The current issue and full text archive of this journal is available on Emerald Insight at:

www.emeraldinsight.com/1012-8255.htm

Archetypes
Prospective study using and system
archetypes and system dynamics
Estudio prospectivo utilizando 181

arquetipos y dinámica de sistemas Received 7 May 2017

Revised 22 December 2017
19 November 2018
Patricio Vera and Christopher Nikulin Accepted 19 November 2018

Department of Engineering in Design of Products,

Federico Santa María Technical University, Valparaiso, Chile
Monica Lopez-Campos
Department of Industrial Engineering,
Federico Santa María Technical University, Valparaiso, Chile, and
Rosa Guadalupe G. Gonzalez Ramirez
Faculty of Engineering and Applied Science, Universidad de los Andes, Santiago, Chile

Abstract
Purpose – The purpose of this paper is to propose a combination of forecasting methods that enables a
holistic understanding of a future situation, given certain influencing variables by a combination of real data
and expert knowledge.
Design/methodology/approach – The proposal combines two well-known methods: first, system
archetypes that correspond to generic structures, allowing us to handle model management issues, and
second, system dynamics that offers technical support on a computational level to assess different scenarios
or problem solutions.
Findings – The case study considers the situation of the mining industry in Chile and its related variables,
including four different scenarios. Based on the proposed methodology, the results indicate that: first, the
price of copper is paramount for the industry and its effects are not limited to company profits; second, a long
period of downfall in copper prices could halt exploration and development projects.
Research limitations/implications – Systemic archetypes are still a subject of research and their
application in different fields of knowledge continues to increase to improve this simulation approach.
Practical implications – The case study illustrates the combination of a Vester matrix and initial system
archetype models that are enriched using the system dynamics approach. Indeed, the case study aims to
understand the consequences of different scenarios based on the problem-driven approach provided by Vester.
Social implications – The goal of prospective studies of large-scale and complex situations is to model the
real situation to obtain solutions that may enhance social welfare.
Originality/value – The proposed methodology contributes to the existing literature by integrating
techniques such as the Vester matrix, system archetype modelling and system dynamics simulation, all of
which were proposed previously in the literature as independent techniques.
Keywords Prospective, Archetypes, System dynamics, Scenarios analysis, Mining
Paper type Research paper

Resumen
Propósito – Este artículo propone una combinación de métodos de pronósticos que permite una
comprensión holística de una situación futura, dadas ciertas variables de influencia mediante una Academia Revista
combinación de datos reales y conocimiento de expertos. Latinoamericana de
Administración
Vol. 32 No. 2, 2019
pp. 181-202
This paper forms part of a special section “Production systems and supply chain management in Latin- © Emerald Publishing Limited
1012-8255
America in association with the international conference on production research – Americas 2016”. DOI 10.1108/ARLA-05-2017-0151
ARLA Diseño/metodología/enfoque – La propuesta combina dos métodos conocidos: (i) arquetipos de sistemas que
corresponden a estructuras genéricas, lo que nos permite manejar los modelos, y (ii) la dinámica de sistemas
32,2 que ofrece soporte técnico a nivel computacional para evaluar diferentes escenarios o soluciones de problemas.
Resultados – El caso de estudio considera la situación de la industria minera en Chile y sus variables
relacionadas, incluidos cuatro escenarios diferentes. Según la metodología propuesta, los resultados indican que
i) el precio del cobre es primordial para la industria y sus efectos no se limitan a las ganancias de la empresa;
ii) un largo período de caída en los precios del cobre podría detener los proyectos de exploración y desarrollo.
Limitaciones en la investigación/implicaciones – Los arquetipos sistémicos siguen siendo un tema de
182 investigación y su aplicación en diferentes campos del conocimiento continúa aumentando para mejorar este
enfoque de simulación.
Implicaciones prácticas – El estudio de caso ilustra la combinación de una matriz de Vester y los modelos
de arquetipos del sistema inicial que se enriquecen utilizando el enfoque de dinámica de sistemas. De hecho,
el caso de estudio apunta a comprender las consecuencias de diferentes escenarios basados en el enfoque
orientado a los problemas proporcionado por Vester.
Implicaciones sociales – El objetivo de los estudios prospectivos para situaciones de gran escala y
complejas es modelar la situación real para obtener soluciones que puedan mejorar el bienestar social.
Originalidad/valor – La metodología propuesta contribuye a la literatura existente mediante la integración
de técnicas como la matriz de Vester, el modelado de arquetipos del sistema y la simulación de dinámica de
sistemas, todo lo cual se propuso anteriormente en la literatura como técnicas independientes.
Palabras clave Prospección, Arquetipos, Dinámica de Sistemas, Análisis de Escenarios, Minería
Tipo de papel Trabajo de investigación

1. Introduction
Competent organisations must be able to make the best possible decisions and face potential
future scenarios in the best possible manner (Godet and Durance, 2007). For this task,
prospective analysis is an invaluable technique that can be used in several research fields
(Chung Pinzás, 2009). Prospective methods have become increasingly popular, given our
rapidly changing world, in many different fields of knowledge, including technology,
economic, social and environmental. In many cases, decision makers must constantly
upgrade the related information in their fields and manage large amounts of data
(Christensen, 2013). Consequently, decision analysis considering different time periods is
relevant for business success. In other words, analytical methods and decision support
systems are key elements that support decision makers.
To study different and complex future scenarios, the analysis is typically divided into
three phases (Bustamante, 2006). The first phase consists of understanding the current
situation in which most of the variables that influence the behaviour of an organisation are
identified. The second phase consists of understanding future scenarios by identifying the
more influential variables and their interrelations. Finally, the third phase consists of
generating the scenarios based on different hypotheses about the future (Bustamante, 2006).
This analysis process requires significant expert knowledge and active participation to
provide insights and support the process to determine the relations among the different
variables under consideration (Flores, 2013).
To achieve these goals and to obtain successful results in the prospective analysis some
well-known limitations and challenges must be mitigated. They include the following: first,
the possibility to create efficient and useful scenarios; second, the possibility to upgrade
data within the models; third, a proper adequacy or adaptation to different scenarios that are
not necessarily controlled (Flores, 2013); and the reduction of the effects of inexperience in
the modelling cases (Godet and Durance, 2007).
Given the limitations of traditional prospective methods, we herein propose a suitable
and generic methodology for the use of the Vester matrix combined with the archetype
system as an initial formalization of problems. They are then integrated into system
dynamics analysis that appears to create a more appropriate strategy to execute
prospective large-scale and complex situations. A Vester matrix aims to identify the
primary problem situations with a large set of variables, thus allowing for
the identification of the relevant variables in a model. Simultaneously, the system Archetypes
archetype technique is a quick simulation method, where representations of variables are and system
pre-established, enabling a model design that can accommodate the future (Senge, 1990).
In other words, system archetypes allow for the development of a causal model based on
modelled events or generic archetypes previously studied in the literature (Wolstenholme,
2003). This strategy accelerates the modelling process. When the model is created, it
is necessary to quantify its elements using real data, knowledge, experience and scenarios. 183
In this case, the systems dynamics method evaluates and controls the system archetypes,
aiming to create a new model in a language that can assess its variables and perform the
analysis of a complex environment (Aracil, 2005).
Finally, this paper proposes a methodology in which the Vester matrix is used to attempt
to obtain the most relevant variables in the context to be analysed (that are strictly related to
problems). Additionally, the system archetypes aim to create a causal model to develop more
complex scenarios by the integration of both variables: Vester matrix and expert knowledge.
Finally, the system dynamics technique evaluates and compares the modelling results.

