Diluição Operacional em Lavra A Céu Aberto
Diluição Operacional em Lavra A Céu Aberto
Diluição Operacional em Lavra A Céu Aberto
ESCOLA DE ENGENHARIA
PROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA DE MINAS, METALÚRGICA
E DE MATERIAIS – PPGE3M
Porto Alegre
2020
TAÍS RENATA CÂMARA
Porto Alegre
2020
TAÍS RENATA CÂMARA
Aprovado em:
__________________________________________________________________
Prof. Dr. Rodrigo de Lemos Peroni
Orientador
__________________________________________________________________
Prof. Dr. João Felipe Coimbra Leite Costa
DEMIN - UFRGS
__________________________________________________________________
Dr. Luciano Nunes Capponi
Mosaic Fertilizantes
__________________________________________________________________
Prof. Dr. Felipe Ribeiro de Souza
Universidade Federal de Ouro Preto - UFOP
À minha família
AGRADECIMENTOS
A minha tese de doutorado não poderia ser finalizada sem o apoio e incentivo de muitas pessoas.
Foi uma tarefa árdua e de muita dedicação, porém de uma imensa satisfação ao ser concluída.
Agradeço à Universidade Federal do Rio Grande do Sul pelo ensino de qualidade na graduação,
mestrado e doutorado. Agradeço à Capes – Coordenação de Aperfeiçoamento de Pessoal de
Nível Superior – pela bolsa concedida em parte do período que fui aluna dedicação exclusiva
do PPGE3M-UFRGS. Agradeço também ao LPM – Laboratório de Pesquisa Mineral e
Planejamento Mineiro, pela estrutura e oportunidades concedidas durante toda a vida
acadêmica.
Aos amigos Ronald Leal e Bruno Kuckartz pelo auxílio fundamental no desenvolvimento da
programação e dos scripts, além das valiosas discussões e colaborações em todo o trabalho. É
muito bom poder trabalhar com profissionais tão qualificados, além de pessoas incríveis.
Ao Luciano Capponi, pela oportunidade concedida de utilizar os dados para os estudos de caso
e pelo apoio e incentivo do trabalho, desde a etapa do mestrado.
Agradeço ao Fernando Rios e à minha família, pelo apoio incondicional e por acreditarem em
mim em todos os momentos.
Um agradecimento especial ao professor Rodrigo Peroni. Obrigada por sempre acreditar neste
trabalho, pela confiança, dedicação e orientação durante toda a minha formação, desde a
graduação até a etapa da pós-graduação.
Por fim, muito obrigada a todos que colaboraram direta ou indiretamente na conclusão deste
trabalho!
RESUMO
A diluição e a perda são fatores importantes que podem impactar significativamente nos
resultados de uma operação de lavra. Quando se trata de lavra a céu aberto, tais fatores nem
sempre são calculados e analisados de maneira detalhada, muitas vezes são tratados como
constantes ou fatores percentuais médios, aplicados para atender necessidades de auditorias ou
reconciliações de dados. Dentre as diferentes abordagens que se pode ter quando se trata de
diluição de lavra, este trabalho irá focar na diluição operacional. A diluição operacional pode
ser definida como a incorporação de material estéril ao minério pela incapacidade operacional
de separar de forma eficiente os materiais durante a lavra, com os equipamentos disponíveis,
considerando os processos vigentes e a configuração geométrica da lavra. Este trabalho propõe
uma metodologia para quantificar a diluição operacional em lavra a céu aberto, utilizando dados
referentes aos planos mensais de lavra e modelos de teores. Para isso, foi desenvolvida uma
rotina automatizada para calcular a diluição, primeiramente baseada em um modelo único de
teores (estimado) e após utilizando modelos equiprováveis de teores (simulados), permitindo
que as incertezas dos teores dos blocos fossem utilizadas para medir a probabilidade de
ocorrência da diluição operacional. Os resultados demonstraram que o uso da metodologia
proposta através das rotinas automatizadas permitiu a sistematização e rastreabilidade de
processos que são executados repetidamente, além de executar tais etapas em menor tempo.
Nos estudos de caso apresentados nos dois primeiros artigos, foi possível concluir que a
variação mensal da diluição é bastante significativa; logo, utilizar um fator fixo não seria o mais
conveniente, mas sim, escolher um fator calculado por essa técnica referente a cada período
analisado, para garantir uma melhor reconciliação. Já o terceiro artigo mostra que a informação
de probabilidade de ocorrência da diluição pode ser utilizada para redefinir os planos de lavra
ou também para adequar o fator de diluição aplicado, dependendo das características da
operação. O estudo demonstra que a utilização de um fator de diluição correto é fundamental
nos processos de reconciliação de massas e teores, já que este tem origem no processo da
operação e varia a partir de diferentes etapas ao longo do planejamento de mina, influenciando
áreas a serem lavradas, orçamentos, eficiência da produção, e com consequência nos resultados
financeiros do projeto.
Dilution and ore loss are important factors that can affect significantly the results of a mining
operation. Regarding open-pit mining, these factors sometimes are not calculated and analyzed
in details, often are treated as constants or average factors, applied to attend audit requirements
or improving the data reconciliation. Among the different approaches that can be taken when
talking about mining dilution, this work will focus on operational dilution. Operational dilution
in open-pit mining can be defined as the incorporation of waste material into the ore due to the
operational inefficiency of separating materials during mining, considering the physical
processes and operational and/or geometric configurations of mining with the available
equipment. This work proposes a methodology to quantify the operational dilution in open-pit
mining, through the use of monthly mine plans and grade block models. An automated
workflow was developed, first based on a single ore block model (estimated by kriging) and
after using equally probable block models (simulated), allowing that the uncertainties of grades
were used to measure the probability of dilution in a given period of time. The results showed
that the use of the proposed methodology through automated routines allowed to the
systematization and traceability of processes that are executed repeatedly, besides executing
such steps in less time. In the case studies presented in the first two papers, it can be noted that
monthly variation of the dilution is quite significant; therefore, using a fixed factor would not
be adequate, but choosing a factor calculated by this technique according to each analyzed
period, to guarantee a better reconciliation. The third article showed that the information about
the propensity to occur dilution can be used to redefine the mining plans or also to adjust the
applied dilution factor, depending on the characteristics of the operation. The study
demonstrates that the use of a correct dilution factor is fundamental in the processes of mass
and grade reconciliation, as this originates from the operation process and varies from different
stages throughout the mine planning, influencing areas to be mined, budgets, production
efficiency, and with consequence in the financial results of the project.
Capítulo 1 .............................................................................................................................. 9
1. INTRODUÇÃO ....................................................................................................... 9
1.1 Estado da arte ........................................................................................................ 12
1.2 Justificativa ............................................................................................................ 16
1.3 Meta ...................................................................................................................... 17
1.4 Objetivos específicos da tese .................................................................................. 17
1.5 Metodologia ........................................................................................................... 17
1.6 Contribuição da tese............................................................................................... 18
1.7 Organização da tese ............................................................................................... 19
Capítulo 2 ............................................................................................................................ 20
2. INTEGRAÇÃO DE ARTIGOS.............................................................................. 20
Capítulo 3 ............................................................................................................................ 41
3. CONCLUSÕES E RECOMENDAÇÕES............................................................... 41
3.1 Conclusões ............................................................................................................ 42
3.2 Recomendações e Trabalhos Futuros...................................................................... 43
Referências .......................................................................................................................... 45
APÊNDICE A - OUTRAS PUBLICAÇÕES CIENTÍFICAS – Periódicos ........................... 49
APÊNDICE B - OUTRAS PUBLICAÇÕES CIENTÍFICAS – Congressos .......................... 51
CAPÍTULO 1
1. INTRODUÇÃO
A diluição também pode ser citada como um importante fator na busca de uma
reconciliação ideal. Ela pode ser definida como material estéril incorporado ao minério durante
sua extração e enviado para a usina de beneficiamento, reduzindo assim os teores de minério
previamente estimados. Já a perda, segundo Villaescusa (1998), é definida como o material
econômico que deixa de ser lavrado ou mesmo que é misturado ao material estéril e
consequentemente descartado devido a condições de operação e execução de lavra. Utilizar
uma estimativa da diluição no planejamento de lavra pode auxiliar muito na obtenção de uma
reconciliação mais realista. É impossível eliminar totalmente a diluição na operação de lavra;
entretanto, ela pode ser controlada e quantificada. Entendendo melhor as causas da diluição e
utilizando estas informações no planejamento, a mesma pode ser reduzida e o planejamento se
torna mais confiável.
10
devido a condições operacionais, levando assim a uma diluição não planejada ou operacional.
Nesta figura pode ser visto como a diluição ocorre em diferentes etapas (estimativa,
planejamento e operação) e que o conceito de diluição pode ser adaptado tanto para lavra
subterrânea (stopes) quanto para lavra a céu aberto (bloco).
