Propuesta de mejora de la priorización de pasivos ambientales mineros en el Perú mediante una metodología basada en inteligencia artificial con Grey Systems
Date
2020-09-11
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Pontificia Universidad Católica del Perú
Abstract
La minería es desde hace unas décadas una parte fundamental del desarrollo
económico en este país, dicho desarrollo trajo consigo muchas cosas positivas; sin
embargo, debido a la poca preocupación ambiental que existía hasta hace unos años,
se generaron aspectos negativos precisamente en este ámbito, como son los pasivos
ambientales.
Estos pasivos se generaron debido a que no existía una legislación que regulara el cese
o finalización de una operación minera, por lo cual, en muchas ocasiones al acabar la
operación se abandonaba las labores tal como estaban, generando así un riesgo para
la salud y seguridad humana así también como para la integridad de los ecosistemas.
En la actualidad se han registrado un total de 8448 pasivos ambientales a lo largo del
territorio nacional, afortunadamente existe una preocupación por parte del estado para
poder tratar esta problemática, habiendo creado una metodología de clasificación de
pasivos ambientales para poder priorizarlos debido a su importancia; sin embargo, esta
metodología se basa en estadística y teniendo en cuenta que en la actualidad se existen
otros métodos de clasificación, esta tesis se propone plantear una metodología basada
en inteligencia artificial con grey systems para mejorar la priorización de pasivos
ambientales mineros en el Perú.
La metodología de Grey Clustering está basada en inteligencia artificial, la cual es una
combinación de matemática con programación para el tratamiento de datos, con dicha
metodología se procesó la información obtenida de la Dirección General de Asuntos
Ambientales Mineros (DGAAM) del Ministerio de Energía y Minas, obteniendo una
clasificación alternativa, nuevas puntuaciones para los pasivos además de nuevos
rangos para clasificarlos según el nivel de riesgo.
Se concluyó que con la nueva clasificación los pasivos que pertenecen al nivel de riesgo
muy alto disminuyen en un 97% al utilizar la metodología de Grey Clustering y la
clasificación porcentual, mientras que los que los pasivos que pertenecen al nivel de
riesgo alto aumentan en un 42% con la nueva clasificación.
Cabe destacar que la metodología y clasificación planteadas en este trabajo aún se
encuentra en análisis, por lo que los resultados presentados en esta tesis son de manera
preliminar y estos aún pueden ser mejorados.
Mining has been a fundamental part of economic development in this country for a few decades, this development brought with it many positive things; however, due to the lack of environmental concern that existed until a few years ago, negative aspects were generated precisely in this area, such as environmental liabilities. These liabilities were generated because there was no legislation regulating the cessation or completion of a mining operation, so that many times at the end of the operation work was abandoned as they were, thus creating a risk to human health and safety as well as to the integrity of ecosystems. A total of 8448 environmental liabilities have now been recorded throughout the national territory, fortunately there is a concern on the part of the state to be able to address this problem, having created a methodology for classifying environmental liabilities to be able to prioritize them because of their importance; however, this methodology is based on statistics and taking into account that there are currently other classification methods, this thesis is proposed to propose a methodology based on artificial intelligence with grey systems to improve the prioritization of mining environmental liabilities in Peru. Grey Clustering's methodology is based on artificial intelligence, which is a combination of mathematics with programming for data processing, with this methodology the information obtained from the “Dirección General de Asuntos Ambientales Mineros” (DGAAM) from the Ministry of Energy and Mines, was processed obtaining an alternative classification, new scores for liabilities in addition to new ranges to classify them according to the level of risk. It was concluded that with the new classification, liabilities belonging to the very high-risk level decrease by 97% when using the Grey Clustering methodology and the percentage rating, while those that are liabilities belonging to the high-risk level increase by 42% with the new classification. It should be noted that the methodology and classification raised in this work is still in analysis, so the results presented in this thesis are preliminary and these can still be improved.
Mining has been a fundamental part of economic development in this country for a few decades, this development brought with it many positive things; however, due to the lack of environmental concern that existed until a few years ago, negative aspects were generated precisely in this area, such as environmental liabilities. These liabilities were generated because there was no legislation regulating the cessation or completion of a mining operation, so that many times at the end of the operation work was abandoned as they were, thus creating a risk to human health and safety as well as to the integrity of ecosystems. A total of 8448 environmental liabilities have now been recorded throughout the national territory, fortunately there is a concern on the part of the state to be able to address this problem, having created a methodology for classifying environmental liabilities to be able to prioritize them because of their importance; however, this methodology is based on statistics and taking into account that there are currently other classification methods, this thesis is proposed to propose a methodology based on artificial intelligence with grey systems to improve the prioritization of mining environmental liabilities in Peru. Grey Clustering's methodology is based on artificial intelligence, which is a combination of mathematics with programming for data processing, with this methodology the information obtained from the “Dirección General de Asuntos Ambientales Mineros” (DGAAM) from the Ministry of Energy and Mines, was processed obtaining an alternative classification, new scores for liabilities in addition to new ranges to classify them according to the level of risk. It was concluded that with the new classification, liabilities belonging to the very high-risk level decrease by 97% when using the Grey Clustering methodology and the percentage rating, while those that are liabilities belonging to the high-risk level increase by 42% with the new classification. It should be noted that the methodology and classification raised in this work is still in analysis, so the results presented in this thesis are preliminary and these can still be improved.
Description
Keywords
Industria minera--Aspectos ambientales--Perú, Responsabilidad por daños al medio ambiente--Perú, Inteligencia artificial--Aspectos ambientales--Perú