2. Framework of system archetype and system dynamics

2.1 Systemic archetypes and relevance for the development of causal models
Systemic archetypes can be considered a representation of variables for problematic
situations, a regular issue in many different contexts. Each archetype is represented through a
causal cycle diagram that exhibits the structure derived from the problematic situation
associated with the archetype (Senge, 1990). It is typical to obtain different scenarios/graphics
in the literature that reveal the behavioural tendencies of some archetype variables and
management principles that facilitate the creation of public policies to mitigate the problem
(Senge, 1990). Although the archetypes are defined by a rigid structure, they are still dynamic
models that change according to contexts and priorities. As such, they can manage changes
and modifications to their structure (Aracil, 2005), making them adaptable to different cases.
To understand archetypes, it is fundamental to determine their key elements. In addition to
the structural elements (i.e. variables that are related), two other primary elements exist that
represent the relationships between the variables. On the one hand, the functioning is set
according to the context in which the work is being performed, implying that two archetypes
with identical structures differ depending on use (Mejía and Díaz, 2006). On the other hand, the
behaviour refers to the type of cycles assigned to the structures, whether the aim is to increase
or decrease any given variable (Mejía and Díaz, 2006). Owing to this condition of system
archetypes, Wolstenholme (2003) proposed that many problematic situations could be modelled
from an archetype composed of two causal loops, including some changes to represent a
function and a behaviour. Thus, archetypes can be realized as predefined structures that also
accept modifications to improve their usefulness and, in some cases, render them simpler.
A primary issue in the literature is the modelling approach for a problematic situation
that may allow for the generation of possible scenarios and potential solutions. For instance,
the modelling process using this approach is not limited to an archetype per problem, thus, it
may be necessary to integrate several structures into an appropriate model analysis. This
occurs when typical variables exist. As an example, the reader can refer to the model created
for the expansion of a hospital presented by Wolstenholme (2004), and to the model
proposed by Kim and Burchill (1992) for the organisation of companies. System archetypes
have been used for highly diverse situations such as software development (Aristizábal
et al., 2015), the definition of a collaborative structure in a telecommunications company
(Aristizábal et al., 2014), the analysis of a specific technique to be used in public
organisations (Alves et al., 2014) and the analysis of hydro-policies of water resources
(Bazrkar et al., 2013). Finally, the actual understanding of archetypes promotes their
usefulness and simplicity for the creation of complex models.
ARLA 2.2 System dynamics and scenario analysis
32,2 System dynamics is a technique used to analyse the behaviour of systems in time (system
archetypes can be included as well), and can be applied to different areas, such as
management, environment, politics, economy, health and engineering (Forrester, 1998). System
dynamics allows for the construction of simulation models in computers based on the causal
relations and feedback from the system elements, thus, it is a useful technique for
184 decision-making in complex environments (Aracil, 2005). Such models use different
approaches including linear regressions, multivariable analysis (Brandmaier et al., 2013)
and stochastic and probabilistic models (Canavos and Medal, 1987). System dynamics as a
modelling technique has been used for a long time; however, its application to large-scale
prospective problems is relatively new (Cascini et al., 2015). Several diverse contributions using
system dynamics in large-scale problems can be found in the literature. For instance, Wäger
et al. (2006) proposed a system dynamics model to assess the potential impact of information
and communication technologies on environmental sustainability in the European Union
within a time horizon that extends to 2020. Wäger et al. (2006) highlighted the advantages and
disadvantages of this technique, emphasising their weaknesses such as model repeatability,
data collection and integration of expert knowledge to create appropriate scenarios. More
recently, Li et al. (2017) created a system dynamics model in order to examine the potential CO2
emission trends of China’s primary aluminium industry over the next 15 years.
In this study, we aim to enhance the overall analysis through the integration of the
Vester matrix to identify problems, the use of system archetypes as an initial approach to
create models and finally, to use system dynamics to model and analyse the different
scenarios encountered by experts. This integration is novel and, because it is generic, it can
be used for any type of large-scale problem.

3. Methodology description for a systematic approach

In this section, the authors propose a methodology by following a step-by-step approach to
illustrate the structure of the proposal. Figure 1 presents the methodology used for the
creation of the model, and the scenarios are analysed subsequently. The proposal consists of
six steps described below.

3.1 Step 1: study of the situation and classification of information

The first step consists of studying the situation to be analysed, followed by the classification
of the information obtained to identify the potential problems or opportunity areas to create
a model of the current situation. Hence, an exhaustive examination of the situation should be
performed to identify the primary variables and problems. Simultaneously, this approach
supports and handles the further classification and synthesis of the data into different risk
factors (David, 2003) such as economic-financial, sociocultural, technological, environmental
and productive areas. This is to encompass, in a general and broad manner, the situations
where problems can arise within an organisation.

3.2 Step 2: prioritisation of variables for modelling

The first step yielded different issues that can be categorised into single variables or a set of
variables. In practice, the use of variables associated to problems allows us to understand the
causal relation of the current situation, making it necessary to prioritise and hierarchise those
problems. To do so, we propose the use of the Vester matrix (Cole, 2006). This is an intrinsic
relation matrix that can be used to prioritise the problems identified in Step 1. This method
allows for the identification of these problems, and the corresponding variables that cause
such problems. This step can be considered a formal approach to understanding the initial
model to be forecasted, which aims to consider the primary problems as the primary causes
that the model should forecast towards a scenario solution with time perspective. In particular,
 Step 1 Step 2 Archetypes
General and system
description Hierarchization of
Identification of
of the problems with
relevant problems
situation and relation matrix
analysis context

185
Classification of
Determination of
problems according
main problems
risk factors

Step 3 Step 4

Integration of Creation of a Creation of the Identify the

archetypes and causal model with system type of
variables model dynamics variables

Integration of
equations

Step 5

Behavioural
analysis
Model Validation

Structural
analysis

Step 6
Figure 1.
Simulation and
Creation of
scenario
Methodology to create
scenarios
analysis a structured and
repeatable system
dynamics analysis

this proposal uses primary problems as they ensure the creation of a base model that may be
used in most situations, owing to its link to all the problems and therefore to their possible
solutions. Finally, it must be highlighted that the problems and related variables obtained in
this step must allow for the creation of a model that can accommodate different situations.

