Homepage arrow Research presentation
Presentation of my projects in modeling domain
Index
Presentation of my projects in modeling domain
- Aggregates exploitation and transport flow (France)
- Artisanal Gold Mining (Burkina Faso)
**    Paper Award
- Ultra-pure Quartz exploitation (Madagascar)
- Collective management of animal wastes (La Réunion)

Artisanal gold mining in Burkina Faso

(Photos: © BRGM)

1- Background and issues

activity 2In Burkina Faso, the activity of artisanal gold mining production has a very large potential in terms of available jobs. Artisanal mines of Burkina Faso, commonly overpopulated, are striking in terms of their teeming and abundant character. For example, some of these mines, such as Bouda (Passoré province) in 1991, or Mogrenore (Boulgou province) in 1998, hosted up to 40,000 individuals during the paroxysmal phase of the gold rush.

Thus, this activity, even if characterized by very low productivity, has a very large potential in terms of available jobs and is then interesting for the development of the artisans population.

activity 3However, gold production has seen a severe and continuous decline since 1998. In addition, more liberal laws rendered such activities more opaque and strengthening clandestine gold buyers’ channels.

This clandestine gold involves a loss of earnings (via tax) for the State. The official purchasing channels rarely collect more than 20% of the gold produced on a site. Experience has shown that the systematic control of gold-producing sites is an expensive and not always efficient proposition for the authorities. For example in 2002, because of insufficient human or material means, the Burkinabe Precious Metals Office (or CBMP, namely the State corporation for management and control of the artisanal sector) estimated to be incapable of recovering more than 3 to 5% of the production in some districts

gold dustA coherent policy is then necessary in order to set simultaneous actions of observation and control of this activity by the authorities and of assistance to the mining population. Technical and socioeconomic analysis resulted to the conclusion that acting on the official gold price seems, for the State, to be one mean to progress towards this coherent policy. To be effective, the control of the site should be accompanied by a reduction in the difference between official and clandestine gold prices.

 

 2- Introduction to prospective Modeling and Simulation

Thus, between 2001 and 2007, the BRGM carried out, in close collaboration with the BUMIGEB (the Geological Survey of Burkina Faso), research on Artisanal and Small-scale Gold Mining in Burkina Faso. We are interested in two points: development and governance. Development concerns the improving of the actors’ income while governance is about the reduction of the importance of the clandestine gold buyers in the sector.
Nevertheless, given the complexity of the sector, determining the appropriate gold price that the authorities could adopt is not an easy task.

Prospective modeling and simulation seems to be an interesting approach as a help to cope with the issues above. In particular, we proposed Multi-Agent System (MAS) modeling and simulation to carry out prospective research on answering to the following questions:
q1)- “with regard to development, how the official gold price should be set if one wishes to improve the income of the ASGMA population by a given x%?”
q2)- “with regard to governance, how the official gold price should be set if one wishes to reduce the importance of the clandestine gold buyers in the ASGMA by
y%?”

As a starting point, our research was particularly focused on Artisanal and Small-scale Gold Mining in Alga (or ASGMA). Alga is a site in Burkina Faso, which is considered as a reference for carrying out such work.

3- Modeling the artisanal gold mining activity

Basic principles of the activity

gold minersExploitation takes place in two areas: the extraction area (EA) and the transformation area (TA). Exploitation means the manipulation of ores by the following successive tasks:
- extraction from the shaft (in EA),
- conveying (between EA and TA),
- crushing, grinding, sluice washing, winnowing and mill grinding (in TA).

The final ore then gives gold dust.

Each task is performed by a team of actors.

pilageIn EA, we find:
- miners: the main workers in the extraction of a shaft.
- shaft owners: each shaft has one shaft owner on behalf of which the miners in this shaft work. A shaft owner takes care of the financial administration (ex: the distribution of the team income, the payment of the production cost …) of the shaft.
- shaft keepers: they are also in the miners team but they rather act as a guard instead of a real worker. Whereas the miners are men, a guard may be an old man.

