Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais
Autor(a) principal: | |
---|---|
Data de Publicação: | 2019 |
Tipo de documento: | Dissertação |
Idioma: | por |
Título da fonte: | Manancial - Repositório Digital da UFSM |
Texto Completo: | http://repositorio.ufsm.br/handle/1/20933 |
Resumo: | Forecasting is an important activity in planning, process control and decision making, especially in the industrial context, and the combination of forecasts is an attractive approach in order to obtain accurate results. The objective of the present research was to investigate the application and combination of prediction methods aiming to verify the accuracy and the gains in terms of error reduction for time series of the industrial sector. The research is structured in two articles: article I discusses a systematic review of the literature on the development and applications of forecasting models in industrial processes. In this study, the Web of Science, Scopus and IEE databases were searched, composing a portfolio of 354 articles published in scientific journals in the last 10 years. The analysis of the literature was carried out in three stages: (i) initially an analysis of the frequencies of the publications was carried out regarding the periodicals, years of publications, authors, countries and number of citations; (ii) analysis of cocitation, bibliographic coupling and similarity analysis of the co-occurrence of the terms in the studies; (iii) finally, a unified framework was developed to classify the applications of forecast methods in industrial processes. Article II addressed the application of the combination of forecasts in a case study conducted in a large mining and logistics company with industrial production series from an integrated port system, in this study the general objective was to fit a model of accurate combined forecasting capable of capturing the serial time behavior of the system. The models of Exponential Smoothing, ARIMA modeling from the Box-Jenkins methodology and the Artificial Neural Networks models were used as individual predictors. The combination of the forecasts was carried out by three different approaches: the combination by arithmetic mean, by the Minimum Variance method, consisting of a linear combination from the variance of the prediction errors, and from Copula models, being a nonlinear approach based on the degree of dependence on individual forecasts. As evaluation measures of the proposed models, the RMSE and U-Theil criteria were used. The results showed that the ARIMA and ANN models were superior in terms of accuracy in relation to the individual predictions, and the methods of combination by model of copula produced more accurate predictions in relation to the other approaches of combinations adopted. |
id |
UFSM_48bb5295a449d52e5b8baf56523fb150 |
---|---|
oai_identifier_str |
oai:repositorio.ufsm.br:1/20933 |
network_acronym_str |
UFSM |
network_name_str |
Manancial - Repositório Digital da UFSM |
repository_id_str |
|
spelling |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriaisCombination of forecasts by linear and non-linear methods: an application in industrial seriesCombinação de previsõesARIMARedes neurais artificiaisVariância mínimaModelos de cópulasCombination of forecastsArtificial neural networksMinimum varianceCopula modelsCNPQ::ENGENHARIAS::ENGENHARIA DE PRODUCAOForecasting is an important activity in planning, process control and decision making, especially in the industrial context, and the combination of forecasts is an attractive approach in order to obtain accurate results. The objective of the present research was to investigate the application and combination of prediction methods aiming to verify the accuracy and the gains in terms of error reduction for time series of the industrial sector. The research is structured in two articles: article I discusses a systematic review of the literature on the development and applications of forecasting models in industrial processes. In this study, the Web of Science, Scopus and IEE databases were searched, composing a portfolio of 354 articles published in scientific journals in the last 10 years. The analysis of the literature was carried out in three stages: (i) initially an analysis of the frequencies of the publications was carried out regarding the periodicals, years of publications, authors, countries and number of citations; (ii) analysis of cocitation, bibliographic coupling and similarity analysis of the co-occurrence of the terms in the studies; (iii) finally, a unified framework was developed to classify the applications of forecast methods in industrial processes. Article II addressed the application of the combination of forecasts in a case study conducted in a large mining and logistics company with industrial production series from an integrated port system, in this study the general objective was to fit a model of accurate combined forecasting capable of capturing the serial time behavior of the system. The models