3.3 Step 3: creation of a causal model based on system archetypes

In this step, a causal model to represent the current situation and the related scenarios is
created using archetype models from the literature to create a more complex model and its
corresponding variables. Hence, the stages described below must be adhered to:
• Stage 1: description of the pre-established systemic archetypes. The model can
combine archetypes both qualitatively and quantitatively, provided that a method to
evaluate them numerically exists.
ARLA • Stage 2: identification of the variables that influence the elements must be incorporated
32,2 into the previous stage directly, allowing for the completion of the model or to create
connections among archetypes and variables.
• Stage 3: previous stages are repeated iteratively until a stopping criterion is reached
(i.e. the maximum number of variables or appropriate influence of different variables
in the primary model).
186
3.4 Step 4: creation of a model using system dynamics
In this step, the system archetype is transformed into a system dynamics model that
consists of integrating the model from Step 3 with the concepts of system dynamics to
understand the data behaviour. The use of flow diagrams allows for the identification of the
gap between a causal model and a dynamic model such that both quantitative and
qualitative analyses can be understood. The following stages are recommended:
• Stage 1: characterize the variables according to their type and indicate whether they
correspond to the level of the model, flux, or auxiliary variables.
• Stage 2: specify the mathematical relations among the variables. In this stage, it is
important to identify the appropriate function that represents the variables, such as
mathematical functions and the inclusion of graphics, tables and equations.
• Stage 3: assign constant values through the simulation time line. The parameters are
not always known; therefore, in many cases, iterations are required. In cases in which
this information is non-existent, reasonable values are assigned based on expert
opinion to leverage real data with the primary hypotheses of the model.

3.5 Step 5: model validation

In this step, the structure, validity and quality of the model are assessed such that it can
represent the defined problematic situation. A variety of possible analyses can be performed
from the logical review of the causal diagrams to a review and comparison of the results.
Particularly, we propose a structural validation analysis and a behavioural validation
analysis (Santa Catalina, 2010).
3.5.1 Structural validation analysis. This validation considers the following aspects: the
first step is to establish whether the exposed relations are coherent, i.e. whether the model
explains the reality being assessed. These relations can be demonstrated frequently through
scientific and technical studies, or expert opinion on the subject. The second validation
considers the assessment of whether the equations of the system are technically correct,
implying that they must be consistent dimensionally, and that the variables and parameters
are specified clearly. If the model does not pass these tests, it will be declared invalid and
therefore of no use (Godoy and Bartó, 2002). If this analysis is not overcome, the model must
be re-evaluated from Step 3 to perform a new causal model analysis.
3.5.2 Behaviour validation analysis. The validation of a behaviour analysis allows for the
indirect analysis of the structure validity by testing the behaviour of the model in different
situations. For example, assigning extreme values for different parameters, and matching the
data output from the model with the expected behaviour or real data, emphasising the manner
in which data exist, and less so on the value provided by the model (Godoy and Bartó, 2002).
This step aims to calibrate the parameters and obtain the realistic results (Step 4). If the model
cannot be calibrated, it is necessary to repeat Step 3 to create the model again.

3.6 Step 6: creation and analysis of scenarios

The primary goal is to handle the uncertainty of the future, identifying or anticipating
changes, reducing future risks and obtaining effective strategies for the solution of the
problems based primarily on scenario creation. To create the scenarios, the behaviours of Archetypes
some model variables are defined for a certain period of time along with their related and system
strategies or policies. It is also possible to include solutions to certain problems as causal
cycles or through new variables corresponding to different archetypes. Following this, the
results provided by the model are analysed. It is recommended that at least two different
scenarios should be created, which must be plausible in order to make decisions that are
consistent with reality. They must be internally consistent and relevant to the case study 187
and propose different approaches to mitigate the problem addressed. Finally, after obtaining
the simulation results with the described scenarios, the analyst must assess these results
objectively in accordance with their purposes (Vergara Schmalbach et al., 2010).

4. Case study: the Chilean mining industry

4.1 Step 1: analysis of the current situation and data classification
In Chile, the mining industry is important to the economy because of its contribution to the
gross domestic product, to the creation of good quality jobs and to attracting high levels of
investment to the country (SONAMI and El Mercurio, 2012). It is noteworthy that the mining
industry has undergone different cycle-stages such as a boom (increasing trend) and decay
(decreasing trend) that have been caused by economic, environmental and productive
variables. Nevertheless, the increasing trend has not been exploited optimally to improve
productivity in order to better handle the decreasing trend when the value of copper is low.
In this context, the case study is a Chilean state-company dedicated to the exploitation of
copper, and the largest copper company in the world, with 7 per cent of the global reserves of
copper, and an estimated useful life for its deposits of approximately 65 more years. The
company also assumes the second position in terms of worldwide molybdenum production.
From 2005 to 2014, Chilean copper production suffered a constant decrease in production
rate, meaning that Chile’s participation in global copper production decreased (Comision
Chilena del Cobre, 2016). Nevertheless, production (in terms of quantity) has still increased
in Chile, but it has done so to a greater degree in many other countries. In Chile, production
has grown by 8 per cent; in Congo, it has grown by 956 per cent; 114 per cent in China;
72 per cent in Zambia; and 37 per cent in Peru (Marticorena, 2016). This is attributable to the
mature stage of the Chilean mining industry. The other countries are considered to be at a
stage similar to that experienced by Chile in the 90s. The existing mining projects cannot
emulate the growing production rates of the 90s, and new projects seem complicated, owing
to the high level of investments and the high prices of critical resources for the industry
(Marticorena, 2016). Even though the cost of critical elements for mining production was
reduced from 2014 to 2015, it did not compensate for the expenditure on materials, extras,
supplies and services, or for the decrease in the copper law, causing the average operational
costs to increase from 216 c/lb to 225 c/lb (Comisión Chilena del Cobre, 2016). The price of
copper has often fluctuated, especially between 2011 and 2015, where it changed from
399.66 c/lb to 243 c/lb (Comisión Chilena del Cobre, 2016). Previous situations partly explain
the reduction in the operating margin over sales in mining companies, i.e. close to 9 per cent
in 2015. These numbers reveal a significant difference considering that in 2014, the margin
over sales reached 32 per cent (Comisión Chilena del Cobre, 2015). In general, the effects
caused by lower incomes generate risks of reductions in the production, potential
disturbance, postponement of investment projects and hired workers, and a decline in the
shareholders’ expectations (Gonzáles, 2015).

4.2 Step 2: hierarchization of variables to create a base model

Once the situation under analysis is clear, and all problems or variables have been
recognised, the problems should be prioritised according to their category, either as central
ARLA problems, causes or consequences. The relevance of this step relies in the fact that the
32,2 creation of a base model starts from the central problems that have been found; therefore,
the model is representative of the numerous problematic situations arising in the Chilean
mining industry.
To analyse and develop this step, we selected the variables or central problems that
influence most of the situations affecting the mining company. This selection was made
188 using the Vester matrix method or relation matrix, which is a tool that allows for the
classification of problems and variables. Based on this classification, the primary problems
are used to create the base model, as they are related to all the other situations. This process
is essential for simplifying the model, as selection depends on variable comparisons using a
Likert scale based on a relation criterion (Table I).
It is possible to establish the relation matrix using the scale above. In this case study, 15
problems were identified, and the values of the relations were assigned to them (Table II).
The sum of all the scores, the sum of the rows that correspond to the value in the “x-axis”,
and the sum of the columns that correspond to the “y-axis”, complete the matrix.
As a summary of the Vester matrix, the different problems are presented in Figure 2.
Figure 2 displays the primary problems identified, as follows: production rates
(Problem 10), increase in production costs (Problem 11), and decrease in investment
(Problem 15). These problems are the basis of the new model leading to a more complex
system dynamics model. Another result of the Vester matrix is the possibility of obtaining
an estimation of the relationships among the problems (shown as an arrow in the model).
The initial model presented in Figure 3 exhibits negative loop behaviour because the
costs affect the investment negatively. This is important because it demonstrates a
steadier causal representation that tends to be balanced (Aracil and Gordillo, 1997).

