och mjukgöra bergmassan vilket i sin tur leder till utveklingen av infrastrukturskador i den skadade volymen när berget avlastas då området hamnar i spänningsskugga från skivraset. Ovanstående studier visar att skadorna som uppkommer i bergmassan, baserat på de konceptuella UDEC-modellerna och mikro-seismiska data, består av ett stort antal lokala skjuvbrott vilka samverkar till att framstå som ett storskaligt trappstegsbrott eller cirkulärt skjuvbrott i skadekarteringsdatat. Utbredningen av de relaterade infrastrukturskadorna förutspås begränsas av utbredningen av bergmasseskadorna uppkomna vid bryningen. Ett enkelt bi-linjär samband föreslås vilket använder malmbredd och brytningsdjup för att uppskatta den slutgiltiga utbredningen av skadezonen i bergmassan för varje brytningssteg, och i förlängningen begräsningen av senare uppkommande infrastrukturskador. Avhandlingen avslutas med rekommendationer för fortsatt arbete samt framtida forskningspotential.

Thesis (M.Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2001.
Includes bibliographical references (leaves 68-71).
by Karin Cheung.
M.Eng.

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado.
This article analyzes the applicability of an appropriate design of blasting engineering for an underground mine to be used in a Peruvian grain context through the data collection and observation method using the Bieniawski rock mass geomechanical classification system, average vein power weighting, and operational dilution of the pits. The generation of solutions is under both the McCarthy approach formula and final results analysis of the dilution behavior obtained in relation to the initial ones. Moreover, the database provided by a mining company is analyzed with respect to the geological conditions for the rock mass classification of the roof box and floor box to determine the quality of the rock, ore grades, power of the vein, and the percentage of dilution. Then we proceed to determine the continuity and thickness of the vein in a given block and analyze the dilution based on the fundamental criteria of blasting engineering.

A lot of different techniques can be used to determine the position of objects in underground mines. What they have in common is that they’re very expensive or require a specific infrastructure. In this master’s thesis the goal was to produce an algorithm to increase the positioning on vehicles in a cheap and easy way by only using gyroscopeand dead reckoning. To illustrate the movement of the vehicle in the underground mines the Mobilaris Mining Intelligence system (MMI) was used. The algorithm is designed to be used alongside the current algorithm that locates objects by using their RSSI value from the Cisco Access points in the underground mine. With this information it’s possible to relocate a vehicle if the algorithm ”chose” the wrong way. Testing shows that this is a good way to increase the positioning of vehicles in underground mines.

Isolated high grade and low tonnage uranium ore pods are prevalent in Saskatchewan. This thesis reviews mining such a pod with both conventional and unconventional methods. The geomechanical characteristics of the overburden and the bedrock in the area of the pods must be investigated before evaluating the technical feasibility of possible mining methods. Conventional ore recovery techniques appraised include open pit and underground mining. The unconventional mining techniques considered are blind boring and jet boring from surface. These last two techniques can offer a greater degree of selectivity, although ground reinforcement techniques had to be considered due to the poor quality of the ground. Ground reinforcement techniques considered were shaft lining, grouting, and ground freezing. This mining design challenge was found to be feasible with the combination of jet boring and freezing. Recommendations are presented to define future areas of study for blind boring, jet boring, grouting, and freezing.

