[ABE-L] Seminário Conjunto UFSCar/ICMC - 05/09/2014 - 14h00 - no ICMC-USP

[Carlos Diniz]

Seminário Conjunto UFSCar/ICMC – 05/09/2014 - 14h00

LOCAL: Sala 4005 - ICMC-USP  

TÍTULO: ARA and ARI Imperfect Repair Models: A Case Study in Industry

PALESTRANTE: Enrico A. Colosimo – UFMG

RESUMO: 

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 understand the behavior of the failure intensity for the systems under study. The usual assumptions of minimal or perfect repair at failures are not adequated for many real world systems, requiring the application of imperfect repair (IR) models. In this paper, the classes ARA and ARI of IR models proposed by Doyen and Gaudoin (2004) are explored. Likelihood functions for such models are derived, assuming Power Law Process and a general memory m. Based on this, punctual and interval parametric estimates were obtained for a real dataset involving failures in trucks used by a mining company for models ARAm and ARIm, m = 1; 2; 4; 6; 8; 12;1, and also for the model based on minimal repair assumption. The maximum of the likelihood function value was used as the criteria for model selection, and the best _tted model (ARI1) estimated parameters, namely, shape and scale parameters for PLP, and the e_ciency of repair parameter were obtained. They provided evidences that the trucks tend to fail more frequently over time, justifying the necessity for preventive maintenance, and also, that the repairs after failures tend to leave the equipment in a state between as good as new and as bad as old. These results are a valuable information for the mining company, and can be used to support decision making regarding preventive maintenance policy.

-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <https://lists.ime.usp.br/archives/abe/attachments/20140902/ec82d952/attachment.html>