Table I. Value Description

Likert scale for
relation criteria of 0 No cause
variables to be used in 1 Weak cause
the Vester matrix 2 Medium cause
analysis 3 Strong cause

No. Problem 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Result x-axis

1 Drop in copper price 0 2 0 0 0 0 0 0 0 0 0 0 0 0 3 5

2 Drop in productivity 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2
3 Social licences to operate 0 0 0 0 0 0 0 0 0 3 0 0 0 0 3 6
4 Nationalisation of the resource 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 2
5 Technology innovation 0 0 0 0 0 0 0 0 0 2 1 1 0 0 0 4
6 Organisational improvements 0 0 0 0 0 0 0 0 0 2 1 1 0 0 0 4
7 Energy resource scarcity 0 3 0 0 0 0 0 0 0 0 3 0 0 0 0 6
8 Hydric resource scarcity 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 3
9 Drop in the copper law 0 2 0 0 0 0 0 0 0 3 0 0 0 0 0 5
10 Production rates 0 3 0 0 0 0 0 0 0 0 2 0 0 0 2 7
11 Increase in production costs 0 3 0 0 0 0 0 0 0 0 0 0 0 0 3 6
12 Scale economics 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 3
13 Demand of copper 2 0 0 0 0 0 0 0 0 1 0 0 0 0 0 3
14 Offer of copper 2 0 0 0 0 0 0 0 0 1 0 0 0 0 0 3
Table II. 15 Investment in mining projects 0 0 0 0 2 2 0 0 0 2 0 0 0 0 0 6
Vester matrix Result y-axis 4 13 2 0 2 2 0 0 0 14 13 2 0 0 13
This balance is related to a specific goal; in this case, it corresponds to the production and Archetypes
how actions are to be conducted regarding the costs and investment for the productions or and system
future model variables.

4.3 Step 3: creation of a causal model based on system archetype

System archetypes have been used to create a causal model that may integrate many of the
already defined archetypes that can be performed by linking at least one variable among 189
them. The reference model used to create the initial system of archetypes towards system
dynamics modelling was the work presented by Naill (1992), where the relations were
established between investment and production for the extraction process of natural gas.
This model contains functional elements such as natural resource extraction and
behavioural elements owing to the relations among the involved variables; these elements
have been adapted for the case study herein addressed.
For the production section of the mining industry (Figure 5), the viability process of
resources and mineral reserves should be considered. These are limited to three primary
variables (Figure 4): the first is oIn SituW, implying all inferred resources that demonstrate
some degree of uncertainty in terms of their real features (i.e. potential available resources)
(Canfield, 2012). The second variable, oresourcesW is related to the mineral already
measured (Canfield, 2012). The amount of mineral that goes from oIn SituW to

Main Problems, Consequences and Causes

18
Consequences Main Problems
16
14
2 11
12
15 10
10
8
Figure 2.
6 Types of problems
6 Causes
4 12 1 emerging from the
4 Vester matrix,
2 14 5 3 classified according
0 8 13 9 7 their relevance
0 1 2 3 4 5 6 7 8 9

+ Production

Investment

–
Total Costs Figure 3.
Base model: a result
Note: The primary elements of the initial system archetype are of the Vester matrix
to create a system
represented in colour to understand their relevance in the overall archetype
system archetype model in Figure 5
ARLA Complete Reduced

32,2 Exploration
In situ
Results
Geological Information
Inferred Engineering
Discovery Results
Resources Results

Resources
190 Indicated Probable
Resources Reserves
Development
Results

Figure 4. Measured
Proven Reserves Reserves
General representation Resources
of resources and
reserves viability:
complete and reduced
in the mining industry The Modifying Factors (Costs, Prices, Environment, Social, The Modifying Factors (Costs,
Marketing, Legal, Others) Prices, Others)

oResourcesW depends on the odiscovery rateW that is the annual amount of copper
measured by the oexploration processW; it only occurs when othe investment decisionW
is approved. The third variable oreserveW is the entire mineral that has already been
verified and can be extracted (Canfield, 2012). This process depends on the odevelopment
rateW that corresponds to the annual amount of copper that becomes a oreserveW, and on
the conditions to oinvestW in the period. Finally, the oproductionW variable is related to
the oproduction capacityW that allows the achievement of the productive goals.
Simultaneously, some oexpansion of the capacityW are required that are translated as
delays between the need for an increase. Furthermore, because this factor is related to the
technology, its obsolescence time would imply a decrease in the capacity (Repenning and D,
2001). This is observed in Figure 5.
To control the oexplorationW, odevelopmentW and oproductionW, the variables
used are as follows: odesired explorationW, odesired developmentW and odesired
productionW, all of which aim to simulate the oproduction plansW adopted by mining
companies (Kim and Anderson, 1998). To complete the cycle, the oinvestmentW is related to
the ocostW and oproductionW cycles. In this step, the decision to invest in
oexplorationW or odevelopmentW depends on the relation among oincomesW,
ocostsW and orisks to investW in that moment (Naill, 1992). This is observed in Figure 5
(green and purple archetypes). Finally, to complement the causal model, a basic utility archetype
is included (Figure 5, blue archetype). The total variables are presented in Appendix 1.

4.4 Step 4: creation of the model with system dynamics

In this step, the model created in Step 3 is used to create an analogous model, by applying the
structure of a Forrester diagram, as the language of the software used for this part of the
process is Vensim (VENSIM Softare by Ventana Systems (http://vensim.com and
http://vensim.com/faq/)). First, it is necessary to assign each variable of the causal model
(Appendix 1), and to add the required mathematical relations and data, and the relations
regarding the operation of the model. It is noteworthy that the behaviour of the model is not
only issued by the established relations in the causal diagram, but also because the manner in
which each variable behaves is issued by the chosen equations and/or data tables.
Consequently, both the behaviour and relation of the variables cause fluctuations among the
case studies, even if different case studies belong to the same field. Appendix 1 presents each
of the variables, and the category and the equations used to represent the system dynamics
model. The details regarding system dynamics modelling are presented in Appendix 2.
 Adjust exploration Adjust exploration
need need

– –

Wanted Wanted
exploration Development

– + – –

In situ Discovery rate Resources Development rate Reserves Production

+ + + + +
+

– Limit of
Decision to investment in – Desired
Limit of Decision to invest
investment in development in development production
investment in exploration
exploration

+ Copper price Copper pricere

+ – –
Exploration Development
relation relation Production Reduction of
+ Capacity increase
capacity capacity
– – – – –
Total costs of +
exploration +