This research is a Master’s thesis that has made for Parker Hannifin Oy in the University of Oulu in Department of Process and Environmental engineering. Parker Hannifin Oy ordered a market research of the Finnish mining market. A subject was a limited research to the northern Fennoscandian and Greenland mining market but the main focus is in Finnish mining industry and technology market. The research produced new information about the mining market for the decision makers of Parker Hannifin Oy. The research problem was to clarify which delivery models would work with mining industry, what is the commercial potential with mining industry and which are the technology areas where the supply of Parker Hannifin Oy and the demand of the mining industry would meet. Qualitative and quantitative methods were used in the market research. A sample for market research was collected from public sources and with interviews and questionnaires. The market research begun with desk research in which the researcher investigated the mining market by public information. After this there were a qualitative interviews and questionnaires for the mining industry and for the company which ordered of research. The data sample was treated with qualitative research methods. The commercial potential for Parker Hannifin Oy is huge in Finnish and Swedish mining market. The mining is developing and growing very fast in both countries. It was estimated that Parker Hannifin Oy could be increased their revenue 10 % by sales to Finnish mining industry in 2012. It has been estimated that the production of Finnish mining industry will triple by the year 2022. Also, the Swedish mining industry will need 10 000–15 000 new employees by year 2025 which two or three times more than Finnish mining industry. The potential mining industry customers considered that site container and a spare part store of mining company are efficient and effective delivery models. One result of the questionnaire was that most of the Finnish mining companies have a store of critical spare parts of mining machines. Products of Parker Hannifin Oy are mostly used in underground mining machines in Finland. In the research came out that the easiest way for Parker Hannifin Oy to increase its sales could be by finding cooperation partners from those local stores and contactors which already do business with local mining companies.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Biological Engineering, 2019
Cataloged from PDF version of thesis.
Includes bibliographical references.
The human microbiome is essential for health and has been implicated in many diseases. DNA sequencing has enabled the detailed characterization of these human-associated microbial communities, leading to a rapid expansion in studies investigating the human microbiome. In this thesis, I describe multiple projects which overcome various data analysis challenges to extract useful clinical insights from microbiome data. In the first project, I present an analysis of lung, stomach, and oropharyngeal microbiomes. I leverage data collected from multiple sites per patient to identify aspiration-associated changes in the relationships between these communities, discovering new properties of the aerodigestive microbiome and suggesting new approaches for treatment. In the second project, I perform a meta-analysis of case-control gut microbiome datasets with standard data processing and analysis methods.
I find consistent patterns characterizing disease-associated microbiome changes and a set of shared associations which could inform clinical treatment and therapeutic development approaches for different microbiome-mediated diseases. Enabled by this work, in the third project I contribute to the development of a method to correct for batch effects in case-control microbiome studies. In the fourth project, I describe a framework for rational donor selection in fecal microbiota transplant clinical trials in which knowledge derived from clinical and basic science research is used to inform which donor is selected for fecal transplants, increasing the likelihood of successful trials. Finally, I present preliminary results analyzing the microbiome and metabolome of residential sewage as a novel platform for community-level public health surveillance.
Together, these projects demonstrate a variety of approaches to mine the human microbiome for clinically-relevant insights and suggests multiple avenues forward for translating findings from microbiome data analyses into clinical and public health impact.
by Claire Marie Noëlle Duvallet.
Ph. D.
Ph.D. Massachusetts Institute of Technology, Department of Biological Engineering

The development of decision support systems for equipment diagnosis has been found to be an iterative process whereby functionality and knowledge are continually added to a prototype until satisfactory performance is achieved. In order to reduce both the dependency on compiled knowledge sources and the number of prototype stages necessary to develop diagnostic decision support systems, this thesis examines, adapts and applies a set theoretical approach to mechanism diagnosis first developed in the field of Artificial Intelligence. The approach does not require the development of computational models to simulate equipment behaviour.
The set theoretical approach was applied to the development of a diagnostic decision support system for a semi-automated Atlas Copco Wagner ST-8B Load-Haul-Dump vehicle. Hypothesis sets were generated for the vehicle's hydraulic circuit and Deutz FL-413-FW diesel engine. A high level of diagnostic resolution was achieved for the hydraulic circuit, but limited resolution was achieved for the diesel engine. This was postulated to be due to the ratio of observable system outputs to input sub-systems, and the number of least repairable units making up each system.
Manual knowledge acquisition was undertaken in an underground mine to refine the diagnostic knowledge developed from the hypothesis sets and to add knowledge to discriminate between competing failure hypotheses. Heuristic failure likelihoods were used to rank hypotheses in order of frequency of occurrence. The knowledge base was implemented as a hypertext decision support system using HyperText Mark-up Language (HTML). The resulting decision support system is platform independent, upgradeable and able to be maintained by site personnel. The system is currently installed at surface level and at 1800 level at INCO Limited's Stobie Mine in Sudbury, Ontario.
The thesis makes a number of original contributions, the first two of which are of generic significance. It is the first work to apply set theoretical concepts to structural models of mobile mining equipment in order to diagnose faults. A number of modifications are advanced to the conventional trace-back analysis technique for generating contributor and normality sets, and heuristic guidelines are provided for estimating the costs and benefits of developing, implementing and maintaining diagnostic decision support systems. It is also the first work to formalise a decision support system in HTML and to suggest the application of company-wide internets ("intranets") to disseminate maintenance knowledge within mines.

Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Materials Science and Engineering, 2004.
Includes bibliographical references (p. 41-42).
Determining the stable structure types of an alloy is critical to determining many properties of that material. This can be done through experiment or computation. Both methods can be expensive and time consuming. Computational methods require energy calculations of hundreds of structure types. Computation time would be greatly improved if this large number of possible structure types was reduced. A method is discussed here to predict the stable structure types for an alloy based on compiled data. This would include experimentally observed stable structure types and calculated energies of structure types. In this paper I will describe the state of this technology. This will include an overview of past and current work. Curtarolo et al. showed a factor of three improvement in the number of calculations required to determine a given percentage of the ground state structure types for an alloy system by using correlations among a database of over 6000 calculated energies.I will show correlations among experimentally determined stable structure types appearing in the same alloy system through statistics computed from the Pauling File Inorganic Materials Database Binaries edition. I will compare a method to predict stable structure types based on correlations among pairs of structure types that appear in the same alloy system with a method based simply on the frequency of occurrence of each structure type. I will show a factor of two improvement in the number of calculations required to determine the ground state structure types between these two methods. This paper will examine the potential market value for a software tool used to predict likely stable structure types. A timeline for introduction of this product and an analysis of the market for such a tool will be included. There is no established market for structure type prediction software, but the market will be similar to that of materials database software and energy calculation software.The potential market is small, but the production and maintenance costs are also small. These small costs, combined with the potential of this tool to improve greatly over time, make this a potentially promising investment. These methods are still in development. The key to the value of this tool lies in the accuracy of the prediction methods developed over the next few years.
by Kevin Tibbetts.
M.Eng.

Thesis: S.M. in Transportation, Massachusetts Institute of Technology, Department of Civil and Environmental Engineering, 2016.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 64-67).
The account receivable is one of the main challenges in the business operation. With poor management of invoice to cash collection process, the over due invoice may pile up, and the increasing amount of unpaid invoice may lead to cash flow problems. In this thesis, I addressed the proactive approach to improving account receivable management using predictive modeling. To complete the task, I built supervised learning models to identity the delayed invoices in advance and made recommendations on improving performance of order to cash collection process. The main procedures of the research work are data cleaning and processing, statistical analysis, building machine learning models and evaluating model performance. The analytical and modeling of the study are based on the real-world invoice data from a Fortune 500 company. The thesis also discussed approaches of dealing with imbalanced data, which includes sampling techniques, performance measurements and ensemble algorithms. The invoice data used in this thesis is imbalanced, because on-time invoice and delayed invoice classes are not approximately equally represented. The cost sensitivity learning techniques demonstrates favorable improvement on classification results. The results of the thesis reveal that the supervised machine learning models can predict the potential late payment of invoice with high accuracy.
by Weikun Hu.
S.M. in Transportation

Thousands of employees in small and medium-sized enterprises (SMEs) in Indonesia’s tin mining industry are working in dangerous environments without an adequate safety and health protection programme. Occupational safety (OS) has become a primary focal point when it comes to preventing dangerous incidents. Such incidents can lead to serious injuries and fatalities, particularly given that mining involves dangerous and high-risk activities, for example, in onshore tin mining, working in muddy areas with heat, humidity and a risk of landslide. Studies addressing the health and safety domain in Indonesia and the mining sector have the least number of ergonomics applications. The research examined the existing OS of a state-owned tin company, namely PT Timah, in Indonesia. The company has a relatively good level of experience in practising tin mining operations, and also employs occupational safety programmes. For approximately a decade, this company has collaborated with 16 private onshore SMEs which work in more than 400 locations under the PT Timah cooperation scheme. Despite applying OS and a health programme in the main company, the same programme has not been included in the tin mining operation cooperation scheme that regulates the rights and obligations of SMEs. This thesis examines the possibilities of using a well-established ergonomics approach and adapting it so that it is more frequently accepted and embraced in Indonesia. The research is focused on the participatory ergonomics (PE) approach as applied to SMEs. Postal data reviews, interviews and observation methods were used to examine the current status of OS in the Indonesian tin mining industry. The scenario-based design (SBD) method was used to investigate the acceptability of PE, while the two-round Delphi technique was applied to reach a consensus on the practical implementations of PE solutions. There is positive acceptance of PE among workers in Indonesian tin mining companies. Regular working safety dissemination through a kinship approach and specific safety training is recommended for resolving communication, cultural and knowledge issues related to OS. In addition to this, a consensus was reached on proposed solutions regarding the implementation of PE within the Indonesian tin mining industry, such as evaluation of the remuneration and bonus system, employing the train-the-trainer programme, and the OSH reporting system.

