Análise de risco e da confiabilidade em sistemas complexos e reparáveis

Abstract

The failure detection and analysis are common procedures in system engineering. The area of engineering for the failure analysis and its consequences is called Reliability Engineering. Reliability can be defined as the resistance to failure of an item or system over time. Another methodology is the risk management. This is a systematic process in order to maximize the positive events and minimize the consequences of negative events. Based on these concepts this work aims to define the guidelines for failure analysis in a complex system based on analysis of Risk and Reliability. Thereunto, the time to failure of equipment is considered a random variable with Weibull’s distribution. It is assumed the minimum share of repair and use the equation that describes a Non-Homogeneous Poisson Process (NHPP) and the maximum likelihood estimation to estimate the Reliability Mission of a complex repairable system. The system risk is measured by the Risk Priority Number (RPN) that is the dot product of the Probability of Occurrence, the Detection and the severity of the failure. The probability of the failure occurrence is estimated by the complementary function of the Reliability Mission. To obtain the severity and the detection is used categorization rules guided by a priori knowledge of the system. This allows the analysis of the reliability of a system is jointly given by reliability mission that is a objective score and RPN that has a portion given by objective scores (Probability of occurrence of the fault) and two subjective scores (detection and severity of failure). Examples are presented to validate the proposed methodology to jointly analysis Reliability Mission and RPN and illustrate its applications. In addition, a case study is presented which assesses the reliability of a Ship-to-Shore crane which is used in container handling at port terminals.