+ Total costs of
development Incomes
Unit cost of
development
Unit cost of +
development + +

Profit
Copper price

+ +
Unit cost of –
production
Total costs

+ Total cost of all

the production
+
Archetypes

191
and system

on the cycles of
Causal model based
Figure 5.

detailed system
production, total cost

archetype model)
and investment (more
ARLA 4.5 Step 5: model validation
32,2 Finally, the model is reviewed and compared with the others in terms of the formulas and
behaviour of the general model. The results of this revision correspond to Steps 2 and 3.
For behaviour validation, this step considers two aspects. The first considers the calibration
of those parameters, about which quantitative data and information are not available, e.g.
the unit development cost, development investment limit and exploration investment limit
192 (Figure 5) (Appendix 1). The second aspect is the comparison of the results of the model, by
contrasting the behaviour of the output (oincomes W and oprofitW) with the sales
obtained by the Chilean company. On the one hand, the variable oIncomes W is assessed
because a similar behaviour implies that the production stage delivers values and
magnitudes close to the real ones. On the other hand, the oprofits W are analysed because
this variable considers the different costs and therefore the investment decisions; a similar
behaviour implies that the remaining archetypes are still fulfilling their purpose.
In the income graph (Figure 6), the model exhibited a similar behaviour to the real
performance, thus corresponding to the expected representation owing to the income
variable being related primarily to the value of the price that exists at that moment, and
their values that are the average for each year. Therefore, the differences that exist currently
are caused by the chosen production plan. Between the years 2005–2008, the most
significant differences appeared, where production was partially dismissed. The value for
the 2008 to 2015 period is similar.
Regarding the profits (Figure 7), differences are found between the real and simulated
data that correspond to extreme situations simulated in years 2009 and 2011. These
differences can occur because of the differences between the costs defined by the model that

Incomes
2E + 10
1.8E + 10
1.6E + 10
1.4E + 10
1.2E + 10
1E + 10
8E + 09
6E + 09
Figure 6. 4E + 09
Comparison between 2E + 09
real incomes and 0
modelled incomes 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Model Real

Profit
1E + 10
9E + 09
8E + 09
7E + 09
6E + 09
5E + 09
4E + 09
3E + 09
2E + 09
Figure 7.
Comparison between 1E + 09
real profit and 0
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
modelled profit
Model Real
are related to exploration, development and production. These costs may differ from those Archetypes
considered by the Chilean company, even though the expected behaviour of the model is and system
maintained, implying the existence of cycles where the profits increase, and decrease in
similar magnitudes according to the data. Finally, the model is proven structurally valid,
because Steps 2 and 3 were carried out using systemic archetypes already used by other
authors (Canfield, 2012). Furthermore, the variables and parameters developed to integrate
the activity of the mining company are coherent to the model. Meanwhile, the behaviour 193
validation step delivers trends and data behaviours which are close to the real ones.

4.6 Step 6: creation of scenarios and results analysis

In this step, the model is already validated, and different analyses are performed. In the
proposed analysis, the situations of national interest for the future of mining are considered,
based on a study by the Chilean government (Ministry of Mining (www.minmineria.gob.cl/),
CORFO (https://corfo.cl/sites/cpp/home) and SONAMI and El Mercurio, 2012), in which two
aspects were considered: the role of governance and the copper market in the 2015–2025 period.
For the first analysis, governance is understood as the relation between industry and
government, between government and civil society, and how they affect the decisions
regarding investment in exploration and development in the mining industry (Mella Polanco
and Berrios Silva, 2013). A ohigh governance W is declared when government institutions,
industry, and community cooperate in a system characterised by transparency, openness,
and respect for a democratic process. The oemerging problems W are solved in a
consensual and collaborative manner where the path to follow is established considering
economic, social and environmental aspects. In general terms of investment decisions, the
risk of losing the investment in this analysis is minimal. However, a olow governance W
results in governmental institutions with poor transparency and dishonesty according to
civil society. Social pressures on companies cause mining projects to develop slowly or even
stop if necessary; in other words, halting investment decisions becomes even more
demanding given the current risk.
In the second analysis, the ocopper marketW responds to a set of factors, such as
demand, offer, new projects, price volatility, relation to other metals, substitute products,
politics and costs (Donoso Muñoz, 2013). As a response to this set of elements, the price of
copper is considered as an indicator of the market status. A ostrong copper market W is
considered by a high demand and a reduced offer at high prices, in which its demand is a
consequence of highly developing countries such as India, China and some African
countries. The price of copper has increased and remained high after its collapse in 2015.
Emerging countries such as Chile, exhibit a slow economic development, causing an
oversupply of copper owing to the high values reached in 2008 and 2011. In addition,
scientific and technological research to develop substitute products affects the price of
copper. Based on the previous description, the scenarios indicated in Figure 8 and Table III
are analysed.
In sum, the copper market is represented by the price of metal, and governance is related
to the limits of exploration and development. The behaviour of each variable is considered
linear, with the limit values presented in Table III.
4.6.1 Results and discussion. In the scenario analysis of the causal model, many variables
have been considered; however, the ones that are affected directly by the previous
configuration are oprofits W, oexploration W and odevelopment projects W associated
to each year. The results generated by each scenario are shown in Figure 9(a) and Figure 10.
Scenario 1 indicates that the high market and governmental conditions promote the
implementation of mining, primarily because the mining company is composed of mature
mines that constitute large reserves, thus overcoming the lack of exploration. Owing to
the decline in the price of copper, in Scenarios 3 and 4, the profits have decreased;
ARLA Copper Market
32,2 Strong

Scenario 2 Scenario 1
<Price increasing> <Price increasing>
<Motivation for <Motivation for
194 investment increasing> investment decreasing>

Governance Governance
Low High
Scenario 3 Scenario 4
<Price decreasing> <Price decreasing>
Figure 8. <Motivation for <Motivation for
Detailed description of investment increasing> investment decreasing>
four primary
scenarios to forecast
the future of the
mining industry Copper Market
Weak

Different scenarios to model Price Exploration limit Development limit

Table III. Year 2015 5,700 1.1 1.05

Initial values Scenario 1 8,500 1 1
associated with each Scenario 2 8,500 1.75 1.4
scenario to Scenario 3 4,000 1.75 1.4
understand the future Scenario 4 4,000 1 1

Scenario 1 Scenario 1
6E + 09 1.2

5E + 09 1

4E + 09 0.8

3E + 09 0.6

2E + 09 0.4
Exploration Projects
1E + 09 0.2
Profits Development Projects
0 0
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025

Years Years

Scenario 2 Scenario 2
1.2
6E + 09
1
5E + 09

4E + 09 0.8

3E + 09 0.6

Figure 9. 2E + 09 0.4
Exploration Projects
Scenarios 1 and 2 1E + 09
Profits
0.2
Development Projects
results (profit and 0
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
0
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
projects Years Years
implementation)
Note: More details in the results appendix

meanwhile, in Scenario 3, projects begin to stop in 2017, a situation that would completely
stop the operation of mines in 2023. In Scenario 4, this will occur in year 2020, in which the
production can be continued at least up to 2025, even though the operation of the company
implies a loss in profit starting from 2022.
 5E + 09
Scenario 3
1.2
Scenario 3
Exploration Projects
Archetypes
Profits
4E + 09 1