This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2019
Cataloged from student-submitted PDF version of thesis.
Includes bibliographical references.
Discovering, designing, and developing a novel material is an arduous task, involving countless hours of human effort and ingenuity. While some aspects of this process have been vastly accelerated by the advent of first-principles-based computational techniques and high throughput experimental methods, a vast ocean of untapped historical knowledge lies dormant in the scientific literature. Namely, the precise methods by which many inorganic compounds are synthesized are recorded only as text within journal articles. This thesis aims to realize the potential of this data for informing the syntheses of inorganic materials through the use of data-mining algorithms. Critically, the methods used and produced in this thesis are fully automated, thus maximizing the impact for accelerated synthesis planning by human researchers.
There are three primary objectives of this thesis: 1) aggregate and codify synthesis knowledge contained within scientific literature, 2) identify synthesis "driving factors" for different synthesis outcomes (e.g., phase selection) and 3) autonomously learn synthesis hypotheses from the literature and extend these hypotheses to predicted syntheses for novel materials. Towards the first goal of this thesis, a pipeline of algorithms is developed in order to extract and codify materials synthesis information from journal articles into a structured, machine readable format, analogous to existing databases for materials structures and properties. To efficiently guide the extraction of materials data, this pipeline leverages domain knowledge regarding the allowable relations between different types of information (e.g., concentrations often correspond to solutions).
Both unsupervised and supervised machine learning algorithms are also used to rapidly extract synthesis information from the literature. To examine the autonomous learning of driving factors for morphology selection during hydrothermal syntheses, TiO₂ nanotube formation is found to be correlated with NaOH concentrations and reaction temperatures, using models that are given no internal chemistry knowledge. Additionally, the capacity for transfer learning is shown by predicting phase symmetry in materials systems unseen by models during training, outperforming heuristic physically-motivated baseline stratgies, and again with chemistry-agnostic models. These results suggest that synthesis parameters possess some intrinsic capability for predicting synthesis outcomes. The nature of this linkage between synthesis parameters and synthesis outcomes is then further explored by performing virtual synthesis parameter screening using generative models.
Deep neural networks (variational autoencoders) are trained to learn low-dimensional representations of synthesis routes on augmented datasets, created by aggregated synthesis information across materials with high structural similarity. This technique is validated by predicting ion-mediated polymorph selection effects in MnO₂, using only data from the literature (i.e., without knowledge of competing free energies). This method of synthesis parameter screening is then applied to suggest a new hypothesis for solvent-driven formation of the rare TiO₂ phase, brookite. To extend the capability of synthesis planning with literature-based generative models, a sequence-based conditional variational autoencoder (CVAE) neural network is developed. The CVAE allows a materials scientist to query the model for synthesis suggestions of arbitrary materials, including those that the model has not observed before.
In a demonstrative experiment, the CVAE suggests the correct precursors for literature-reported syntheses of two perovskite materials using training data published more than a decade prior to the target syntheses. Thus, the CVAE is used as an additional materials synthesis screening utility that is complementary to techniques driven by density functional theory calculations. Finally, this thesis provides a broad commentary on the status quo for the reporting of written materials synthesis methods, and suggests a new format which improves both human and machine readability. The thesis concludes with comments on promising future directions which may build upon the work described in this document.
by Edward Soo Kim.
Ph. D.
Ph.D. Massachusetts Institute of Technology, Department of Materials Science and Engineering

Recirculation of air is a widely used technique in general ventilation and air conditioning practise.
Interest in the application of controlled recirculation of ventilation air in mines has grown considerably in the last decade. This is evident by the number of controlled recirculation schemes which have been implemented or are being considered or planned in several countries and in different types of mines.
A detailed description is given of the operating and proposed schemes in South African gold and U.K. coal mines. This includes safety, dust filtration and blast contaminant removal systems.
The monitoring and feasibility methodology required to evaluate controlled recirculation for heat reclamation, and thus cost saving, in Canadian hard rock mines, is then examined.
Experimental work pertaining to air quality study, carried out at three mines, is described in detail, and backed up by extensive appendices.
The observed deposition of dust and reduction in gaseous pollutants concentration in exhaust shafts are analyzed and depicted in several figures and tables.
Simplified theoretical aspects of dust deposition, and statistical analysis of the obtained results, are outlined.

