A Big Data approach to forestry harvesting productivity

Autores
Rossit, Daniel Alejandro; Olivera, Alejandro; Viana Céspedes, Víctor; Broz, Diego Ricardo
Año de publicación
2019
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
Modern industrial technology enables to collect and process large amount of data, providing valuable information for different industry activities. A representative case of this evolution is the Forest industry, since modern forest harvesters are equipped with automatic data collection devices. The collected data can be extracted and communicated to computers using special forestry protocols, as StanForD, where it can be analysed. This skill of modern harvesters allows to study harvest productivity with thousands of records, instead of having a few hundred as it would be possible by recording through traditional methods (visual inspection or filming). However, traditional analytical methods, as linear regression, are not capable to deal with this volume of data (or, at least, does not take full advantage of the data potential), consequently, new approaches must be considered. Our proposal is to address this shortcoming using data mining methods, specially, we consider decision trees and k-means algorithms. We study how different variables (DBH, species, shift and operator) affect the productivity of a forest harvester considering real scenario data. The harvest data comes from Eucalyptus spp. plantations in Uruguay where the harvest system implemented is cut-to-length. To analyse the data, firstly, productivity is modelled in a categorical manner considering two different approaches: ranges of equal intervals and ranges calculated using k-means clustering algorithm. Then, Decision Trees methods are applied to analyse the influence of the mentioned variables in productivity. The results show that clustering is a proper approach to categorically model scalar productivity and that DBH is the most influential factor in productivity. Moreover, Decision Trees, after setting DBH values, allowed to use new variables to describe productivity, achieving very high levels of accuracy, in many cases greater than 90%.
Fil: Rossit, Daniel Alejandro. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Bahía Blanca. Instituto de Matemática Bahía Blanca. Universidad Nacional del Sur. Departamento de Matemática. Instituto de Matemática Bahía Blanca; Argentina
Fil: Olivera, Alejandro. Universidad de la República; Uruguay
Fil: Viana Céspedes, Víctor. Universidad de la Republica; Uruguay
Fil: Broz, Diego Ricardo. Universidad Nacional de Misiones; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Nordeste; Argentina
Materia
DATA SCINCE
FOREST HARVEST
DECISION MAKING
FOREST ENGINEERING
Nivel de accesibilidad
acceso abierto
Condiciones de uso
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
Repositorio
CONICET Digital (CONICET)
Institución
Consejo Nacional de Investigaciones Científicas y Técnicas
OAI Identificador
oai:ri.conicet.gov.ar:11336/118819

id CONICETDig_03821c4b7be78b6e758a4eddaace83e5
oai_identifier_str oai:ri.conicet.gov.ar:11336/118819
network_acronym_str CONICETDig
repository_id_str 3498
network_name_str CONICET Digital (CONICET)
spelling A Big Data approach to forestry harvesting productivityRossit, Daniel AlejandroOlivera, AlejandroViana Céspedes, VíctorBroz, Diego RicardoDATA SCINCEFOREST HARVESTDECISION MAKINGFOREST ENGINEERINGhttps://purl.org/becyt/ford/4.1https://purl.org/becyt/ford/4Modern industrial technology enables to collect and process large amount of data, providing valuable information for different industry activities. A representative case of this evolution is the Forest industry, since modern forest harvesters are equipped with automatic data collection devices. The collected data can be extracted and communicated to computers using special forestry protocols, as StanForD, where it can be analysed. This skill of modern harvesters allows to study harvest productivity with thousands of records, instead of having a few hundred as it would be possible by recording through traditional methods (visual inspection or filming). However, traditional analytical methods, as linear regression, are not capable to deal with this volume of data (or, at least, does not take full advantage of the data potential), consequently, new approaches must be considered. Our proposal is to address this shortcoming using data mining methods, specially, we consider decision trees and k-means algorithms. We study how different variables (DBH, species, shift and operator) affect the productivity of a forest harvester considering real scenario data. The harvest data comes from Eucalyptus spp. plantations in Uruguay where the harvest system implemented is cut-to-length. To analyse the data, firstly, productivity is modelled in a categorical manner considering two different approaches: ranges of equal intervals and ranges calculated using k-means clustering algorithm. Then, Decision Trees methods are applied to analyse the influence of the mentioned variables in productivity. The results show that clustering is a proper approach to categorically model scalar productivity and that DBH is the most influential factor in productivity. Moreover, Decision Trees, after setting DBH values, allowed to use new variables to describe productivity, achieving very high levels of accuracy, in many cases greater than 90%.Fil: Rossit, Daniel Alejandro. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Bahía Blanca. Instituto de Matemática Bahía Blanca. Universidad Nacional del Sur. Departamento de Matemática. Instituto de Matemática Bahía Blanca; ArgentinaFil: Olivera, Alejandro. Universidad de la República; UruguayFil: Viana Céspedes, Víctor. Universidad de la Republica; UruguayFil: Broz, Diego Ricardo. Universidad Nacional de Misiones; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Nordeste; ArgentinaElsevier2019-06info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articuloapplication/pdfapplication/pdfapplication/pdfhttp://hdl.handle.net/11336/118819Rossit, Daniel Alejandro; Olivera, Alejandro; Viana Céspedes, Víctor; Broz, Diego Ricardo; A Big Data approach to forestry harvesting productivity; Elsevier; Computers and Eletronics in Agriculture; 161; 6-2019; 29-520168-1699CONICET DigitalCONICETenginfo:eu-repo/semantics/altIdentifier/doi/10.1016/j.compag.2019.02.029info:eu-repo/semantics/altIdentifier/url/https://www.sciencedirect.com/science/article/abs/pii/S0168169917316368info:eu-repo/semantics/openAccesshttps://creativecommons.org/licenses/by-nc-sa/2.5/ar/reponame:CONICET Digital (CONICET)instname:Consejo Nacional de Investigaciones Científicas y Técnicas2025-09-29T10:43:59Zoai:ri.conicet.gov.ar:11336/118819instacron:CONICETInstitucionalhttp://ri.conicet.gov.ar/Organismo científico-tecnológicoNo correspondehttp://ri.conicet.gov.ar/oai/requestdasensio@conicet.gov.ar; lcarlino@conicet.gov.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:34982025-09-29 10:43:59.344CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicasfalse
dc.title.none.fl_str_mv A Big Data approach to forestry harvesting productivity
title A Big Data approach to forestry harvesting productivity
spellingShingle A Big Data approach to forestry harvesting productivity
Rossit, Daniel Alejandro
DATA SCINCE
FOREST HARVEST
DECISION MAKING
FOREST ENGINEERING
title_short A Big Data approach to forestry harvesting productivity
title_full A Big Data approach to forestry harvesting productivity
title_fullStr A Big Data approach to forestry harvesting productivity
title_full_unstemmed A Big Data approach to forestry harvesting productivity
title_sort A Big Data approach to forestry harvesting productivity
dc.creator.none.fl_str_mv Rossit, Daniel Alejandro
Olivera, Alejandro
Viana Céspedes, Víctor
Broz, Diego Ricardo
author Rossit, Daniel Alejandro
author_facet Rossit, Daniel Alejandro
Olivera, Alejandro
Viana Céspedes, Víctor
Broz, Diego Ricardo
author_role author
author2 Olivera, Alejandro
Viana Céspedes, Víctor
Broz, Diego Ricardo
author2_role author
author
author
dc.subject.none.fl_str_mv DATA SCINCE
FOREST HARVEST
DECISION MAKING
FOREST ENGINEERING
topic DATA SCINCE
FOREST HARVEST
DECISION MAKING
FOREST ENGINEERING
purl_subject.fl_str_mv https://purl.org/becyt/ford/4.1
https://purl.org/becyt/ford/4
dc.description.none.fl_txt_mv Modern industrial technology enables to collect and process large amount of data, providing valuable information for different industry activities. A representative case of this evolution is the Forest industry, since modern forest harvesters are equipped with automatic data collection devices. The collected data can be extracted and communicated to computers using special forestry protocols, as StanForD, where it can be analysed. This skill of modern harvesters allows to study harvest productivity with thousands of records, instead of having a few hundred as it would be possible by recording through traditional methods (visual inspection or filming). However, traditional analytical methods, as linear regression, are not capable to deal with this volume of data (or, at least, does not take full advantage of the data potential), consequently, new approaches must be considered. Our proposal is to address this shortcoming using data mining methods, specially, we consider decision trees and k-means algorithms. We study how different variables (DBH, species, shift and operator) affect the productivity of a forest harvester considering real scenario data. The harvest data comes from Eucalyptus spp. plantations in Uruguay where the harvest system implemented is cut-to-length. To analyse the data, firstly, productivity is modelled in a categorical manner considering two different approaches: ranges of equal intervals and ranges calculated using k-means clustering algorithm. Then, Decision Trees methods are applied to analyse the influence of the mentioned variables in productivity. The results show that clustering is a proper approach to categorically model scalar productivity and that DBH is the most influential factor in productivity. Moreover, Decision Trees, after setting DBH values, allowed to use new variables to describe productivity, achieving very high levels of accuracy, in many cases greater than 90%.