Em grande parte das abordagens do tema, a diluição em lavra a céu aberto é tratada
como um problema de estimativa do modelo de blocos, de erros de classificação dos blocos em
minério ou estéril (SINCLAIR; BLACKWELL, 2004). Também conhecida como diluição
planejada, geralmente é utilizada em horizontes de planejamento de longo prazo e aplicada
como um fator nos modelos de estimativas. Entretanto, outro tipo de diluição, aqui chamada de
diluição operacional, a qual ocorre no momento da lavra, muitas vezes não é quantificada e
considerada para comparação de massas planejadas e apontadas na execução. Tal informação
muitas vezes é negligenciada e somente será avaliada no momento da reconciliação dos dados,
onde baseado em um mining call factor (MCF), ou fator de reconciliação, é feita uma correção
dos dados para anular as discrepâncias observadas sistematicamente (SINCLAIR;
BLACKWELL, 2004).
O uso de fatores de correção para atingir reconciliações mais aderentes é muito comum
em mineração, entretanto, para que estes fatores sejam estimados de uma maneira sistemática,
11
é necessário conhecer a maior quantidade de características possível do depósito e da operação.
Segundo Matthews (2015), em muitos casos, as reservas são estimadas sem dados de produção
disponível para calcular o MCF; entretanto, ainda assim é necessário estimar fatores razoáveis
de perda e diluição de minério para o depósito. Se a perda e diluição do minério forem mal
estimadas, isso poderá ter implicações significativas nos resultados econômicos do projeto.
Diversos autores como Pitard (1998), Best e Gallant (2004), Morley (2003), Fouet et
al. (2009), Parker (2012), entre outros, apontam para o fato de que a qualidade dos dados é um
dos aspectos mais importantes no processo de reconciliação e que a análise dos mesmos deve
ser feita sempre em conjunto. Segundo Bertinshaw e Lipton (2007), a reconciliação deve ser
realizada em intervalos apropriados de tempo para checar o desempenho dos modelos de
12
recursos e reservas e fornecer parâmetros científicos para ajustar as estimativas de diluição e
perdas. Além de intervalos de tempo adequados, a reconciliação também deverá ser realizada
em intervalos de espaço monitorados, para que características específicas do depósito não
mascarem possíveis problemas de operação (continuidade de um corpo, por exemplo).
Souza (2002) diz que as principais razões para uma reconciliação insatisfatória estão
relacionadas, na maioria das vezes, com a incerteza na interpretação geológica e na estimativa
de parâmetros. Muitos fatores podem afetar o grau de incerteza no processo de estimativa, por
exemplo: o efeito da mudança de suporte; a insuficiência de amostras, a má amostragem ou a
técnica de preparação; erros de interpolação na geração do modelo de blocos; efeito de
suavização dos interpoladores, etc. Desta forma, independentemente das razões, é importante
reconhecer que não existem modelos de blocos perfeitos, no sentido de que sempre existirá um
erro associado (incerteza). Schofield (1998) aponta para o fato de que nenhum método de
estimativa será bom em todos os aspectos. Na seleção do método mais adequado, devem ser
levadas em consideração algumas questões como a variação de minério no depósito e qual a
expectativa de recuperação da porção mineralizada em minério.
13
Segundo Abichequer (2010), os métodos de estimativas tradicionalmente utilizados
para construção do modelo de blocos não são capazes de prever a variabilidade associada à
estimativa dos teores de maneira efetiva. Diferente da krigagem, os métodos de simulação
geoestatística têm como objetivo reproduzir a variabilidade in situ, e a continuidade espacial
dos dados originais, pela geração de imagens equiprováveis condicionadas aos dados que
reproduzem as características estatísticas de 1ª e 2ª ordem dos dados amostrais. Desse modo, a
intensidade de incerteza associada às estimativas pode ser avaliada (GOOVAERTS, 1997). O
objetivo da krigagem e da simulação são muito diferentes: enquanto a krigagem busca a
minimização do erro associado ao valor previsto localmente, a simulação tem como objetivo
construir modelos de incerteza que devem reproduzir as características de variabilidade natural
dos fenômenos. Os mapas provenientes da krigagem fornecem a visualização das melhores
estimativas locais, porém quando analisados globalmente apresentam alguns desvios. Por outro
lado, o modelo simulado é capaz de reproduzir estatisticamente os dados originais, ou seja,
reprodução de histogramas e modelos de variogramas mais próximos e semelhantes aos dos
dados originais.
Alguns trabalhos mais recentes no tema de diluição em lavra a céu aberto como
Domingo et al. (2015) e Eshun e Dzigbordi (2016) apontam para o uso de sistemas de controle
de movimento do material durante o desmonte de rocha por explosivos, a fim de minimizar a
perda de minério, diluição e, às vezes, erros de classificação. Ambos trabalhos focam na
minimização da diluição através de métodos que propõem um melhor sequenciamento no
desmonte, utilizando para isso detonadores eletrônicos e modelos que preveem o movimento
dos materiais desmontados. Estes estudos mostram que o uso de tais sistemas pode contribuir
eficientemente para o controle da diluição, porém, o custo de sua aplicação pode ser bastante
alto.
15
Zarnshenas e Saeedi (2016) fazem uma avaliação do risco da diluição em minas a céu
aberto através de um modelo que quantifica o risco de diluição de acordo com características
específicas da mina, no que diz respeito a riscos ambientais, econômicos, técnicos, sociais
(comunidade), de saúde e segurança, etc. Cada um destes riscos principais irá receber um peso
de acordo com seus subitens, por exemplo: o risco social e comunidade, onde a agricultura entra
como um subitem e os riscos associados à mesma se referem à poluição na água, ar e solo.
Outro exemplo é a respeito do risco econômico, onde a diluição pode influenciar de maneira
direta. Ainda segundo Zarshenas e Saeedi (2016), se as propriedades do minério e do estéril
forem diferentes, a planta, equipamentos e todo o sistema que é feito baseado no minério irão
encontrar problemas. Da mesma forma, problemas técnicos como erros no destino do material
(despacho) poderão acontecer se as propriedades físicas do estéril forem muito semelhantes às
do minério.
1.2 Justificativa
1.3 Meta
Esta tese de doutorado tem como meta desenvolver uma metodologia para
quantificação da diluição operacional em lavra a céu aberto, utilizando para isso informações
referentes a modelos de blocos e planos mensais de produção.
Para atingir o objetivo geral e meta propostos, os seguintes objetivos específicos serão
buscados:
1.5 Metodologia
17
o uso de explosivos. A exemplo de mineração subterrânea, a diluição é calculada bloco a bloco,
porém é contabilizada dentro do painel ou do avanço programado para um determinado período,
ponderando os resultados pelas massas lavradas.
18
uma rotina de fácil uso permitirá a sua aplicação em casos práticos, contribuindo assim para
melhorar resultados do planejamento de mina.
19
CAPÍTULO 2
2. INTEGRAÇÃO DE ARTIGOS
Câmara, T.R., Leal, R.S., Peroni, R.L. and Capponi, L.N. Controlling operational
dilution in open-pit mining. Mining Technology, v. 128, Issue 1, p. 1-8, 2018 (TIMM
A), 2018. https://doi.org/10.1080/25726668.2018.1470275
Câmara, T.R., Leal, R.S. and Peroni, R.L., Accounting for operational dilution by
incorporating geological uncertainties in short-term mine planning. DYNA, 87(213),
pp. 178-183, April - June, 2020.
http://doi.org/10.15446/dyna.v87n213.83661
20
O Artigo 1, publicado no periódico REM - International Engineering Journal, mostra
a quantificação da diluição operacional em lavra a céu aberto devido à ineficiência de operação.
Esta primeira abordagem levou em consideração aspectos como a diluição interna dos blocos e
a ineficiência da operação foi medida de acordo com a capacidade do equipamento em lavrar o
bloco de maneira fidedigna e seletiva. Neste trabalho, a partir da informação correspondente às
poligonais de planejamento e modelo de teores, a diluição interna dos blocos foi calculada de
forma não automática, quantificada para cada poligonal de planejamento de forma individual.
A principal contribuição do Artigo 1 é a de evidenciar que a utilização de um fator fixo de
diluição para corrigir planos de lavra e posteriores reconciliações é suplantada pela utilização
de um fator baseado em informações específicas do plano em questão e que esses valores variam
ao longo do tempo.
O Artigo 2, publicado na revista Mining Technology (TIMM A), mostra uma evolução
considerável na metodologia desenvolvida no primeiro artigo. Além da diluição interna dos
planos, neste artigo também foi considerada a diluição que ocorre pela influência do ângulo de
face de bancada. Outro ponto de grande contribuição foi a sistematização do cálculo da diluição
de forma automática, apresentado através de um script (baseado em funções presentes no
software comercial de planejamento de lavra e uso de linguagem JavaScript e HTML) que é
capaz de calcular a diluição de maneira a garantir a confiabilidade e rastreabilidade das
informações, em curto período de tempo. O segundo artigo também mostrou duas diferentes
abordagens para o cálculo de diluição, a depender se o desmonte das frentes é realizado via
explosivos ou escavação mecânica.