of Exponential Smoothing, ARIMA modeling from the Box-Jenkins methodology and the Artificial Neural Networks models were used as individual predictors. The combination of the forecasts was carried out by three different approaches: the combination by arithmetic mean, by the Minimum Variance method, consisting of a linear combination from the variance of the prediction errors, and from Copula models, being a nonlinear approach based on the degree of dependence on individual forecasts. As evaluation measures of the proposed models, the RMSE and U-Theil criteria were used. The results showed that the ARIMA and ANN models were superior in terms of accuracy in relation to the individual predictions, and the methods of combination by model of copula produced more accurate predictions in relation to the other approaches of combinations adopted.Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPESRealizar previsões é uma importante atividade no planejamento, controle de processos e tomada de decisões, sobretudo no contexto industrial, sendo a combinação de previsões uma abordagem atrativa a fim de obter de resultados acurados. Dessa forma, o objetivo da presente pesquisa foi investigar a aplicação e combinação de métodos de previsões objetivando verificar a acurácia e os ganhos em termos de redução de erros para séries temporais do setor industrial. A pesquisa está estruturada em dois artigos: o artigo I aborda uma revisão sistemática de literatura acerca do desenvolvimento e aplicações de modelos de previsão em processos industriais. Nesse estudo foi realizada a busca nas bases Web of Science, Scopus e IEE, compondo um portfólio de 354 artigos publicados em periódicos científicos nos últimos 10 anos. A análise da literatura foi realizada em três etapas: (i) inicialmente foi realizada uma análise das frequências das publicações quanto aos periódicos, anos de publicações, autores, países e número de citações; (ii) foi realizada a análise de cocitação, de acoplamento bibliográfico e análise de similitude da coocorrência dos termos nos estudos; (iii) por fim foi elaborado um framework unificado para classificação das aplicações de métodos de previsão em processos industriais. O artigo II abordou a aplicação da combinação de previsões em um estudo de caso realizado em uma empresa de grande porte do setor de mineração e logística, com séries de produção industrial provenientes de um sistema portuário integrado, neste estudo o objetivo geral foi ajustar um modelo de previsão combinada acurado capaz de capturar o comportamento seriado temporal do sistema. Foram utilizados como preditores individuais os modelos de Suavização Exponencial, a modelagem ARIMA a partir da metodologia de Box-Jenkins e os modelos de Redes Neurais Artificiais. A combinação das previsões foi realizada por três abordagens distintas: a combinação por média aritmética, pelo método da Variância Mínima, consistindo em uma combinação linear a partir da variância dos erros de previsão, e a partir de modelos de cópulas, sendo uma abordagem não linear baseada no grau de dependência das previsões individuais. Como medidas de avaliação dos modelos propostos foram utilizados os critérios RMSE e U-Theil. Os resultados encontrados demonstraram que os modelos ARIMA e ANN foram superiores em termos de acurácia em relação as previsões individuais, e os métodos de combinação por modelo de cópula produziram previsões mais acuradas em relação às demais abordagens de combinações adotadas.Universidade Federal de Santa MariaBrasilEngenharia de ProduçãoUFSMPrograma de Pós-Graduação em Engenharia de ProduçãoCentro de TecnologiaSouza, Adriano Mendonçahttp://lattes.cnpq.br/5271075797851198Silva, Wesley Vieira daVeiga, Claudimar Pereira daAgostino, Icaro Romolo Sousa2021-05-20T10:59:08Z2021-05-20T10:59:08Z2019-02-20info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://repositorio.ufsm.br/handle/1/20933porAttribution-NonCommercial-NoDerivatives 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-nd/4.0/info:eu-repo/semantics/openAccessreponame:Manancial - Repositório Digital da UFSMinstname:Universidade Federal de Santa Maria (UFSM)instacron:UFSM2022-06-13T16:57:50Zoai:repositorio.ufsm.br:1/20933Biblioteca Digital de Teses e Dissertaçõeshttps://repositorio.ufsm.br/ONGhttps://repositorio.ufsm.br/oai/requestatendimento.sib@ufsm.br||tedebc@gmail.comopendoar:2022-06-13T16:57:50Manancial - Repositório Digital da UFSM - Universidade Federal de Santa Maria (UFSM)false |
dc.title.none.fl_str_mv |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais Combination of forecasts by linear and non-linear methods: an application in industrial series |
title |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais |
spellingShingle |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais Agostino, Icaro Romolo Sousa Combinação de previsões ARIMA Redes neurais artificiais Variância mínima Modelos de cópulas Combination of forecasts Artificial neural networks Minimum variance Copula models CNPQ::ENGENHARIAS::ENGENHARIA DE PRODUCAO |