Fil: Rossit, Daniel Alejandro. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Bahía Blanca. Instituto de Matemática Bahía Blanca. Universidad Nacional del Sur. Departamento de Matemática. Instituto de Matemática Bahía Blanca; Argentina
Fil: Olivera, Alejandro. Universidad de la República; Uruguay
Fil: Viana Céspedes, Víctor. Universidad de la Republica; Uruguay
Fil: Broz, Diego Ricardo. Universidad Nacional de Misiones; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - Nordeste; Argentina
description Modern industrial technology enables to collect and process large amount of data, providing valuable information for different industry activities. A representative case of this evolution is the Forest industry, since modern forest harvesters are equipped with automatic data collection devices. The collected data can be extracted and communicated to computers using special forestry protocols, as StanForD, where it can be analysed. This skill of modern harvesters allows to study harvest productivity with thousands of records, instead of having a few hundred as it would be possible by recording through traditional methods (visual inspection or filming). However, traditional analytical methods, as linear regression, are not capable to deal with this volume of data (or, at least, does not take full advantage of the data potential), consequently, new approaches must be considered. Our proposal is to address this shortcoming using data mining methods, specially, we consider decision trees and k-means algorithms. We study how different variables (DBH, species, shift and operator) affect the productivity of a forest harvester considering real scenario data. The harvest data comes from Eucalyptus spp. plantations in Uruguay where the harvest system implemented is cut-to-length. To analyse the data, firstly, productivity is modelled in a categorical manner considering two different approaches: ranges of equal intervals and ranges calculated using k-means clustering algorithm. Then, Decision Trees methods are applied to analyse the influence of the mentioned variables in productivity. The results show that clustering is a proper approach to categorically model scalar productivity and that DBH is the most influential factor in productivity. Moreover, Decision Trees, after setting DBH values, allowed to use new variables to describe productivity, achieving very high levels of accuracy, in many cases greater than 90%.
publishDate 2019
dc.date.none.fl_str_mv 2019-06
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/11336/118819
Rossit, Daniel Alejandro; Olivera, Alejandro; Viana Céspedes, Víctor; Broz, Diego Ricardo; A Big Data approach to forestry harvesting productivity; Elsevier; Computers and Eletronics in Agriculture; 161; 6-2019; 29-52
0168-1699
CONICET Digital
CONICET
url http://hdl.handle.net/11336/118819
identifier_str_mv Rossit, Daniel Alejandro; Olivera, Alejandro; Viana Céspedes, Víctor; Broz, Diego Ricardo; A Big Data approach to forestry harvesting productivity; Elsevier; Computers and Eletronics in Agriculture; 161; 6-2019; 29-52
0168-1699
CONICET Digital
CONICET
dc.language.none.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/doi/10.1016/j.compag.2019.02.029
info:eu-repo/semantics/altIdentifier/url/https://www.sciencedirect.com/science/article/abs/pii/S0168169917316368
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
eu_rights_str_mv openAccess
rights_invalid_str_mv https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
dc.format.none.fl_str_mv application/pdf
application/pdf
application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:CONICET Digital (CONICET)
instname:Consejo Nacional de Investigaciones Científicas y Técnicas
reponame_str CONICET Digital (CONICET)
collection CONICET Digital (CONICET)
instname_str Consejo Nacional de Investigaciones Científicas y Técnicas
repository.name.fl_str_mv CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicas
repository.mail.fl_str_mv dasensio@conicet.gov.ar; lcarlino@conicet.gov.ar
_version_ 1844614476162138112
score 13.070432