O Artigo 3, formalmente aceito pela revista DYNA, acrescenta aos resultados dos
artigos anteriores as medidas de incerteza sobre os teores do depósito. O artigo apresenta como
principais contribuições o desenvolvimento de uma rotina de cálculo de diluição (script
Python), independente do software comercial de planejamento de lavra e, principalmente, a
abordagem do uso de informações de incertezas geológicas para a previsão e controle da
diluição. Neste trabalho, a quantificação da diluição utilizou informações de modelos de blocos
simulados, com n-teores (equiprováveis), onde a partir destas informações foi possível
identificar locais com maior probabilidade de ocorrência de diluição dentro de um plano de
lavra. A informação de áreas com maior propensão de ocorrência de diluição é muito útil para
possíveis correções de tais planos, caso as condições operacionais permitam. Mais uma vez,
neste trabalho é salientado que a informação da diluição não precisa ser utilizada somente como
21
um fator na reconciliação, mas também pode e deve ser usada como parâmetro de entrada no
planejamento da lavra.
22
o teor do bloco é igual em toda a sua extensão. Desta forma, foi possível quantificar a diluição
causada pela influência do ângulo de face de bancada e a diluição interna, uma vez que foi
assumido que o teor correspondente às porções adicionadas (ou deixada de ser lavrada, no caso
da perda) era o mesmo do que o bloco inteiro. Outra premissa assumida foi a respeito da
abordagem utilizando o método de desmonte via explosivos. Geralmente, o desmonte de rochas
com uso de explosivos resulta em maior diluição do que o método de escavação mecânica,
porém no estudo de caso apresentado no Artigo 2 foi assumida a premissa que, após o desmonte
via explosivos, a separação dos materiais dentro de uma mesma poligonal se torna inviável.
Desta forma, tal abordagem considerou o cálculo da diluição apenas para os blocos localizados
na borda das poligonais planejadas. Para que o resultado não seja interpretado de maneira
equivocada, é importante ressaltar que para utilizar a metodologia de cálculo da diluição para
operações com desmonte de rochas com uso de explosivos, os locais previamente definidos a
serem desmontados deverão considerar a melhor separação possível entre áreas de minério e
estéril.
23
Tais Renata Câmara and Rodrigo de Lemos Peroni
Mining
Mineração
Quantifying dilution
http://dx.doi.org/10.1590/0370-44672014690006 caused by execution efficiency
Taís Renata Câmara Abstract
Pesquisadora assistente de Lavra de Minas
Instituto Tecnológico Vale In open pit mining, dilution is not always a factor systematically analyzed and
Ouro Preto – Minas Gerais - Brasil calculated. Often it is only an adjusted number, for example, calculated or even em-
[email protected] pirically determined for a certain operational condition perpetuating along time in the
form of a constant applied to calculating reserves or mine planning in attendance of
Rodrigo de Lemos Peroni audit requirements. Dilution and loss are factors that should be always considered for
Professor Associado tonnage and grade estimates. These factors are always associated and can be deter-
Universidade Federal do Rio Grande do Sul – UFRGS mined considering several particularities of the deposit and the operation itself. In this
Departamento de Engenharia de Minas study, a methodology was determined to identify blocks adjacent to the blocks previ-
Porto Alegre – Rio Grande do Sul – Brasil ously planned to be mined. Thus, it is possible to estimate the dilution caused by poor
[email protected] operating efficiency, taking into account the inability of the equipment to perfectly
remove each block, respecting its limits. Mining dilution is defined as the incorpora-
tion of waste material to ore due to the operational incapacity to efficiently separate
the materials during the mining process, considering the physical processes, and the
operating and geometric configurations of the mining with the equipment available.
1. Introduction
All mining operations experience estimated. And the loss, according to the complexity of the contact between
dilution at some time and in most Villaescusa (1998), is defined as the ore and waste, while the degree of
cases, the elimination of the waste economical material that is not mined influence will be dependent on how
within the ore blocks is impossible. due to geological aspects and operat- abrupt is the change in grades in this
However, experience has shown that ing conditions. contact. External dilution can also
dilution can be controlled to accept- Dilution can be subdivided into be a result of the size and position
able levels by the implementation two categories: internal and external of the block to be mined. In a block
of correct mining engineering prin- to the ore. According to Sinclair and model representation, each one of
ciples (Butcher, 2000). According to Blackwell (2004), both categories of the six blocks contiguous to the con-
Pakalnis et al. (1996), ore losses and dilution can be further subdivided on sidered one have influence on a type
dilution are present at all stages of the basis of geometric considerations of dilution, that can be horizontal or
mining and while several models can about the deposit itself or of the dilut- vertical (Figure 1). If the block to be
investigate the influence of dilution, it ing material. External dilution can be mined is in contact with blocks that
is its quantification that represents the related to minimum mining width, are not mineralized (or if they are of
most serious challenge. Furthermore it contact dilution and overbreak of low grade ore), the dilution will occur
is now recognized that what is consid- wallrock relative to planned mining in this contact when imperfect separa-
ered acceptable dilution is a function margins. Internal dilution can be tion methods are used or to maintain
of ore grade, percentage of dilution, considered from the perspective of the geotechnical slope angle of the
production costs and ultimately of volumes of barren rock within an ore wall face.
ore selling prices. Consequently the zone or the inherent diluting effect Rarely can the block be removed
degree of acceptable dilution differs resulting from either increasing the without incorporating any part of the
from site to site. size of SMUs or the effect of blocks lateral blocks. Also, considered must
A definition for dilution is a por- misclassification resulting from sam- be the fact that the operator should
tion of waste material incorporated pling and analytical errors occurring have full control of positioning, and
into the ore during its extraction and in grade control. that the equipment selected for the
transport to the processing plant, thus External dilution is a result of block extraction should be properly
reducing the grades of ore previously the mining method chosen, as well as sized, thus minimizing the dilution.
REM: Int. Eng. J., Ouro Preto, 69(4), 487-490, oct. dec. | 2016 487
Quantifying dilution caused by execution efficiency
Figure 1
Block to be mined and the six
blocks contiguous to the considered one.
In short term planning, polygons other physical marking on site. causing differences in relation to planning.
are usually defined to delineate the loca- In some companies that use more This aspect can certainly be cited as one
tions where mining should be made based technology in their operations, many of the factors that cause differences during
on the processing plant requirements or mining machines have Global Positioning the reconciliation process. The dilution
product specification, such as average head Systems (GPS) installed. However, even that occurs due to these problems in the
grade, contaminants, rock hardness and with the use of technological resources, execution can be calculated using the
total milling capacity (Câmara, 2013). The often due to operating problems (such as methodology shown in this study and this
delimitation of these areas in the field can equipment's selectivity or, operator’s skills) dilution can be associated as short term
be made through the use of flags or any these polygons are not strictly followed, mine planning dilution.
Methodology
To apply the methodology, the block’s height or to the bench height, by the mine planning. Figure 2 shows a
first step is to select the blocks falling depending on the operation approach. plan view of the deposit, where it can be
within the polygons of the monthly The union of the base and top polygons observed that the model is represented
plan. Thus, the type of rock that is now sets a solid that represents the in the colored blocks in the background
present in each polygon planned to be volume to be removed from the block representing different rock types that
mined can be determined. These poly- (or portion) and weighted average occur in the ore deposit, according
gons are contained in a given reference considering the masses, allowing one to the legend provided. In pink, are
level, normally associated with the base to calculate the value of the variables representations of what would be the
of the blocks. After, the polygons are of interest that are expected to be ac- mining areas planned for a given period
projected to a distance equivalent to the complished according to targets defined (monthly plans, for example).
Figure 2
Planning polygons
superposed to the block model.
488 REM: Int. Eng. J., Ouro Preto, 69(4), 487-490, oct. dec. | 2016
Tais Renata Câmara and Rodrigo de Lemos Peroni
After creating the solids from plan- includes 1 meter of each block that is in the of the polygon, i.e., the blocks that will
ning polygons, it is possible to identify contact area. For a block with dimensions cause dilution if they are in contact with
the blocks that are inserted within these of X, Y and Z corresponding to 25m, 25m the planned blocks, considering they are
solids and the blocks that are in contact and 10m, respectively, this addition of 1 low-grade ore or waste, the polygon was
with the boundary blocks. Considering meter represents 4% of the total mass to be expanded horizontally taking into account
that the process of mining is imperfect, mined. In this example it was considered the block dimensions in X and Y direc-
contact blocks will be diluted because the that the equipment removes more material tions and transformed into a solid using
equipment is not able to select the block to calculate dilution, but the equipment the same process described above to select
exactly where it ends or at the dig line could also remove less material, thus caus- the planned blocks. After performing
previously determined, if this line does not ing losses. Therefore, this methodology the step of selecting the blocks planned,
coincide exactly with the boundary of the allows defining both dilution and losses in the process is repeated for the expanded
blocks. For the definition of the dilution, it a block inside a planned area. To identify polygon, so that the contact blocks and
was considered that the mining equipment which blocks are on the external limit their grades can be identified (Figure 3).