In TA, we successively find team of crushers, grinders, sluice washers, winnowers and mill grinders, following the TA task organization described above.

The general process of the exploitation is the following: after a phase of extraction, each shaft owner subcontracts the transformation of its ore to the actors in TA. And since EA generally contains several shafts for one TA, each task in TA possesses a “First In First Out” (FIFO) list, used to temporarily store the ores as and when they arrive from the preceding task.

gold dustBetween the two zones commonly, we find
- one or more collectors, also called buyers, which can indeed be in EA and/or in TA. There are two types of collector: official collectors, from the CBMP, and collectors which receive the products in a clandestine way and buy them at a more advantageous price for the workers.
- conveyors which carry the ores from a shaft to crushers. However, in this study, we have not included the conveying process since we do not yet have all of the necessary data concerning them. For example, we can mention the spatial data describing their path during the transport, given that their income depends, inter alia, on the length of these ways they covered.

Modeling of the ASGMA application

ASGMA is built on the ADK platform. In EA, the group body is the shaft. In TA, the body, for each group, is the zone where this group works. In concrete terms, it is a demarcated area that we call a taskzone. The FIFO list of the respective group, i.e. the element that contains the ores normally treated by the group is, for example, physically stored inside this area.

asgma application based on adkThe behavior layer is focused on group than individual behavioral patterns. Notions like shaft owners, keepers, miners, crushers, grinders, etc. become roles. They are executed by agents, who are actually the translation of the concept of actors in a MAS context. These roles are associated with actions, which are the translation of the concept of tasks in a MAS context.

An additional role exists in TA: the group representative. When a taskzone has finished treating one element ore_el of its FIFO, and ore_el is still in transformation, the role of the representative is to transfer it to the next taskzone. Otherwise, if ore_el is the final task that  gives Wdust, the representative sends it to its owners.

Figure on the right presents a (partial) view of the application of ADK to the ASGMA. For simplification, we only have taken the case of winnowers as an example on the TA side.

4- Simulation results obtained concerning the prospective research

Given the two questions q1 (development) and q2 (governance) presented in the introduction, the figure below presents the possible values of the official gold price that result from our prospective research.

According to these results, the authorities in Burkina Faso may, for example, take the following decision:
“if we want to raise the actors’ income by 1.5% (for better development) while decreasing the part of clandestine buyers from 84% to 60% (for better governance), we should raise the official gold price, from 4500 Cfa/gAu up to, at least, 5500 Cfa/gAu”.

simulation results

The values we obtained during the simulations is probably a solution for the validation of the model but we agree that it is not certainly unique. There are surely further ways to make progress in determining how far the real scope (and limitations) of the general domain of our values lies: by continuing the simulation runs and by directly discussing the results with stakeholders. Such a participative approach is important, since it gives the modeller a better picture of the real society, with a more accurate view than that of experts/scientists. In the medium term, this will help us in integrating sustainable development indicators in this work (perhaps some of these indicators are already among the inputs/outputs data we have studied herein).

Meanwhile, we believe that what we have achieved so far is promising. Technically, we have constructed a simulation mechanism that has been validated and that has allowed us to carry out initial prospective research. The realization of the decision issuing from the simulation results (for example the increase of the gold price to 5500 Cfa/gAu) then depends on the financial possibility for the State itself. Our tool aims at proposing values and the possible impact of these values on the socioeconomic situation of the sector.

At term, the work should lead to a tool that is as reliable as possible and which may be used as an aid for answering questions other than q1 and q2. This idea of analysis, modeling and simulation should also result in a detailed and quantified evaluation of the artisanal gold sector in Burkina Faso, in a rigorous identification of the best targets from the perspective of supporting small mines and, more generally, in proposals for concrete actions to improve the contribution of ASM to sustainable development since remind that ASM provides a potential in terms of available jobs.

(Photos: © BRGM)