Abstract (EN):
In a very complex technical industrial system, sophisticated methods of life data analysis are required to understand its technical behavior. To model this life data, the development of models and methods is needed, and these should take into account the random behavior of such industrial equipment, depending on its working conditions during the life cycle. The type of mathematical formulations used for modeling requires a preliminary analysis of the available data. In this paper a framework to analyze the field data related with equipment failures during its lifetime is presented. The purpose of this framework is to understand the equipment behavior and to estimate the failure pattern, so its maintenance management can be adjusted and better planned. This paper proposes to review methodologies to analyze preliminary maintenance life data. The methodologies are based on statistical models and techniques, also most practical problems in reliability data analysis involve censored data and missing information. Life data from centrifugal pumps that are in petroleum refinery were used to illustrate the framework capabilities.
Language:
English
Type (Professor's evaluation):
Scientific