Figure 3
Blocks located in the dig line contact limit.
The selections of blocks are ex- contact and which are not. Then, for the that are in contact with blocks of waste
ported to an Excel spreadsheet. The first contact blocks, identify which blocks or low-grade ore, dilution is calculated
step is to analyze in the worksheet of the are the adjacent ones, and if they are and a new grade is now applied to that
expanded model, which blocks are of the blocks of ore or waste. To those blocks diluted block.
Considering that the dimensions in that XC and YC represent the coordinates boring blocks can be identified using the
the X and Y directions are of 25 m and of the centroid of each block, the neigh- calculation shown in Table 1.
After identifying the block, it is tion can be calculated for each block the block located to the North of the
possible to query other important fields analyzed. This analysis process evaluates original block is identified as low-grade
related to these blocks, such as ROCK each block individually, but the results ore, and it will cause dilution in the grade
TYPE and P2O5AP grade (apatite phos- can be computed together for the whole of the original block. This analysis is
phate grade). By knowing the contacts dig line. Table 2 shows an example of repeated for all the blocks that are in the
and the corresponding grades, dilu- analysis for a specific block. As shown, contact region.
REM: Int. Eng. J., Ouro Preto, 69(4), 487-490, oct. dec. | 2016 489
Quantifying dilution caused by execution efficiency
First of all, the location of the block according to the mining equipment selec- grades of about 3.12%.
is evaluated along upon direction this tivity for this specific operation; however, The dilution results show consider-
block has contact with unplanned blocks. it can vary according to the characteristics able values, which demonstrate that even
Secondly, the rock type and P2O5AP grade of each operation. where the reconciliation adherence is quite
of the contact is analyzed and if the rock With the grades diluted according to satisfactory, dilution may occur. In this
type of the contact block is different from the type of contact, the total dilution of the example it occurs due to the lack of the
the rock type of the analyzed block, then planned polygon can be calculated. In the selectivity capacity of the equipment and/
dilution is calculated. As stated earlier, it example tested in this case, the plan was to or operator skills and mining delimitation
is considered that the mining equipment mine the total mass with an average grade practices. This demonstrates the impor-
incorporates 1 meter of the adjacent block of 10.82%. After calculating the dilution, tance of reconciliation between planning
in the mined block, a value that is also in- the new average grade was of 10.48%. and execution in mining to control dilu-
corporated into the calculation of the total The difference in grades is 0.34%, a value tion, and if there are problems in reconcil-
mass mined. This 1 meter was defined that corresponds to a dilution of P2O5AP ing these, the dilution will be much higher.
4. Conclusions
Controlling dilution in a system- systematically approach this subject mine life without considering that the
atic way allows for a better understand- that is commonly neglected or masked origin and consequences of this choice
ing of problems during mass and grade during mining. It was attempted to can lead to differences in reconcilia-
reconciliation processes and where make a simple and direct approach to tion and generalizing to deposits that
they occur. According to Câmara et a parameter that is notoriously complex have different contexts. The analysis
al. (2013), ore losses and dilution are and difficult to control. It is known that of the dilution caused by the execution
problems that are often difficult to most mining companies either simply efficiency proved to be quite satisfac-
locate and quantify, but with a good disregard this effect or use a factor tory and with this methodology, it is
method of control, it is possible to mini- without considering the particularities possible to identify the contacts of the
mize them. The calculation proposed of the deposit, just applying a fixed planned blocks and after determining
within this methodology is a way to number for the entire deposit along the their grades, calculate the dilution.
5. Acknowledgements
The authors acknowledge the post- eral Research Laboratory) from DEMIN/ and support during the work. And to
graduating program (PPGE3M) of Fed- UFRGS, for the support and infrastruc- CNPq, National Council of Technology
eral University of Rio Grande do Sul ture to develop this study. The company and Science Development – Brazil, for the
(UFRGS); LPM (Mine planning and Min- Vale Fertilizantes S.A., for providing data scholarship grant.
6. References
490 REM: Int. Eng. J., Ouro Preto, 69(4), 487-490, oct. dec. | 2016
MINING TECHNOLOGY
https://doi.org/10.1080/25726668.2018.1470275
Introduction
to the operational inefficiency of separating materials
Dilution refers to the material below the cut-off grade during mining, considering the physical processes,
that is mixed with ore during mining, thereby reducing operational and/or geometric configurations of mining
the grade of the excavated material (Tommila, 2014). with the available equipment. Short-term operational
Dilution can be classified into two categories according dilution that occurs during mining is quantified
to Sinclair and Blackwell (2004): internal or external according to the contact between the planned blocks
with regard to the ore. Internal dilution is the waste and their neighbourhood, and according to the way
material that is necessary to extract the ore and its sep- the blocks are mined, primarily considering whether
aration during mining is considered impossible, and it the dig line limits (i.e. mine plans) are respected.
is intrinsic to the mining method and equipment used. In mechanical excavation, dilution depends on the
Ebrahimi (2013) defined external dilution (also called selectivity of the mining equipment, skill of the oper-
contact dilution) as the waste that is mined within ator and correct identification of blocks in the field.
the mining block. External dilution varies based on If the blocks are correctly identified and the mining
the geology, shape of the orebody, drilling and blasting equipment has Global Positioning System (GPS) con-
techniques, scale of operation and equipment size. Vil- trol, the errors tend to be smaller and dilution therefore
laescusa (1998) defined ore loss as the economical tends to be reduced. For operations that use drill and
material that is not mined or that is mixed with the blast disaggregation, dilution can be measured by the
waste material because of geological factors and operat- amount of waste incorporated in the ore after blasting.
ing conditions. In this case, dilution can be caused by problems in dril-
Dilution is extremely important in some types of ling patterns, detonation times or by operational pro-
underground mining because of the higher selectivity blems, such as back break, sub-drilling and hole
of those mining methods compared to open-pit mining deviation (Fernberg, 2008).
methods. In the case of open-pit mining, the most According to Pakalnis, Poulin, and Hadjigeorgiou
common way to represent dilution is through the use (1996), dilution and ore losses can be present at all stages
of resource/reserve models, but in such models, blocks of mining and the degree of acceptable dilution differs
can be misclassified in the estimates or simulations. from site to site. Dilution cannot be completely avoided,
Ebrahimi (2013) demonstrated the importance of cal- but it can be controlled at acceptable levels by imple-
culating dilution in open-pit mining considering this menting correct mining engineering principles (Butcher,
factor when designing or evaluating mines. 2000). This study aims to quantify operational dilution
Câmara (2013) defined operational dilution as the in open-pit mining by evaluating the effect of ineffi-
incorporation of the waste material into the ore due ciency in mining on the planned masses and grades.
CONTACT Taís Renata Câmara [email protected] Instituto Tecnológico Vale (ITV), Avenida Juscelino Kubitschek, 31, First Floor, Bauxita, Ouro
Preto 35400-000, Minas Gerais, Brazil
© 2018 Institute of Materials, Minerals and Mining and The AusIMM Published by Taylor & Francis on behalf of the Institute and The AusIMM
2 T. R. CÂMARA ET AL.
Figure 1. Ore block surrounded by waste blocks. The dashed lines represent the portion of the waste blocks that will be added to
the ore block during mining.
MINING TECHNOLOGY 3
Figure 2. Dilution for the blocks located at the borders of the mine plans due to the influence of bench face angle.
and Y dimensions) and in the equipment selectivity, Deposit 2 is also a phosphate deposit but composed
which was chosen to be 1 m in this case. The mine of rock types different from the rock types that com-
plans used are the monthly advances represented by posed deposit 1. In this case, drilling and blasting was
the polygons (as shown in Figure 5) with 700–1500 performed to disaggregate the material. The definition
planned blocks in the evaluated months. The total of ore and waste was also based on the P2O5AP grade.
planned mass varies according to the periods and The cell size of the block model used for the evaluation
areas available to access. of deposit 2 is 10 m in the X, Y and Z directions; the
Figure 6. Drill and blast plan represented by polygons and block model.
6 T. R. CÂMARA ET AL.
bench height used is the block height (Z), as in the pink polygons are the mine plans and the squares
deposit 1. Figure 6 shows a plan view of deposit 2, are the blocks coloured according to the grades of the
where the white dots represent the drilling pattern, blocks.
