The three most common rock mass classification systems in use in the South African mining industry today are Bieniawski’s (1976) Geomechanics or RMR System, Barton et al.’s (1974) Q-System and Laubscher’s (1990) MRMR System respectively. Of these three systems, only the MRMR Classification System was developed specifically for mining applications, namely caving operations. In response to the increased use of the MRMR Classification System in the mining industry, and concerns that the MRMR System does not adequately address the role played by discontinuities, veins and cemented joints in a jointed rock mass, Laubscher and Jakubec introduced the In-Situ Rock Mass Rating System (IRMR) in the year 2000. A quantitative comparison of the MRMR and IRMR Classification Systems has been undertaken to determine a correlation between the two classification systems, the results of which indicate that there is not a major difference between the resultant rock mass rating values derived from the two Classification Systems. Therefore, although the IRMR System is more applicable to a jointed rock mass than the MRMR System, the MRMR System should not be regarded as redundant, as it still has a role to play as a mine design tool.

Sweden's population is increasing every year, which means that cities must adapt their infrastructure to be able to follow the demographic trend. In the current situation, Stockholm plans a major expansion of its metro network where evaluating the bedrock quality is of paramount importance. For the new SL metro connection project Gula linjen that stretches between Odenplan and Arenastaden in Solna, the company WSP has previously done quality surveys of the bedrock in the area and dimensioning of the reinforcements in the planned tunnels. Previous survey of the lineaments and weakness zones in the area has been done by means of core drilling and observations in the field. This study examines if the lineaments in the area are associated to any structures below ground surface in the bedrock by examining rock samples from drill cores. A digital presentation tool called Astrock hyperdata report is evaluated in this study by comparing manual measurements of the orientation of the joints in drill cores and the software´s measurements. The studied bedrock in the drill cores varies in quality from very poor to very good. Further examinations of zones of particularly poor rock quality done in this study suggest that the two identified lineaments have different characteristics. The north-west/south-east lineament which runs parallel to the planned metro route is indicated to represent the surface trace of a water-bearing fracture zone in the bedrock. The lineament with the E/W direction is dominated by structures and fault rocks, such as fault gouge and crushed rock in the drill cores, which indicates a brittle deformation zone. For the Gula linjen project, this information becomes useful as the underground reinforcements such as bolting and grouting have to / Sveriges befolkning ökar varje år, vilket gör att städer måste anpassas och expandera för att kunna följa den demografiska utvecklingen. I Stockholm planeras i dagsläget en stor utbyggnad av dess tunnelbanenätverk där det sker mycket arbete kring utvärdering av berggrunden för att göra detta möjligt. För Stockholms lokaltrafiks (SL) tunnelbaneprojektet Gula linjen som ska gå mellan Odenplan och Solna har företaget WSP tidigare gjort kvalitetsundersökningar av berggrunden i området samt dimensioneringar av förstärkningar i de planerade tunnlarna. Studier av svaghetzoner och lineamenten i området har gjorts med hjälp av kärnborrning och fältobservationer. Denna studie undersöker om lineamenten som finns i området är kopplade till strukturer längre ner i berggrunden genom att undersöka bergprov i form av borrkärnor. Som komplement har det digitala redovisningsverktyget Astrock hyperdata report använts för att kunna jämföra strukturer i berget med de egna mätningarna i studien. Programmet kommer även att utvärderas i studien. De erhållna värdena från undersökningen på borrkärnorna varierar i kvalitet från väldigt dålig till väldigt bra. Närmare studier på zoner av särskilt dålig kvalitet tyder på att de båda lineamenten karaktäriseras av olika typer av strukturer. Det nordväst-/sydöstliga lineamentet, som går parallellt med den planerade tunnelbanesträckan, domineras av strukturer som tyder på att lineamentet representerar en vattenförande sprickzon under markytan. I lineamentet med öst-västlig riktning dominerar strukturer som tyder på förekomsten av en förkastningszon. För projektet Gula linjen blir denna information användbar då den vattenförande sprickzonen kan innebära att tunnelförstärkning med injektering måste göras för ytterligare skydd mot vatteninträngning.