Situation awareness is a crucial capability of any autonomous system, including mining vehicles such as drill rigs and mine trucks. Typically situation awareness is interpreted as the capability of an autonomous system to interpret its surroundings and the intentions of other agents. The internal system awareness however, is often not receiving the same focus, even though the success of any given mission is completely dependent of the condition of the agents themselves. The internal system awareness in the form of vehicle health is the focus of this thesis. As the mining industry becomes increasingly automated, and vehicles become increasingly advanced, the need for condition monitoring and prognostics will continue to rise. This thesis explores data-driven methods that estimate the health of mining vehicles to accommodate those needs. We do so by utilizing available sensor signals, common on a large amount of mining vehicles, to make assessments of the current vehicle condition and tasks. The mining industry is characterized by small series of highly specialized vehicles, which affects the possibility to use more traditional prognostic solutions. The resulting health information can be used both to aid in tasks such as maintenance planning, but also as an important input to decision making for the planning system, i.e. how to run the vehicle for minimum wear and damage, while maintaining other mission objectives. The contributions include: a) A method to use operational data to estimate damage on the frame of a mine truck. This is done using system identification to find a model describing stresses in the structure with input from other sensors such as accelerometers, load sensors and pressure sensors. The estimated stress time signal is in turn used to calculate accumulated damage, and is shown to reveal interesting conclusions on driver behavior. b) A method to characterize the different driving tasks by using an accelerometer and a convolutional neural network. We show that the model is capable of classifying the vehicle task correctly in 96 % of the cases. And finally c), a system for underground road monitoring, where a quarter car model and a Kalman filter are used to generate an estimate of the road profile, while positioning the vehicle using inertial measurements and access point signal strength.

Ytterligare forskningsfinansiär: Epiroc Rock Drills AB

Opencast coal mining using mechanical excavators has taken place in Northern England for over sixty years. In the early years the excavations for coal were relatively shallow and of limited area, typically less than 20 m deep and 50 ha in plan. Nowadays with the deployment of very large draglines and hydraulic shovels, opencast mines can be over 200 m deep and up to 1,000 ha in area. The investigations, excavations and earthworks failures associated with this activity have provided a unique opportunity to study several geotechnical engineering aspects of the drift and solid geology of Northern England, and how they impact on the mine planning, design and operations processes.

Over the last decade, the state-of-the-art in text mining has moved towards the adoption of machine learning as the main paradigm at the heart of approaches. Despite significant advances, machine learning based text mining solutions remain costly to design, develop and maintain for real world problems. An important component of such cost (feature engineering) concerns the effort required to understand which features or characteristics of the data can be successfully exploited in inducing a predictive model of the data. Another important component of the cost (knowledge engineering) has to do with the effort in creating labelled data, and in eliciting knowledge about the mining systems and the data itself. I present a series of approaches, methods and findings aimed at reducing the cost of creating and maintaining document classification and information extraction systems. They address the following questions: Which classes of features lead to an improved classification accuracy in the document classification and entity extraction tasks? How to reduce the amount of labelled examples needed to train machine learning based document classification and information extraction systems, so as to relieve domain experts from this costly task? How to effectively represent knowledge about these systems and the data that they manipulate, in order to make systems interoperable and results replicable? I provide the reader with the background information necessary to understand the above questions and the contributions to the state-of the- art contained herein. The contributions include: the identification of novel classes of features for the document classification task which exploit the multimedia nature of documents and lead to improved classification accuracy; a novel approach to domain adaptation for text categorization which outperforms standard supervised and semi-supervised methods while requiring considerably less supervision; and a well-founded formalism for declaratively specifying text and multimedia mining systems.

The mining industry is a very important sector of the South African national economy. A major factor threatening the sustainability of this industry is the worrying effect of mining accidents. These accidents usually lead to the destruction of property, injury/death of mine workers, and pollution of the environment. Although mining is generally seen as a hazardous operation worldwide, the accident rates in South African mines are still unacceptably high. Another worrying phenomenon is the fact that since 2003 reduction in fatalities and injuries has been 20– 25% short of annual targets set by stake holders. These factors make the safety of the industry a very important subject. The understanding of accident causality is a major step in the quest to reduce accidents. It is only with a good understanding of the accident process that effective remedies can be designed. Accident modelling techniques provide the necessary platform for the interpretation and understanding of accidents at workplaces. The Swiss Cheese Model of accidents has proven to be a very efficient way of analysing industrial accidents. In this model, an accident is seen as a combination of unsafe acts by front line operators and latent conditions in the organization. The model helps to identify factors in an organizational structure that influence human behaviour/performance at workplaces. This study is aimed at demonstrating how a systemic approach can be applied to the analysis of the causes of accidents in South African mines. In this study, an accident analysis framework has been developed from the Swiss Cheese Model, combining the Mark III version of the Swiss Cheese Model, the Nertney Wheel and safety management principles. The main section of the framework is made up of three layers of accident causality: proximal causes, workplace factors and systemic factors. The second section (metadata) of the framework incorporates contextual data pertaining to each accident such as age, experience, task being performed, and time of accident. These data enhance the understanding of accident causality. The third and final section of the framework incorporates information about accident causing agencies and the nature of barriers breached in the accident process.