Results and discussions In both case studies presented, the bench heights
and bench face angles were set according to the specific
To validate the methodology, the script was tested in
characteristics of each operation, demonstrating the
two case studies using monthly dig lines related to
adaptability of the routine created. Nevertheless, it is
the planning polygons and calculating the
extremely important that the data representing the
dilution and losses based on the block model. All
planning polygons (areas to be mined) be designed cor-
procedures take an average of 1–2 min to execute
rectly such that the blocks (or parts of them) within the
using the files for the case studies considered in
polygonal can be identified. Polygons with small grades
this work.
or with complex contacts between ore waste are more
The first case study, considering mechanical exca-
likely to present higher dilution; hence, the use of a
vation, yielded the data shown in Figure 7. The graph
constant dilution factor might lead to missing mass
shows a comparison between the planned average
of ore in those periods.
grades (triangles) versus the diluted average grades
The dilution calculated is considered one of the
(dots). The bars are related to the calculated dilution
possible causes of the lack of reconciliation between
for each period that can be read in the right-hand
planning and execution. By incorporating dilution
axis, and the solid lines are related to the P2O5 grades
in the short-term plans, it is possible to have greater
read in the left-hand axis.
control on the grades and masses fed to the
As can be observed in Figure 7, operational
processing plant. The results demonstrate that
dilution occurred in all mining periods, demonstrat-
dilution and ore loss are variable for each mining
ing to be variable through the periods. The different
period, and that they are not small to be disregarded.
results obtained for each period may be attributed
This work does not suggest that studies like those
to different factors, such as the area planned for a
presented here should be done to find out a
particular period being in a region of close contact
dilution factor that will be used ahead, but stimulate
with waste zones. The highest dilution observed was
that this must be implemented in the operation rou-
in a period that had the smallest planned grade
tine to have the best estimate as possible of the real
(period 5), which could indicate a low-grade polygon
dilution for a particular operation and possibly
or a large ore/waste contact zone.
reduce that factor to acceptable levels when it is
In the second case study (drill and blast disaggre-
too high.
gation), the same test was performed employing the
second ‘Mining method’ option in the script. The
results shown in Figure 8 present similar behaviour Acknowledgements
to that observed in the previous test. For all periods, The authors acknowledge Vale Fertilizantes, Inc., for provid-
small differences were observed between the planned ing data and support during the work. Special thanks are due
and diluted grades, but even these small differences to the Universidade Federal do Rio Grande do Sul and the
result in grade dilution that must be taken into Vale Institute of Technology.
account. As in the previous case study, the period
with the smallest planned ore grade presented higher
dilution. Disclosure statement
No potential conflict of interest was reported by the authors.
Conclusions
References
In open-pit mining, dilution is generally accounted
Butcher RJ. 2000. Dilution control in Southern African
for only as the misclassification of blocks in the
mines. Proceedings, MassMin 2000 Conference.
block model estimates. The results of this work Australasian Institute of Mining and Metallurgy.
demonstrated that in addition to misclassification in Brisbane, Australia, p. 113–118.
these models, it is important to calculate the dilution Câmara TR. 2013. Sistematização do cálculo de diluição e
that occurs during mining, i.e. the operational perdas operacionais para reconciliação de teores e massas
dilution. It is clear that using a constant number to em lavra a céu aberto [master’s thesis]. Federal University
of Rio Grande do Sul - UFRGS. 89p. Portuguese.
express dilution in the operation is not valid, consid- Câmara TR, Peroni RL. 2016. Quantifying dilution caused by
ering the dilution is dependent on a number of fac- execution efficiency. REM. 69(4):487–490.
tors that vary from period to period. As a general Crawford GD. 2004. Dilution and ore recovery. Pincock
rule, for larger periods (years, semester), it might Perspectives. 60(4):487–490
become easier to estimate dilution, but the operation Ebrahimi AB. 2013. The importance of dilution factor for
open pit mining projects. World Mining Conference
needs to deal with monthly, weekly or even daily
2013, Montreal, Canada, 12p.
production; hence, providing a way to understand Fernberg H. 2008. Surface drilling – Atlas Copco Reference
and explain the source of the differences is Book. 4th ed.; p. 15–16. www.fing.edu.com.uy/iiq/
paramount. SurfaceDrilling.pdf.
8 T. R. CÂMARA ET AL.
Pakalnis RC, Poulin R, Hadjigeorgiou J. 1996. Quantifying Tommila E. 2014. Mining method evaluation and dilution
the cost of dilution in underground mines. Int J Rock control in Kittilä mine [master’s thesis]. Aalto
Mech Min Sci Geomech Abstr. 33(5):1136–1141. University School of Engineering. 63p.
Rossi ME. 2009. Accounting for dilution in ore resource esti- Villaescusa E. 1998. Geotechnical design for dilution control
mation. Proceedings APCOM 2009. Vancouver, Canada, in underground mining. Proceedings of the Seventh
p. 84–92. International Symposium on Mine Planning and
Sinclair AJ, Blackwell GH. 2004. Applied mineral inventory esti- Equipment Selection. Calgary, Balkema, Rotterdam,
mation. Cambridge: Cambridge University Press. 401p. p. 141–149.
DYNA
http://dyna.medellin.unal.edu.co/
Received: November 21th, 2019. Received in revised form: April 3rd, 2020. Accepted: April 13th, 2020
Abstract
Operational dilution in open-pit mining can influence short-term mine planning and affect the adherence between designed and executed
plans, impacting the reconciliation results. This type of dilution may occur due to several reasons, such as erroneous assumptions regarding
contacts, inefficient mining performance and uncertainties associated with the planned grades. The use of uncertainties enables better
predictability and achievement of planned targets. Using the mine planning polygons and grade models, a methodology is proposed to
measure the dilution ranges in each region. The methodology considers the uncertainty associated with the ore grades to verify the
probability of occurrence of dilution in the short-term mine plan. The results demonstrated that the use of uncertainties enabled
identification of the areas with a higher potential of incurring operational dilution. Using this information, the plans can be adjusted to
minimize dilution, or the result can be used as a factor to correct the planned data.
Resumen
La dilución operativa en la minería en rajo puede afectar la adherencia entre los planes diseñados y ejecutados, lo que afecta los resultados
de la reconciliación. Este tipo de dilución puede ocurrir debido a suposiciones erróneas con respecto a los contactos, desempeño minero
ineficiente e incertidumbres asociadas con las leyes planificadas. Utilizando los polígonos de planificación y los modelos de bloques, se
propone una metodología para medir los rangos de dilución en cada región. La metodología considera la incertidumbre asociada con las
leyes de mineral para verificar la probabilidad de ocurrencia de dilución en el plan de mina a corto plazo. Los resultados demostraron que
el uso de incertidumbres permitió la identificación de las áreas con un mayor potencial de incurrir en una dilución operati va. Con esta
información, los planes pueden ajustarse para minimizar la dilución, o el resultado puede usarse como un factor para corregir los datos
planificados.
Block Schedule [6]) are implemented as several software calculated factor, often it is just a factor applied to the whole
algorithms that consider as many parameters as possible to deposit or some zones of the deposit to satisfy the needs of
obtain results that are more realistic. An optimization model audits [15]. Dilution as well as ore loss are factors that should
must involve the costs, especially mining and processing be always considered for tonnage and grade estimates during
costs, and the generated income. The basic input data to the mine planning.
optimization problem is a set of economic block values According to Câmara et al. [16], in open-pit mining,
representing the net value of each block. These block values dilution is generally accounted for only as a misclassification
are determined using the cut-off grade, costs, prices, of blocks in the block model estimates. Rossi [17] showed
recovery, dilution, density and operational parameters (e.g., the importance of incorporating an appropriate amount of
slope angle, capacities, and production) [7]. dilution in the “undiluted” resource model such that a
Over the years and with the technological advances, more recoverable reserves model would eventually be obtained.
variables have been added to the optimization process, The study of Bertinshaw and Lipton [18] reviewed various
making it more realistic and adherent to the operational data. approaches to the estimation of dilution and ore loss, as
Among these improvements, simulated block models can be implemented in computer block models, and considered
cited, i.e., multiple equally probable models considering these variables as an essential part of any ore reserve estimate
grade and geological uncertainty. A method for capturing the or mine schedule. A probabilistic approach to calculate the
uncertainty is to create stochastic simulations of the resource dilution and the ore loss of a long-term model, considering
model, where the stochastic behaviour expresses the the simulation results was presented in [19].