0.8
Development Projects
and system
3E + 09
0.6
2E + 09
0.4
1E + 09 0.2

0 0
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
–1E + 09
Years Years 195
Scenario 4 Scenario 4
5E + 09 1.2
Profits Exploration Projects
4E + 09 1
Development Projects
3E + 09 0.8

2E + 09 0.6

1E + 09 0.4
Figure 10.
0
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
0.2 Scenarios 3 and 4
–1E + 09 0
2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025
results (profit
–2E + 09
Years
and projects
Years
implementation)
Note: More details in the results appendix

The results indicate the importance of the price of copper in the mining industry, which does
not only affect the profit; a large period of downfall could lead to the ceasing of exploration
and development projects. In this situation, governance can serve as a cushion/buffer, albeit
for only a few years to maintain mining companies if the price is low, as this is not a
sustainable long-term solution. In relation to low governability, after several years, this
could stop company production. Finally, this type of model is generally used to understand
the general behaviour of a situation, rather than a strictly exact forecast.

5. Conclusions
A methodological proposal aimed at designing a model to explain complex scenarios within
a system with multiple variables was presented herein. The model considered several
techniques, including a decision matrix, systemic archetypes and system dynamics to create
better simulation by a combination of both real data and knowledge of experts. An integral
scheme was proposed to use such techniques in which the Vester matrix was employed to
explore relevant problems. System archetypes allowed for the creation of causal models, and
system dynamics used the model to simulate different scenarios. In this case, the scenarios
were related to the mining industry. The strength of the proposed methodology is that even
though the design stage comprises a structure, it is sufficiently broad in terms of what it
enables, implying that it does not limit the number of variables or structure size. This
situation promotes the free use of the properties of the archetypes, as it does not limit
the creative stage. In this case, the analyst can design an appropriate model in terms of the
available data and obtained experience.
It is noteworthy that systemic archetypes are still a subject of research, whose concepts
and properties have varied over time. An indirect achievement and consequence of this
work is that these properties must be established accurately such that an efficient design
stage of the causal model can be created through the integration of the archetypes (models).
The proposed methodology can be used easily by any analyst in the mining industry (and
particularly in the case study presented herein). However, it could also be used in other
strategic and large-scale sectors: energy, manufacturing, agriculture, logistics, etc.
In the scenarios’ creation stage using the properties and systemic archetypes, it is
possible to analyse more specific scenarios aiming to understand the different conditions
that the Chilean mining industry might encounter in the future. In this case, the variables of
ARLA investment intention, in which the upward and downward behaviours were included, were
32,2 interpreted according to the governability closely related to the situation in Chile.
Additionally, in the model, either by understanding the behaviour of some variables or by
integrating other archetypes, it is possible to carry out a great variety of studies that might
include solutions or politics to mitigate the problems that may arise from the experience of
professionals or from the use of other methodologies.
196 Regarding the results of the analysed scenarios, it is possible to observe the strong
impact of the price of copper on the profits earned by the mining industry. This is
represented in Scenarios 1 2, where a olow governability W could not be accomplished to
mitigate the effect of the copper price on profits. On the contrary, owing to the weak price of
copper in Scenario 3, the mining company stopped production owing to the extremely
unfavourable conditions within the country. In Scenario 4, the country exhibits high
governability, enabling production to be continued temporarily while mineral reserves exist,
despite the red numbers indicated for those years owing to the low price. This implies that
ogovernability W may weaken the impact of reduced copper prices but this does not imply
that it is a method that can be used to mitigate those problems. Further analysis of this
situation should focus on better understanding partial solutions for scenario creation, such
as: first, different policies that may support cost reduction and increase the productivity of
the operations; and second, cycles of investment when the price is high to encounter the
period of time when the price of copper is low.
Finally, it is noteworthy that as mentioned previously, the proposed methodology can be
applied to any situation that implies the analysis of variables, causes and effects to develop
a prospective analysis of future conditions to be encountered by a particular industry. For
instance, it can be applied to a prospective analysis of the market conditions of shipping
lines, or to evaluate decisions related to infrastructure investments, e.g. where the logistics
platform should be located.