This project investigated the possibility of tailoring the crystallinity of electrospun fibres (crystallinity studies), and the electrospinning of bi- and multicomponent scaffolds of PLLA. During the crystallinity studies, the effects of various electrospinning process parameters on the crystallinity of electrospun poly(L-lactic acid) (PLLA) fibres were investigated. It was observed that the electrospun fibres had crystallinities between 23 and 46% while that for the as-received granule was 37%, suggesting that the crystallinity of electrospun fibres can be controlled by optimizing the electrospinning process. These results showed that the degree of crystallinity of the electrospun fibres decreased with increasing the polymer solution concentration. Furthermore, an optimum electrospinning voltage at which the maximum degree of crystallinity can be obtained was observed. Finally, bi-component scaffolds based on PLLA and gelatin were electrospun. Multicomponent scaffolds based on PLLA, gelatin and hydroxyapatite (HA) were electrospun followed by electrospraying of the HA phase. Blending gelatin with PLLA resulted in an approximate 50% decrease in fibre diameter. Biocompatibility studies revealed that all scaffolds permitted cell attachment with best results observed on the PLLAGel-HA scaffolds. This was attributed to the exposed HA particles on the surface of the PLLAGel-HA scaffolds which promoted better binding with integrins for osteoconductivity.

碩士
國立高雄第一科技大學
資訊管理所
95
The general principle of IC maker to arise marketing rate and revenue is to lead in advance manufacture technical and help customers to release new product to the market. Products that begin from design to mass production stage , the major problem of back-end testing is how to feedback the analysis result to FAB and enhance product yield. In back-end testing process, the engineering data and analysis performance will affect the speed of FAB to enhance new products’ yield and these kinds of problems always be interesting topics to IC makers. In order to fast response to FAB and resolve product problems or enhance yield , the IC makers always build up engineering data management platform or develop SPC monitor systems that can help engineers to analysis and watching the yield variants. Engineers can collect data via automatic equipments for new products and monitor its or analysis defect patterns. Although, collected mass data could help engineers to analysis the problems of new products but engineers did not have a simply tool to discover the potential risks, especially for unknown domains. This research is using one of data mining methods – SVM (Support Vector Machine) and try to build up the forecast and classification models to analysis 2 production cases of IC makers. Case one is relative to wafer acceptance yield and case two is relative to interface cards of equipments. And demonstrate the analysis results and hope it can help engineers to resolve problems quickly and find out the hiding knowledge.

碩士
義守大學
資訊工程學系
89
In this thesis, we present a systematic approach which integrates data mining techniques to the processes of database reverse engineering. Three new set-oriented (SQL language based) validation techniques and two heuristic searching methods of inclusion dependencies for static and dynamic databases are proposed. In addition, an efficient data mining algorithm to incrementally discover all functional dependencies from a relational database, in the presence of a new set of tuples added to the database, is proposed. The purpose of database reverse engineering is to understand the data semantics of legacy databases in order to facilitate the maintenance of old database and redesign of new databases. Most of the proposed researches assume that inclusion dependencies and functional dependencies had been known. It is hoped that under minimum pre-requisites, a systematic approach combining functions of database management systems, and data mining to facilitate the data structure extraction in database reverse engineering will be proposed. The discovery of inclusion dependencies consists of two basic tasks: validation and searching. Many set and non-set oriented data mining techniques of inclusion dependencies for static database relations have been proposed. In this work, we propose three new set-oriented (SQL Language) validation techniques and two heuristic searching methods of inclusion dependencies for static and dynamic databases. Computational complexities of the new validation techniques are analyzed and compared with Bell & Blockhausen’s [1] method. Numerical simulations of the heuristic searching methods are performed. The results show good computational efficiencies at the time complexity and numbers of query. In addition, we present an efficient data mining algorithm to incrementally discover all FD’s in the presence of a new set of tuples added to an old database. Based on the concept of SQL validation [1] and the monotonicity of FD’s [4], we avoid re-scanning of the database and thereby reduce computation time. The average numbers of SQL queries required by the proposed algorithm and Bell & Blockhausen’s [1] algorithm are estimated. The result shows that the proposed algorithm requires less SQL validation queries.