uncertainty or variability of the model’s properties [8]. A In addition to misclassification of blocks, another type of
common technique for creating multiple realizations is dilution − operational dilution − should be considered in
through geostatistical simulation procedures, e.g., turning mine planning, especially over a short-term horizon. An
bands, LU decomposition and sequential Gaussian approach to determine operational dilution as the
simulation. Simulation also provides a means for incorporation of waste material into the ore due to the
incorporating a quantified risk analysis within the planning operational inefficiency of separating materials during
and design procedures [9]. mining, considering the physical processes and operational
Considering uncertainties in mine planning is a necessary and/or geometric configurations of mining with the available
task for achieving the expected targets, regardless of the equipment was presented in [15].
planning stage considered. Spleit [10] stated in his thesis that This paper will show how the use of information of
one of the key risks in a mining project was geological geological uncertainties can contribute to control the
uncertainty because the understanding of the geology, spatial operational dilution in open-pit mining, besides enabling the
distribution and variability of the ore qualities could only be development of mine plans seeking to minimize dilution. To
inferred from limited data that was not necessarily this end, the proposed methodology will present an evolution
representative of the entire deposit. The larger the amount of of the routine presented in [16].
available information is, the greater the chance of obtaining
better results. The use of geological uncertainties is an 2 Methodology
already well-understood task and is a part of the planner's
routine when some variables have not been fully included to The methodology developed by [16,20] consisted of an
date and require more attention. Dilution is an example of a automated routine for calculating operational dilution
variable with few developments regarding its impact on mine through the analysis of short-term mine plans (subsets of
planning, especially in regard to dilution in open-pit mining. blocks to be extracted), analysing blocks’ grades, contacts
Dilution can be defined as waste or low-grade materials and neighbourhoods to verify if there were differences
(below the cut-off grade) that are not separated from the ore between the planned grades and the measured data. This
during the operation; it reduces the quality of the mined ore work proposes improving the mentioned methodology by
[11]. Matthews [12] defined ore loss as a reduction of ore incorporating the information of grade uncertainty to
volumes in the mining process due to misclassification of ore calculate the operational dilution. Another improvement was
as waste in the ore control process, exclusion of to rewrite the previous routine in the Python language to
discontinuous mineralization from mining due to geometry make the routine faster and more flexible, regardless of the
or continuity, and mechanical losses in the mining process. mine planning software used. The routine is very
Xingwana [13] affirmed that the inability to minimize ore straightforward and can be used for different deposits with
loss can thus impact negatively on mine output through the different numbers of simulation scenarios. To use the
loss of revenue. He also cited that ore dilution can impact proposed routine, it is necessary to input the data on block
negatively on the mine profit margin as a result of the extra classification (ore or waste). Typically, this classification is
costs involved in mining and processing of the waste that is based on rock type and grades prior to model definition.
treated as ore. Another necessary input data are the columns that correspond
Instead of quantifying dilution in mining studies, it was to the grades.
common to assume a general dilution such as 5% for massive The process of calculating the operational dilution begins
deposits, and 10% for tabular-shaped deposits [14]. The by characterizing each block planned to be extracted, and its
dilution is not always an analysed and systematically neighbourhood. Depending on the block position (inside the
2
Câmara et al / Revista DYNA, 87(213), pp. 178-183, April - June, 2020.
mine plans or touching the borders of mine plan polygons), To calculate the diluted grade due to the influence of bench
dilution will be calculated differently. For blocks located face angle (eq. 4), the mass of the triangular prism designated
inside the polygon, dilution will be calculated based on the as dilution (eq. 2) will be added to the total block mass and
premise that equipment imprecision will keep the blocks the mass of the triangular prism designated ore loss (eq. 3)
from being extracted perfectly. If the contact between blocks will be subtracted from it [16].
is ore-waste, each portion of the waste blocks in contact with
the analysed block will be incorporated with it (Fig. 1). As a 𝑥 𝑍𝐼𝑁𝐶
(2 × 2 ) × 𝑌𝐼𝑁𝐶 × 𝐷𝑐𝑜𝑛𝑡 (2)
result, the diluted grade will be weighed considering the new Dilution = × 𝐵𝐺𝑐𝑜𝑛𝑡
2
mass (eq. 1) incorporated in the ore block. The ore loss is not
considered at this stage because the methodology assumes Where,
that it is better to recover more ore tonnage, even with the 𝑥 is the base of the triangular prism formed due to the bench
consequence of generating higher dilution, than having ore face angle;
losses [16]. Crawford [21] also showed that reducing dilution 𝑌𝐼𝑁𝐶 and 𝑍𝐼𝑁𝐶 are the block dimensions in Y and Z,
through more selective mining often resulted in higher ore respectively;
losses. 𝐷𝑐𝑜𝑛𝑡 is the density of the contact block;
(𝐵𝑀𝑜 × 𝐵𝐺𝑜 ) + (𝐵𝑀𝑎𝑑𝑗 × 𝐵𝐺𝑤 )
Diluted Grade = (1) 𝐵𝐺𝑐𝑜𝑛𝑡 is the contact block grade;
(𝐵𝑀𝑜 + 𝐵𝑀𝑎𝑑𝑗 )
Where, 𝑥 𝑍𝐼𝑁𝐶
( 2 × 2 ) × 𝑌𝐼𝑁𝐶 × 𝐷𝑜 (3)
𝐵𝑀𝑜 is the ore block mass; Ore loss = × 𝐵𝐺𝑜
𝐵𝐺𝑜 is the ore block grade; 2
𝐵𝑀𝑎𝑑𝑗 is the adjacent mass portion; Where,
𝐵𝐺𝑤 is the waste block grade; 𝐷𝑜 is the density of the ore block;
Where,
𝐵𝑀𝑡𝑜𝑡𝑎𝑙 is the total block mass;
3
Câmara et al / Revista DYNA, 87(213), pp. 178-183, April - June, 2020.
Figure 4. 8th month: P90 of dilution (right) and the rock type (left).
Source: The authors.
In the 6th, 10th and 11th months there was a large variation
in the dilution results with values considerably lower than the
maximum values. This fact indicates that these very distinct
values may have occurred in isolated areas. Fig. 5 shows the
Figure 3. Box plot showing the dilution distribution considering 30 grade blocks planned to be extracted in the 10th month (right side -
simulations per month. dilution legend (P90); left side - rock types). In contrast to the
Source: The authors. previous analysis (of the 8th month), the areas with the highest
dilution were in this case concentrated at specific points,
As each analysed scenario presents a possible distribution
which leads to the conclusion that it is possible to correct the
of dilution values, to visually analyse the results, a legend
mining plan at these specific locations.
was created based on the 90th percentile (P90) of 30
simulations. The value of P90 was chosen because it means
that 90% of results were lower than this value, i.e., this
scenario would be considered conservative.
4
Câmara et al / Revista DYNA, 87(213), pp. 178-183, April - June, 2020.
4 Conclusions References
Usually, the block grades are diluted by a given fixed [1] Blom, M., Pearce, A. R. and Stuckey, P. J. Short-term planning for
value, considering both operational dilution and dilution open pit mines: a review, International Journal of Mining, Reclamation
related to misclassification of blocks. Quantifying and Environment. Taylor & Francis, 0930, pp. 1–22, 2018. doi:
10.1080/17480930.2018.1448248.
operational dilution through techniques applied to block
models and short-term mine plans can better control dilution [2] Dimitrakopoulos, R., Farrelly, C. T. and Godoy, M. Moving forward
from traditional optimization: grade uncertainty and risk effects in
than merely applying a best-guess factor to the block model. open-pit design, Mining Technology, 111(1), pp. 82–88, 2002. doi:
Through the results of operational dilution calculated 10.1179/mnt.2002.111.1.82.
using the information of each simulated model, it was [3] Dimitrakopoulos, R. Strategic mine planning under uncertainty,
possible to understand the probability distribution of dilution, Journal of Mining Science, 47(2), pp. 138–150, 2011. doi:
allowing the verification of the areas that were most prone to 10.1134/S1062739147020018.
suffering dilution. With this information, the mine plan can [4] Lerchs, H. and Grossmann, I. Optimum design of open-pit mines.
be directed to control these areas through more selective dig Canadian Mining and Metallurgical Bulletin, LXVIII:17–24, January
1965.
lines or a different short-term schedule, aiming to reduce the
dilution effect. [5] Hochbaum, D. A new-old algorithm for minimum cut in closure
graphs. Networks, 37(4), Anniversary Issue 2001, pp. 171-193, 2001.
It is important to compare the use of information from doi: 10.1002/net.1012
deterministic block models and that of information from
[6] Johnson, T. Optimum open pit mine production scheduling. In A.
stochastic block models. It is known that deterministic block Weiss, editor, A Decade of Digital Computing in the Mining Industry,
models, usually estimated by ordinary kriging, result in chapter 4. American Institute of Mining Engineers, New York, 1969.
smoothing. This drawback can be overcome by using [7] Torres, V. F. N. et al. Classical and stochastic mine planning
stochastic block models that guarantee global accuracy. In techniques, state of the art and trends, REM - International Engineering
the case study presented, the use of stochastic models gives a Journal, 71(2), pp. 289–297, 2018. doi: 10.1590/0370-
higher-confidence evaluation of uncertainty, as the result is 44672016710165.
obtained through a probabilistic analysis and is not limited to [8] Birge, J. R. and Louveaux, F. Introduction to stochastic programming.