References
Alves, C., Valença, G. and Santana, A.F. (2014), “Understanding the factors that influence the adoption
of BPM in two Brazilian public organizations”, Enterprise, Business-Process and Information
Systems Modeling, Springer, Berlin and Heidelberg, pp. 272-286.
Aracil, J. (2005), Dinámica de Sistemas, Publicaciones de Ingeniería de Sistemas, Madrid, Isdefe, p. 87.
Aracil, J. and Gordillo, F. (1997), Dinamica de Sistemas, Alianza Editorial, Madrid, p. 109.
Aristizábal, M., Echeverri, J., Aristizábal, M., Aristizábal, C., Gómez, Z.S. and Gómez, G.S. (2015),
“Systemic archetypes use to developing an interactive software: a children with moderated
autism case”, Proceedings of 10th Iberian Conference on Information Systems and Technologies,
CISTI 2015, pp. 1-6.
Aristizábal, M., Urrego, G., Pérez, A., Echeverry, J., González, L. and González, M. (2014), “Systemic
archetypes for value co-creation based on collaborative cores: case study at a telecommunication
company”, Proceedings of the 9th Iberian Conference on Information Systems and Technologies,
CISTI, pp. 1-4.
Bazrkar, M.H., Tavakoli-Nabavi, E., Zamani, N. and Eslamian, S. (2013), “System dynamic approach to
hydro-politics in hirmand transboundary river basin from sustainability perspective”,
International Journal of Hydrology Science and Technology, Vol. 3 No. 4, pp. 378-398,
doi: 10.1504/IJHST.2013.060338.
Brandmaier, A.M., von Oertzen, T., McArdle, J.J. and Lindenberger, U. (2013), “Structural equation
model trees”, Psychological Methods, Vol. 18 No. 1, pp. 71-86.
Bustamante, M. (2006), “Análisis Estratégico: Metodología de Análisis Prospectivo”, CEGIS
Universidad de Talca, Talca, available at: www.cegis.utalca.cl/doc/miguel_bustamante/
direccion_sistemas_salud_talca/9%20S%C3%8DNTESIS%20DE%20PROSPECTIVA.pdf
(accessed 15 October 2016).
Canavos, G.C. and Medal, E.G. (1987), Probabilidad y estadística, McGraw Hill, México, p. 651. Archetypes
Canfield, M. (2012), “Etapas del Proceso Productivo de una Mina”, Curso de Minería para Periodistas, and system
SONAMI, Santiago, available at: www.sonami.cl/site/wp-content/uploads/2016/04/01.-Etapas-
del-Proceso-Productivo-de-una-Mina.pdf (accessed 15 October 2016).
Cascini, G., Ramadurai, B., Słupiński, M., Rabie, M., Becattini, N., Kikov, I., Kucharavy, D., Nikulin, C.,
Koziolek, S. and Festa, E. (2015), Forecasting and Roadmapping Manufacturing Technologies,
FORMAT Consortium, 2nd ed., November, available at: http://handbook.format-project.eu/wp- 197
content/Downloads/FORMAT-handbook_ed2.pdf (accessed 11 November 2017).
Christensen, C. (2013), The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail,
Harvard Business Review, Boston, p. 253.
Chung Pinzás, A. (2009), “Prospectiva Estratégica: más allá del plan estratégico”, Industrial Data,
Vol. 12 No. 2, pp. 27-31.
Cole, A. (2006), “The influence matrix methodology: a technical report”, Landcare Research Contract
Report No. LC0506/175, Palmerston North.
Comisión Chilena del Cobre (2015), “Caracterizacón de los costos de la gran minería del cobre”,
available at: www.cochilco.cl/Listado%20Temtico/Informe_caracterizacion_de_los_costos.pdf
(accessed 15 October 2016).
Comisión Chilena del Cobre (2016), “Estudio de Insumos Criticos en la Mineria Chilena”, available at:
www.cochilco.cl/Mercado%20de%20Metales/Estudio_Insumos_Cr%C3%ADticos_2016v2.pdf
(accessed 15 October 2016).
David, F.R. (2003), Conceptos de Administración Estratégica, 9th ed., Pearson Educación, México, p. 336.
Donoso Muñoz, M.J. (2013), “El Mercado del Cobre Mundial: evolución, riesgos, características y
potencialidades futuras”, Ingeniere. Revista chilena de ingeniería, Vol. 21 No. 2, pp. 248-261.
Flores, H.R. (2013), “Métodos y modelado matemático para el análisis de procesos complejos en las
organizaciones”, Clerinet, Vol. 1, pp. 64-69.
Forrester, J.W. (1998), “Diseñando el Futuro”, Universidad de Sevilla, D-4808, Sevilla, available at:
http://static.clexchange.org/ftp/documents/sdintro/D-4808.pdf (accessed 15 October 2016).
Godet, M. and Durance, P. (2007), “Prospectiva Estratégica: problemas y métodos”, Cuadernos de
LIPSOR, 2nd ed., Cuaderno No. 20, p. 104, available at: http://prospektiker.es/prospectiva/
Documentos/caja-herramientas-2007.pdf (accessed 15 October 2016).
Godoy, L.A. and Bartó, C.A. (2002), “Validación y Valoración de modelos en la Dinámica de Sistemas”,
Revista Argentina de Enseñanza de la Ingeniería, Vol. 3 No. 5, pp. 31-47.
Gonzáles, F. (2015), “La Tercera”, available at: www.latercera.com/noticia/nacional/2015/11/680-65640
5-9-alvaro-merino-gerente-de-estudios-de-sonami-esperamos-tres-anos-complejos.shtml
(accessed 15 October 2016).
Kim, D. and Anderson, V. (1998), System Archetype Basics: From Story to Structure, Pegasus
Communications, Whaltham, p. 188.
Kim, D. and Burchill, G. (1992), “System archetypes as a diagnostic tool: a field-based study of TQM
implementations”, 10th International Conference of the System Dynamics Society Proceedings
System Dynamics, pp. 311-320.
Li, Q., Zhang, W., Li, H. and He, P. (2017), “CO2 emission trends of china’s primary aluminum industry:
a scenario analysis using system dynamics model”, Energy Policy, Vol. 105, pp. 225-235,
doi: 10.1016/j.enpol.2017.02.046.
Marticorena, J. (2016), “Chile pierde terreno como productor de cobre en medio del desplome
del meta”, La Tercera, available at: www.latercera.com/noticia/negocios/2016/01/655-664520-9-
chile-pierde-terreno-como-productor-de-cobre-en-medio-del-desplome-del-metal.shtml (accessed
15 October 2016).
Mejía, A. and Díaz, G. (2006), “Tipos de Arcos y hacia dónde disparan: Sobre la naturaleza y las
posibilidades de los arquetipos”, Proceedings of IV Congreso Anual del Capítulo Latinoamericano
de la Sociedad de Dinámica de Sistemas, Cancún.
ARLA Mella Polanco, M. and Berrios Silva, C. (2013), “Gobernabilidad, democratización y conflictividad
32,2 social en Chile: escenarios posibles para un nuevo equilibrio”, Polis (Santiago), Vol. 12 No. 35,
pp. 429-458.
Naill, R.F. (1992), “A system dynamics model for national energy policy planning”, System Dynamics
Review, Vol. 8 No. 1, pp. 1-19.
Repenning, N. and Sterman, J. (2001), “Nobody ever gets credit for fixing problems that never
198 happened: creating and sustaining process improvement”, California Management Review,
Vol. 43 No. 4, pp. 64-88.
Santa Catalina, I.M. (2010), “Modelo de Dinámica de Sistemas para la implantación de Tecnologías de la
Información en la Gestión Estratégica Universitaria”, Servicio Editorial de la Universidad del
País Vasco/Euskal Herriko Unibertsitatearen Argitalpen Zerbitzua, available at: https://addi.
ehu.es/handle/10810/12277 (accessed 15 October 2016).
Senge, P.M. (1990), The Fifth Discipline: The Art and Practice of the Learning Organization, Doubleday,
Random House, p. 445.
SONAMI and El Mercurio EMOL (2012), “Chile, País Minero”, Sociedad Nacional de Mineria, pp. 1-22,
available at: www.sonami.cl/site/wp-content/uploads/2016/06/Chile-Pais-Minero-SONAMI-El-
Mercurio.pdf
Vergara Schmalbach, J.C., Fontalvo Herrera, T.J. and Maza Ávila, F. (2010), “La Planeación por
escenarios: Revisión de conceptos y propuestas metodológicas”, Prospectiva, Vol. 8 No. 2,
pp. 21-29.
Wäger, P.A., Hilty, L.M., Arnfalk, P., Erdmann, L. and Goodman, J. (2006), “Experience with a system
dynamics model in a prospective study on the future impact of ICT on environmental
sustainability”, Proceedings of the iEMSs 3rd Biennial Meeting, Summit on Environmental
Modelling and Software, Burlington, VT.
Wolstenholme, E. (2003), “Towards the definition and use of a core set of archetypal structures in
system dynamics”, System Dynamics Review, Vol. 19 No. 1, pp. 7-26.
Wolstenholme, E. (2004), “Using generic system archetypes to support thinking and modelling”, System
Dynamics Review, Vol. 20 No. 4, pp. 341-356.