title_short |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais |
title_full |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais |
title_fullStr |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais |
title_full_unstemmed |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais |
title_sort |
Combinação de previsões por métodos lineares e não-lineares: uma aplicação em séries industriais |
author |
Agostino, Icaro Romolo Sousa |
author_facet |
Agostino, Icaro Romolo Sousa |
author_role |
author |
dc.contributor.none.fl_str_mv |
Souza, Adriano Mendonça http://lattes.cnpq.br/5271075797851198 Silva, Wesley Vieira da Veiga, Claudimar Pereira da |
dc.contributor.author.fl_str_mv |
Agostino, Icaro Romolo Sousa |
dc.subject.por.fl_str_mv |
Combinação de previsões ARIMA Redes neurais artificiais Variância mínima Modelos de cópulas Combination of forecasts Artificial neural networks Minimum variance Copula models CNPQ::ENGENHARIAS::ENGENHARIA DE PRODUCAO |
topic |
Combinação de previsões ARIMA Redes neurais artificiais Variância mínima Modelos de cópulas Combination of forecasts Artificial neural networks Minimum variance Copula models CNPQ::ENGENHARIAS::ENGENHARIA DE PRODUCAO |
description |
Forecasting is an important activity in planning, process control and decision making, especially in the industrial context, and the combination of forecasts is an attractive approach in order to obtain accurate results. The objective of the present research was to investigate the application and combination of prediction methods aiming to verify the accuracy and the gains in terms of error reduction for time series of the industrial sector. The research is structured in two articles: article I discusses a systematic review of the literature on the development and applications of forecasting models in industrial processes. In this study, the Web of Science, Scopus and IEE databases were searched, composing a portfolio of 354 articles published in scientific journals in the last 10 years. The analysis of the literature was carried out in three stages: (i) initially an analysis of the frequencies of the publications was carried out regarding the periodicals, years of publications, authors, countries and number of citations; (ii) analysis of cocitation, bibliographic coupling and similarity analysis of the co-occurrence of the terms in the studies; (iii) finally, a unified framework was developed to classify the applications of forecast methods in industrial processes. Article II addressed the application of the combination of forecasts in a case study conducted in a large mining and logistics company with industrial production series from an integrated port system, in this study the general objective was to fit a model of accurate combined forecasting capable of capturing the serial time behavior of the system. The models of Exponential Smoothing, ARIMA modeling from the Box-Jenkins methodology and the Artificial Neural Networks models were used as individual predictors. The combination of the forecasts was carried out by three different approaches: the combination by arithmetic mean, by the Minimum Variance method, consisting of a linear combination from the variance of the prediction errors, and from Copula models, being a nonlinear approach based on the degree of dependence on individual forecasts. As evaluation measures of the proposed models, the RMSE and U-Theil criteria were used. The results showed that the ARIMA and ANN models were superior in terms of accuracy in relation to the individual predictions, and the methods of combination by model of copula produced more accurate predictions in relation to the other approaches of combinations adopted. |
publishDate |
2019 |
dc.date.none.fl_str_mv |
2019-02-20 2021-05-20T10:59:08Z 2021-05-20T10:59:08Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://repositorio.ufsm.br/handle/1/20933 |
url |
http://repositorio.ufsm.br/handle/1/20933 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.rights.driver.fl_str_mv |
Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/ |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal de Santa Maria Brasil Engenharia de Produção UFSM Programa de Pós-Graduação em Engenharia de Produção Centro de Tecnologia |
publisher.none.fl_str_mv |
Universidade Federal de Santa Maria Brasil Engenharia de Produção UFSM Programa de Pós-Graduação em Engenharia de Produção Centro de Tecnologia |
dc.source.none.fl_str_mv |
reponame:Manancial - Repositório Digital da UFSM instname:Universidade Federal de Santa Maria (UFSM) instacron:UFSM |
instname_str |
Universidade Federal de Santa Maria (UFSM) |
instacron_str |
UFSM |
institution |
UFSM |
reponame_str |
Manancial - Repositório Digital da UFSM |
collection |
Manancial - Repositório Digital da UFSM |
repository.name.fl_str_mv |
Manancial - Repositório Digital da UFSM - Universidade Federal de Santa Maria (UFSM) |
repository.mail.fl_str_mv |
atendimento.sib@ufsm.br||tedebc@gmail.com |
_version_ |
1805922174099259392 |