Springer Science & Business Media. 2011.
a deterministic single value providing the average value and
the interval that it can vary within. [9] Dare-Bryan, P. C. and Dowd, P. A. Assessing Ore Loss and Ore
Dilution — Quantifying the Effects of Geological and Technical
In addition to the block grades, density is another Uncertainties, in PACRIM 2004, pp. 19–22, 2004.
parameter that could have a major impact on the dilution
[10] Spleit, M. Stochastic Long-Term Production Scheduling of the
result. If there is great variability in the density values, this LabMag Iron Ore Deposit in Labrador, Canada. M. S. Thesis. Mc Gill
attribute must also be simulated, to ensure greater precision University, Montreal, Canada, 2014.
in the dilution results. However, for this case study, there was [11] Zarshenas, Y. and Saeedi, G. Risk assessment of dilution in open pit
no need to simulate the density values, since the they were mines, Arabian Journal of Geosciences, 9(209), p. 1-11, 2016. doi:
very similar in the blocks, which would result in a non- 10.1007/s12517-015-2214-8.
significant impact to dilution control. [12] Matthews, T. Dilution and ore loss projections: strategies and
The approach presented in this study demonstrates that it considerations, in SME Annual Meeting. Denver, Colorado, pp. 1–4,
2015.
is possible to measure and control operational dilution values
resulting from a given mine plan. However, how this [13] Xingwana, L. Monitoring ore loss and dilution for mine-to-mill
integration in deep gold mines: A survey-based investigation, Journal
information will be used will depend on each specific of the Southern African Institute of Mining and Metallurgy, 116(2), pp.
situation. If the choice is to reduce dilution as much as 149–160, 2016. doi: 10.17159/2411-9717/2016/v116n2a6.
possible, the consequence will be lower productivity, which [14] Ebrahimi, A. The importance of dilution factor for open pit mining
can be quite detrimental to the operation, depending on the projects. 2013. Available at:
orebody type. If it is not possible to increase the selectivity http://www.srk.com.au/files/File/papers/dilution_factor_openpit_a_eb
of the operation, the obtained results can be used as an rahimi.pdf.
adjustment factor to the plans, resulting in better adherence [15] Câmara, T. R. Sistematização do cálculo de diluição e perdas
of the planned data to those actually executed. operacionais para reconciliação de teores e massas em lavra a céu
aberto. M.S. Thesis. Mining Engineering Department, Federal
The dilution calculated in this study is operational, University of Rio Grande do Sul, Porto Alegre, Brasil, 2013.
meaning that the observed values are not particularly low;
[16] Câmara, T. R. et al. Controlling Operational Dilution in Open-Pit
generally, the dilution used in mine plans will take into Mining, Mining Technology, 128:1, pp. 1-8, 2018. doi:
account factors other than operational. For this reason, it is 10.1080/25726668.2018.1470275.
important to try to minimize the operational dilution as much [17] Rossi, M. E. Accounting for dilution in ore resource estimation, in
as possible (or feasible, given the operating conditions and APCOM 2009. Vancouver, Canada, pp. 84–92, 2009.
geometry of the orebody), since among the dilution types, [18] Bertinshaw, R. and Lipton, I. Estimating Mining Factors (Dilution and
this is one that can be controlled. Ore Loss) in Open Pit Mines, Large Open Pit Mining Conference,
(September), pp. 13–17, 2007.
[19] Roldão, D. et al. Combined Use of Lithological and Grade Simulations
for Risk Analysis in Iron Ore, Brazil, in Geostatistics Oslo 2012, pp.
423–434, 2012. doi: 10.1007/978-94-007-4153-9
5
Câmara et al / Revista DYNA, 87(213), pp. 178-183, April - June, 2020.
6
CAPÍTULO 3
3. CONCLUSÕES E RECOMENDAÇÕES
O objetivo desta tese foi desenvolver uma rotina para o cálculo da diluição operacional
em lavra a céu aberto utilizando informações de planos de curto prazo e modelos de blocos
contendo informações de massas e teores. Os três artigos que compõem a tese demonstram que
os objetivos previamente definidos no primeiro capítulo foram cumpridos:
i. Foi desenvolvido um procedimento para determinar a diluição causada pela
ineficiência de execução da operação de lavra.
ii. Tal procedimento foi automatizado em software comercial de planejamento de
lavra em um primeiro momento e, posteriormente, atualizado para uma rotina
em linguagem Python, para que tal tarefa pudesse ser executada rapidamente e
ter rastreabilidade.
iii. A metodologia foi aplicada em três diferentes tipos de depósitos considerando
como input tanto informações determinísticas (modelos estimados) quanto
informações estocásticas (modelos equiprováveis simulados).
iv. O valor da diluição operacional foi quantificado de duas diferentes formas: valor
médio para cada período e distribuição da probabilidade de ocorrência de
diluição por período nas áreas analisadas.
41
3.1 Conclusões
Analisando os resultados apresentados nos estudos de caso deste trabalho, fica claro
que o uso de um número constante para expressar diluição na operação não é a abordagem ideal
a ser utilizada, considerando que a diluição depende de vários fatores e que pode ter grande
oscilação entre diferentes períodos. Como regra geral, para períodos maiores como anos ou
semestres, pode ser mais fácil estimar a diluição. Todavia, a operação precisa lidar com a
produção mensal, semanal ou mesmo diária; logo, é fundamental fornecer uma maneira de
entender e explicar a fonte das diferenças entre valores planejados e dados obtidos na produção.
Os resultados obtidos na aplicação da metodologia demonstraram que a diluição teve bastante
variação para cada período analisado e que, em alguns casos, não foram valores baixos que
pudessem ser ignorados ou desconsiderados. Geralmente, a diluição usada nos planos de mina
leva em consideração outros fatores além dos operacionais, por esse motivo, é importante tentar
minimizar a diluição operacional o quanto for possível, dadas as condições de operação e a
geometria do corpo de minério, uma vez que, entre os tipos de diluição, a diluição operacional
pode ser controlada.
Para que a metodologia apresentada neste trabalho possa ter seu uso e aplicação
comprovados na esfera operacional, é sugerida uma etapa de reconciliação dos dados de
produção versus os dados obtidos com a diluição operacional aplicada. Esta etapa seria uma
validação da informação teórica com a informação prática, contribuindo para que a metodologia
tenha aplicação prática em operações de mina.
Uma abordagem adicional que pode ser desenvolvida é a análise de sensibilidade dos
parâmetros utilizados como dados de entrada nos estudos de caso aqui apresentados, como por
exemplo: o tamanho dos blocos, o desenho das poligonais de planejamento e o ângulo de face
da bancada. Desta forma, seria possível verificar o impacto de cada uma das variáveis na
quantificação da diluição operacional, o que serviria como ponto de partida para que o
engenheiro de planejamento possa corrigir e diminuir a diluição operacional de curto prazo.
Algoritmos de detecção de borda também poderiam ser aplicados na definição das poligonais
de planejamento, demonstrando ser outra ferramenta de grande contribuição na área de estudo.
Outro trabalho que pode ser elaborado utilizando modelos de incertezas de teores e a
metodologia de quantificação da diluição operacional, é a correlação das informações de
probabilidade de incerteza dos teores dos blocos (e consequente propensão a ocorrer mais ou
menos diluição) com o lucro associado a cada bloco a ser lavrado. O trabalho de Neufeld,
43
Norrena e Deutsch (2005) apresenta uma metodologia onde os modelos de incertezas de teores
são transformados em modelos de lucro esperado, levando em consideração questões
operacionais, recuperação do processo, custos e preço de venda. A partir destes modelos de
lucro esperado é feita a classificação dos blocos em minério e estéril. Utilizando tal metodologia
associada ao estudo apresentado nesta tese, a variação da diluição pode ser mensurada, uma vez
que a mesma só é calculada para contatos de blocos de minério com blocos de estéril após a
classificação. Esta abordagem também poderia contribuir para o melhor aproveitamento de
blocos com classificação imprecisa devido à incerteza dos teores, como minério marginal, por
exemplo.
44
REFERÊNCIAS
AMIRÁ, R.; MORALES, N.; CÁCERES, A. Analysis of the impact of the dilution on the
planning of open-pit mines for highly structural veined-shaped bodies. In: 27th
INTERNATIONAL SYMPOSIUM ON MINE PLANNING AND EQUIPMENT
SELECTION-MPES 2018, 27., 2018, Santiago, Chile. Proceedings... Springer, Cham: 2019.
p. 187–196.
