Publication
To what extent is the implementation of tree-based models effective and feasible in predictive maintenance under Industry 4.0?
| datacite.subject.fos | Ciências Sociais::Economia e Gestão | pt_PT |
| dc.contributor.advisor | Fernandes, Pedro Afonso | |
| dc.contributor.author | Bernards, Martin Kairay | |
| dc.date.accessioned | 2023-06-26T11:16:32Z | |
| dc.date.available | 2023-06-26T11:16:32Z | |
| dc.date.issued | 2023-01-26 | |
| dc.date.submitted | 2023-01 | |
| dc.description.abstract | The technology available in Industry 4.0, combined with the constantly evolving machine learning techniques have allowed managers and stakeholders of the man ufacturing, transportation, and logistics sectors to predict the likelihood of their machines failing at an unprecedented reliability. Although Machine learning models have become highly complex in recent years, tree-based models starting from Deci sion Trees to Random Forests and boosting models, continue to exist in academia due to their feasibility, interpretation and even effectiveness. The purpose of this study is to grasp the relevance of such models in the real-world and whether they are worth investing into. To guide the process of building tree-based models, the Cross-Industry Standard Process for Data-Mining (CRISP-DM) will be utilized in order to understand the business as well as the data, prepare the data, create and evaluate the models, and build a deployment strategy in an iterative and flexible manner. A synthetic dataset that simulated the sensor data of a milling machine was used for the research, and the results through multiple evaluation techniques, indicated that the boosting model, XGBoost, outperformed the Random Forest, Decision Tree, and Logistic Regression models. Though the models from other re search outperformed XGBoost, however, XGBoost along with Random Forests are advised to still be taken under consideration due to their feasibility to be produced, trained, and interpreted, while still generating good results. | pt_PT |
| dc.description.abstract | A tecnologia dispon´ıvel na Ind´ustria 4.0, combinada com as t´ecnicas de apren dizagem de m´aquinas em constante evolu¸c˜ao, permitiram aos gestores e interve nientes dos sectores de fabrico, transporte e log´ıstica prever a probabilidade das suas m´aquinas falharem com uma fiabilidade sem precedentes. Embora os modelos de aprendizagem de m´aquinas se tenham tornado altamente complexos nos ´ultimos anos, os modelos baseados em ´arvores a partir de Arvores de Decis˜ao para Ran- ´ dom Forests e modelos de refor¸co, continuam a existir no meio acad´emico devido `a sua viabilidade, interpreta¸c˜ao e mesmo efic´acia. O objectivo deste estudo ´e captar a relevˆancia de tais modelos no mundo real e se vale a pena investir neles. Para orientar o processo de constru¸c˜ao de modelos baseados em ´arvores, ser´a utilizado o Processo Padr˜ao Cruzado de Minera¸c˜ao de Dados (CRISP-DM) para compreen der o neg´ocio, bem como os dados, preparar os dados, criar e avaliar os modelos, e construir uma estrat´egia de implanta¸c˜ao de uma forma iterativa e flex´ıvel. Um conjunto de dados sint´eticos que simulava os dados dos sensores de uma fresadora foi utilizado para a investiga¸c˜ao, e os resultados atrav´es de m´ultiplas t´ecnicas de avalia¸c˜ao, indicaram que o modelo de impulsionamento, XGBoost, superou os mod elos de Random Forests, Arvore de Decis˜ao, e Regress˜ao Log´ıstica. Embora os ´ modelos de outras pesquisas tenham superado o XGBoost, contudo, o XGBoost, juntamente com as Random Forests, s˜ao aconselhados a serem ainda tomados em considera¸c˜ao devido `a sua viabilidade de serem produzidos, treinados e interpreta dos, gerando ao mesmo tempo bons resultado. | pt_PT |
| dc.identifier.tid | 203278950 | pt_PT |
| dc.identifier.uri | http://hdl.handle.net/10400.14/41435 | |
| dc.language.iso | eng | pt_PT |
| dc.subject | Tree-based models | pt_PT |
| dc.subject | Industry 4.0 | pt_PT |
| dc.subject | Predictive maintenance | pt_PT |
| dc.subject | Decision trees | pt_PT |
| dc.subject | Random forest | pt_PT |
| dc.subject | Boosting | pt_PT |
| dc.subject | XGBoost | pt_PT |
| dc.title | To what extent is the implementation of tree-based models effective and feasible in predictive maintenance under Industry 4.0? | pt_PT |
| dc.type | master thesis | |
| dspace.entity.type | Publication | |
| rcaap.rights | openAccess | pt_PT |
| rcaap.type | masterThesis | pt_PT |
| thesis.degree.name | Mestrado em Análise de Dados para Gestão | pt_PT |