A caracterização geotécnica de maciços rochosos é um dos aspectos mais importantes para o sucesso de um empreendimento mineiro. Desta forma é preciso ter um investimento significativo para a aquisição de dados durante a fase de pesquisa mineral, bem como um grande esforço para manipular os dados obtidos. Para se determinar as propriedades de um maciço, é necessário estabelecer, primeiramente, as diferenças entre rochas intacta e maciço rochoso. O comportamento mecânico destas duas classes pode apresentar grandes diferenças quando analisadas em laboratório. Um dos métodos mais utilizados no ambiente mineiro para classificar o maciço rochoso é o Rock Mass Rating (RMR), sistema que consiste em ranquear os seguintes parâmetros: Rock Quality Designation (RQD), espaçamento entre fraturas, resistência a compressão simples (UCS), qualidade das descontinuidades e presença de água no maciço. Este trabalho tem por utilizar a geoestatística para primeiro se conhecer algumas das particularidades destas variáveis, tais como a não linearidade, o comportamento em diferentes direções e a união de diferentes populações e posteriormente usar as técnicas de krigagem para criar um modelo geotécnico tridimensional. A geoestatística nos permite obter uma estimativa mais precisa e ter um maior conhecimento da incerteza do fenômeno, entretanto o RMR, por não ser uma variável aditiva, não se pode aplicar diretamente o uso das técnicas de krigagem. Assim uma metodologia é proposta para se estimar cada um dos parâmetros, ranqueá-los de acordo com seus valores e por fim obter a classificação RMR através do somatório destes ranques. Os resultados são comparados com a geologia estrutural da região de estudo, para se analisar a relação desta com a qualidade do maciço. A metodologia proposta apresentou uma melhora significante na qualidade de informações no local estimado e mostrou ser uma tentativa consistente para se criar um modelo geotécnico 3D. / The geotechnical characterization of rock masses is one of the most important aspects for the success of a mining enterprise. In this way, it is necessary to have a significant investment to acquire data during the mineral research phase and to make a great effort to manipulate and interpret the obtained data. In order to know a rock, it is necessary to establish the differences of properties between rock intact and rocky mass. The mechanical behavior of these two classes can present great differences when analyzed in the laboratory. One of the methods most used in the mining environment to classify the rock mass is the Rock Mass Rating (RMR) system, which consists of ranking the following parameters: Rock Quality Designation (RQD), fracture spacing, simple compression strength (UCS) quality of the discontinuities and presence of water in the massif. This work has to use geostatistics to first know some of the peculiarities of these variables, such as nonlinearity, behavior in different directions and the union of different populations and later to use kriging techniques to create a three - dimensional geotechnical model. The geostatistics allows us to obtain a more accurate estimate and to have a better knowledge of the uncertainty of the phenomenon, however, the RMR is not an additive variable, so cannot be directly applied the use of kriging techniques. Thus a methodology is proposed to estimate each of the parameters, rank them according to their values and finally obtain the RMR classification through the sum of these ranks. The results are compared with the structural geology of the study region, in order to analyze its relation with the quality of the rock mass. The proposed methodology presented a significant improvement in the quality of information at the estimated site and proved to be a consistent attempt to create a 3D geotechnical model.

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 paper illustrates how the Bieniawski rock mass rating geomechanics classification, within the support system used by medium-scale mining companies, allows for the development of a new anchor rock bolt selection method aimed at improving horizontal access stability in underground mines. However, this case study only seeks to select the most efficient anchor rock bolt for any given horizontal access. A proper support selection method is very important for mining companies because this decision will safeguard both miners and mining infrastructure. This selection process will ultimately prevent fatal accidents, which is critical for mining companies today, especially considering the constant operation standstills reported in Peru.

Una caracterización precisa y el modelamiento de la heterogeneidad geomecánica del macizo rocoso conducen a una planificación y diseño de mina eficientes. El uso de técnicas convencionales como el Kriging para el modelamiento del macizo rocoso es limitado, pues no considera la variabilidad espacial y la heterogeneidad del macizo rocoso, dando como resultado estimaciones que no representan el comportamiento real del macizo rocoso. En este contexto, se propone como alternativa de solución usar la Simulación Gaussiana para estimar la heterogeneidad espacial del macizo rocoso basado en el análisis de los valores de UCS, RQD, Condición de aguas y condición de juntas para el modelado del RMR, esta técnica consiste en simular diferentes valores a partir de datos conocidos, además permite analizar la incertidumbre de las simulaciones obtenidas. La metodología que se propone en esta investigación considera el análisis variográfico de las variables regionalizadas geomecánicas en diferentes direcciones, con el fin de determinar su comportamiento anisotrópico; asimismo, se desarrolla el análisis de la incertidumbre mediante la técnica de la validación cruzada que consiste en dividir los datos originales en dos subconjuntos al azar. El 85% de la base de datos se usó para estimar los valores de RMR, mientras que el 15% se utilizó como subconjunto de prueba. El caso de estudio es el tramo crítico con presencia de aguas hidrotermales del túnel de acceso principal de un proyecto minero al sur de Perú, se utilizaron los datos de registro geomecánico tomados en campo, con ellos se produjeron un total de 5 simulaciones por cada variable, produciéndose aproximadamente 3 millones de valores por cada variable. El error medio absoluto del modelo generado con Simulación Gaussiana es de solo 6.58%, que es considerado admisible comparado con el 38.01% obtenido con Kriging. / Accurate characterization and modeling of the rock mass geomechanical heterogeneity leads to efficient mine planning and design. The use of conventional techniques such as Kriging for modeling the rock mass is limited, since it does not consider the spatial variability and heterogeneity of the rock mass, resulting in estimates that do not represent the real behavior of the rock mass. In this context, it is proposed as an alternative solution to use the Gaussian Simulation to estimate the spatial heterogeneity of the rock mass based on the analysis of the values of UCS, RQD, Water condition and joint condition for the RMR modeling, this technique consists in simulating different values from known data, in addition to allowing to analyze the uncertainty of the obtained simulations. The methodology proposed in this research considers the variographic analysis of the regionalized geomechanical variables in directions, in order to determine their anisotropic behavior; Likewise, the uncertainty analysis is developed using the cross-validation technique that consists of dividing the original data into two subsets at random. 85% of the database was used to estimate the RMR values, while 15% was used as a test subset. The case study is the critical section with the presence of hydrothermal waters of the main access tunnel of a mining project in southern Peru, the geomechanical record data taken in the field was used, with them a total of 5 simulations were produced for each variable, producing approximately 3 million values for each variable. The mean absolute error of the model generated with Gaussian Simulation is only 6.58%, which is considered admissible compared to the 38.01% obtained with Kriging. / Tesis