BERTINSHAW, R.; LIPTON, I. Estimating mining factors (dilution and ore loss) in open pit
mines. In: 6th LARGE OPEN PIT MINING CONFERENCE, 6., 2007, Perth, WA.
Proceedings...2007. p. 13–17.
BEST, J.; GALLANT, K. Reconciliation wars – Who’s data is right? In: SME ANNUAL
MEETING, 2004, Denver, Colorado. Proceedings...2004. p. 1–5.
45
em Engenharia de Minas, Materiais e Metalúrgica, Universidade Federal do Rio Grande do Sul,
Porto Alegre, 2012.
DAGASAN, Y. et al. Pilot point optimization of mining boundaries for lateritic metal deposits:
finding the trade-off between dilution and ore loss. Natural Resources Research, v. 28, n. 1,
p. 153–171, 2019.
DOMINGO, J. et al. Dilution, ore grade and blast movement calculation model. In: 11th
INTERNATIONAL SYMPOSIUM ON ROCK FRAGMENTATION BY BLASTING, 2015,
Sydney, Austrália. Proceedings... Melbourne: The Australasian Institute of Mining and
Metallurgy, 2015. p. 71–78.
DOMINY, S. C.; NOPPÉ, M. A.; ANNELS, A.E. Errors and uncertainty in mineral resource
and ore reserve estimation: the importance of getting it right. Exploration and Mining
Geology, v. 11, n. 1–4, p. 77–98, Fev. 2004.
EBRAHIMI, A. The importance of dilution factor for open pit mining projects. In: 23rd
WORLD MINING CONGRESS - WMC, 2013, Montreal, Canadá. Proceedings... 2013. p. 1–
12.
ELBROND, J. Economic effect of ore loss and rock dilution. CIM Bulletin, v. 87, p. 1–7, 1994.
ESHUN, P. A.; DZIGBORDI, K. A. Control of ore loss and dilution at AngloGold Ashanti,
Iduapriem mine using Blast Movement Monitoring System. Ghana Mining Journal, v. 16, n.
46
1, p. 49–59, 2016.
FOUET, T. et al. Standardising the reconciliation factors required in governance reporting. In:
7th INTERNATIONAL MINING GEOLOGY CONFERENCE, 2009, Perth, WA.
Proceedings... Melbourne: The Australasian Institute of Mining and Metallurgy, 2009. p. 127–
139.
MATTHEWS, T. Dilution and ore loss projections: Strategies and considerations. In: SME
ANNUAL MEETING, 2015, Denver, Colorado. Proceedings...2015. p. 1–4.
MORLEY, C. Beyond reconciliation — a proactive approach to using mining data. IN: 5th
LARGE OPEN PIT MINING CONFERENCE, 2003, Kalgoorlie, WA. Proceedings...2003, p.
185–192.
NOPPÉ, M. Reconciliation: importance of good sampling and data QA-QC. In: MINING AND
RESOURCE GEOLOGY SYMPOSIUM, EGRU Contribution, n. 62, 2004, Queensland,
Austrália. Proceedings... Queensland: 2004. p. 1–7.
PITARD, F. F. A strategy to minimise ore grade reconciliation problems between the mine and
the mill. In: MINE TO MILL, 1998, Brisbane, Australia. Proceedings...Brisbane: 1998. p. 77–
82.
47
Geosciences and Geomatics, v. 3, n. 5, p. 116–121, 2015.
RISKE, R.; MARK, P. DE; HELM, S. Mine reconciliation – for better or worse! In:
PROCEEDINGS THE CANADIAN INSTITUTE MINING METALLURGY AND
PETROLEUM (CIM), 2010, Vancouver, Canada. Proceedings...Vancouver, Canada: 2010. p.
1 – 8.
ROLDÃO, D. et al. Combined use of lithological and grade simulations for risk analysis in iron
ore, Brazil. In: GEOSTATISTICS OSLO, 2012, Oslo, Noruega. Proceedings... Oslo: 2012. p.
423–434.
ROSSI, M. E. Accounting for dilution in ore resource estimation. In: APCOM, 2009,
Vancouver, Canada. Proceedings...Vancouver, Canada: 2009. p. 84–92.
SCHOFIELD, N. The myth of mine reconciliation. In: MINE TO MILL, 1998, Brisbane,
Australia. Proceedings...Brisbane: 1998. p. 11–14.
XINGWANA, L. Monitoring ore loss and dilution for mine-to-mill integration in deep gold
mines: A survey-based investigation. Journal of the Southern African Institute of Mining
and Metallurgy, v. 116, n. 2, p. 149–160, Fev. 2016.
ZARSHENAS, Y.; SAEEDI, G. Risk assessment of dilution in open pit mines. Arabian
Journal of Geosciences, v. 9, n. 3, p. 1–11, Mar. 2016.
48
APÊNDICE A - OUTRAS PUBLICAÇÕES CIENTÍFICAS – PERIÓDICOS
Chaves, L.S.; Carvalho, L.A.; Souza, F. R.; Nader, A. S.; Ortiz, C.E.A.; Navarro-Torres,
V. F.; Câmara, T. R.; Napa-Garcia, G. F.; Valadão, G.E.S. Analysis of the impacts of
slope angle variation on slope stability and NPV via two different final pit definition
techniques. REM - International Engineering Journal, v. 73, p. 119-126, 2020.
DOI: https://doi.org/10.1590/0370-44672019730011
Souza, F. R.; Câmara, T. R.; Navarro-Torres, V. F.; Nader, A. S.; Galery, R. Mine fleet
cost evaluation - Dijkstra's optimized path. REM - International Engineering
Journal, v. 72, p. 321-328, 2019. DOI: https://doi.org/10.1590/0370-44672018720124
Souza, F. R.; Câmara, T. R.; Navarro-Torres, V. F.; Nader, A. S.; Galery, R. Optimum
mine production rate based on price uncertainty. REM - International Engineering
Journal, v. 72, p. 625-634, 2019. DOI: https://doi.org/10.1590/0370-44672018720093
Souza, F. R.; Burgarelli, H.R.; Nader, A. S.; Ortiz, C.E.A.; Chaves, L.S.; Carvalho,
L.A.; Navarro-Torres, V. F.; Câmara, T. R.; Galery, R. Direct block scheduling
technology: analysis of avidity. REM - International Engineering Journal, v. 71, p.
97-104, 2018. DOI: https://doi.org/10.1590/0370-44672017710129
Burgarelli, H.R.; Souza, F. R.; Nader, A. S.; Navarro-Torres, V. F.; Câmara, T. R.;
Ortiz, C.E.A.; Galery, R. Direct block scheduling under marketing uncertainties. REM
49
- International Engineering Journal, v. 71, p. 275-280, 2018. DOI:
https://doi.org/10.1590/0370-44672017710128
Navarro-Torres, V. F.; Nader, A. S.; Ortiz, C.E.A.; Souza, F. R.; Burgarelli, H.R.;
Chaves, L.S.; Carvalho, L.A.; Câmara, T. R.; Zanetti, E.; Galery, R. Classical and
stochastic mine planning techniques, state of the art and trends. REM - International
Engineering Journal, v. 71, p. 289-297, 2018. DOI: https://doi.org/10.1590/0370-
44672016710165
Carvalho, L.A.; Souza, F. R.; Chaves, L.S.; Nader, A. S.; Câmara, T. R.; Navarro-
Torres, V. F.; Galery, R. Impact of grade distribution on the final pit limit definition.
REM - International Engineering Journal, v. 71, p. 621-629, 2018. DOI:
https://doi.org/10.1590/0370-44672017710150
50
APÊNDICE B - OUTRAS PUBLICAÇÕES CIENTÍFICAS – CONGRESSOS
Napa-Garcia, G. F.; Navarro-Torres, V. F.; Tropia, I.R.; Capelli, R.B.; Câmara, T. R.;
Three-dimensional integral modeling of large open-pit slopes: an innovative stability
analysis. In: MPES 2018, 2018, Santiago. 27th MINE PLANNING AND EQUIPMENT
SELECTION INTERNATIONAL SYMPOSIUM, 2018.
Câmara, T.R.; Leal, R.S.; Peroni, R.L.; Capponi, L.N. Dilution caused by mine
execution inefficiency in open pit mining. In: 24th WORLD MINING CONGRESS,
2016, Rio de Janeiro. 24th WORLD MINING CONGRESS PROCEEDINGS, 2016. p.
68-75.
Câmara, T.R.; Kuckartz, B.T.; Peroni, R.L.; Fontoura, D.M.; Capponi, L.N.
Reconciliação: ferramentas de avaliação do fator mina/usina. In: 8º CONGRESSO
BRASILEIRO DE MINA A CÉU ABERTO, 2014, Belo Horizonte. 8º Congresso
Brasileiro de Mina a Céu Aberto, 2014. v. 1.
51