Further reading
Achachlouei, M.A. and Hilty, L.M. (2015), “Modeling the effects of ICT on environmental sustainability:
revisiting a system dynamics model developed for the European commission”, in Hilty, L. and
Aebischer, B. (Eds), ICT Innovations for Sustainability: Advances in Intelligent Systems and
Computing, Vol. 310, Springer, Cham, pp. 449-474.
Jablonka, E. (2007), “The relevance of modelling and applications: relevant to whom and for what
purpose?”, Modelling and Applications in Mathematics Education, Springer, Boston, pp. 193-200.
Salas, A. (2012), “Chile es una potencia minera a nivel mundial”, Chile país minero, SONAMI-El
Mercurio-EMOL, available at: www.sonami.cl/site/wp-content/uploads/2016/06/Chile-Pais-
Minero-SONAMI-El-Mercurio.pdf (accessed 15 October 2016).
Appendix 1. Model variables Archetypes
This section details the variables introduced in the model, their descriptions and their mathematical
relations.
and system

Variable Type Description Equation

Adjust Auxiliary Lasting time of the 40 199

development activity in the mine
needed
Adjust Auxiliary Duration of the mineral 30
exploration extraction
needed
Capacity Flow Increase of the required DELAY1(MAX (0, wanted production + capacity
increase production capacity to decrease − production capacity), 0.5
fulfil the wanted
production
Production Level Maximum amount of INTEG (capacity increase-decrease Cap,1.6e+006)
capacity mineral the company
produces
Total cost of Auxiliary Cost for the total of the Development rate × development unit cost
development developed mineral
Total cost of Auxiliary Cost for the total of copper Production × production unit cost
production production at this moment
Unit cost of Auxiliary Cost per unit of developed 10
development ton at that moment
Unit cost of Auxiliary Cost per unit of produced WITH LOOKUP (Time, ([(1998, 1400) −
production ton, data are introduced (2020, 6000)], (1999, 1426.64), (2000, 1439.86),
with a table (2001, 1446.48), (2002, 1360.48), (2003, 1620.67),
(2004, 1792.66), (2005, 2156.49), (2006, 2548.98),
(2007, 3137.71), (2008, 3924.9), (2009, 3479),
(2010, 4357), (2011, 4533.48), (2012, 5832.22),
(2013, 5104.57), (2014, 5084.73), (2015, 4476.15))
Total costs Auxiliary Cost of al mining activity Total cost of production + development cost +
total cost of exploration
Total costs of Auxiliary Total cost of exploration Discovery rate × unit costs of exploration
exploration activity
Unit costs of Auxiliary Cost per unit of ton in 20
exploration exploration
Exploration Auxiliary If an investment in IF THEN ELSE (Relación de Exploración ⩾ Limite
investment exploration is going to de Inversión en Exploración, 1, 0)
decision happen at the moment or
not
Wanted Auxiliary Amount of resources the 1e+006+RAMP (150000, 2000, 2000+adjust
development company has planned to development need)
develop
Decision to Auxiliary Whether to invest in IF THEN ELSE (development relation ⩾
invest in development or not at the development investment limit, 1, 0)
development time
Reduction of Flow Obsolescence of the 25
capacity production capacity
Wanted Auxiliary Amount of mineral the 1e+006+RAMP (200000, 2000, 2000 + adjust need
exploration company wants to explore to explore)
In situ Level Amount of mineral in the INTEG (−discovery rate, initial mineral)
earth
Incomes Auxiliary Copper price × production Table AI.
Variables, types,
descriptions and
(continued ) model equations
ARLA Variable Type Description Equation
32,2
Limit of Auxiliary Minimum 1.05
investment in
development
Limit of Auxiliary Minimum 1.1
investment in
200 exploration
Initial mineral Auxiliary Maximum amount of 6.00E+07
mineral in an area
Copper price Auxiliary Value per ton of copper WITH LOOKUP (([(1999, 60) − (2020, 9000)],
(1999, 1573.9), (2000, 1814.11), (2001, 1578.11),
(2002, 1557.83), (2003, 1780.09), (2004, 2870.91),
(2005, 3684.5), (2006, 6731.86), (2007, 5832.22),
(2013, 5104.57), (2014, 5084.73), (2015, 4476.15))
Production Flow MIN (Production capacity, MIN (Wanted
production, Proved reserves))
Wanted Auxiliary Amount of copper WITH LOOKUP (Time, ([(1999,0) − (2020,2e+006)],
production planned to be produce at (1999,1.6e+006), (2015,1.8e+006)))
that period
Resources Level Amount of mineral that INTEG (Discovery rate − Development rate, Initial
can be transformed in resources)
reserves
Initial resources Auxiliary Amount of resources in 1.00E+07
the year the simulation
starts
Development Auxiliary Predisposition to invest in Copper price/(Unit cost of production + Unit cost of
relation development at that development)
period
Exploration Auxiliary Predisposition to invest in Copper price/(Unit cost of production + Unit costs
relation exploration at that period of exploration + Unit cost of development)
Proved reserves Level Amount of reserves ready INTEG (Rate of development-Production,4e+006)
to be transformed in
commercial copper
Development Flow Resources transformed in MIN (Resources, Wanted development) ×
rate reserves Investment decision in development
Discovery rate Flow Mineral transformed in MIN (Resources, Wanted development) × Decisión
resources de Inversión en Desarrollo
Profit Auxiliary Difference between Incomes − Total costs
Table AI. incomes and costs
 Adjust
exploration need
Adjust
development
Initial need
Wanted resources Wanted
Initial mineral development
exploration

Proved
In situ Resources reserves
Discovery Production
Development
rate
rate
Decision to
Limit of investment invest in
in development development <Time>
Unit costs of Wanted Total cost of
exploration production production
Appendix 2. System dynamics model

Exploration Total costs of

investment exploration Development
decision relation
Total cost Production
of development Unit cost of capacity
Capacity Reduction of
Limit of investment development
increase capacity
in exploration
Copper
price
Exploration
relation

Unit cost of
production

Incomes

Profit
Total costs
Archetypes

201
and system

analysis based on

archetype
System dynamics

initial system
Figure A1.

model for simulation

ARLA Appendix 3. Results of system dynamics model
In this appendix, some of the behaviours of other variables used in the model starting from 1999
32,2 are presented.

Total Costs Incomes

20bn 20bn

202 15bn 15bn

10bn 10bn

5bn 5bn

0 0
1999 2003 2007 2011 2015 2019 2023 1999 2003 2007 2011 2015 2019 2023
Time (Year) Time (Year)
Scenario 1 Scenario 3 Scenario 1 Scenario 3
Scenario 2 Scenario 4 Scenario 2 Scenario 4

Profit Production
9bn 3m

6.25bn 2.25m

3.5bn 1.5m

750m 750,000

–2bn 0
1999 2003 2007 2011 2015 2019 2023 1999 2003 2007 2011 2015 2019 2023
Time (Year) Time (Year)
Scenario 1 Scenario 3 Scenario 1 Scenario 3
Scenario 2 Scenario 4 Scenario 2 Scenario 4

Resources Proved Reserves

20m 30m

15m 22.5m

10m 15m

5m 7.5m

0 0
1999 2003 2007 2011 2015 2019 2023 1999 2003 2007 2011 2015 2019 2023
Time (Year) Time (Year)
Scenario 1 Scenario 3 Scenario 1 Scenario 3
Scenario 2 Scenario 4 Scenario 2 Scenario 4

Production
3m

2.25m

1.5m

750,000

Figure A2. 0
Results of the system 1999 2003 2007 2011
Time (Year)
2015 2019 2023

dynamics model Scenario 1 Scenario 3

Scenario 2 Scenario 4

Corresponding author
Rosa Guadalupe G. Gonzalez Ramirez can be contacted at: [email protected]

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: [email protected]