Testes de hipóteses para modelos de Reparo Imperfeito

• Ano de defesa: 2018
• Autor(a) principal: Daysemara Maria Cotta
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Engenharia de produção
• Link de acesso: http://hdl.handle.net/1843/BUOS-B6FHQ9

An appropriate maintenance policy is essential to reduce expenses and risks related to equipment failures. A fundamental aspect to be considered when specifying such policies is to be able to predict the reliability of the systems being studied, based on a well tted model. In this work, the classes of models Arithmetic Reduction of Age (ARA) and Arithmetic Reduction of Intensity (ARI) are explored. Likelihood functions for such models are derived. In developing methods that aim to determine the optimum periodicity of preventive maintenance interventions, one should assume which model best ts the reality of the analyzed data, so that the probability functions for such models are derived, and the parameters are estimated, allowing to calculate reliability indicators to predict future process failure behavior. Therefore, before calculating an optimal maintenance policy, it would be interesting to develop a general statistical test procedure in order to allow professionals to rst answer whether the data are under a minimum repair situation (ABAO eect) or a situation of imperfect repair. A set of real data involving pulp pump rotor failures used by a Brazilian mining industry is analyzed considering models with dierent memories. The exact binomial and multinomial tests were applied in the data, as well as the estimated form and scale parameters for PLP and the repair eciency for dierent memories, which allowed to apply the model selection tests based on the maximum log-likelihood, on the weight of the evidence and in the goodness of t graphic method. The estimation of the parameters of the best adjusted model allowed us to calculate the optimal periodicity of preventive maintenance. These results are a valuable information for the mining company and can be used to support in decision making.