Search Results

This work presents an application of the simulated annealing algorithm to redundant system reliability optimization, and its objective is to analyze this method and compare to other optimization methods, which are Genetic Algorithm, Evolution Strategy and Lagrange Multipliers. A hidrid optimization algorithm, composed by Simulated Annealing and Genetic Algorithm, is also developed in this work, to conciliate the advantages of both methods. The results presented by the Simulated Annealing and the hybrid algorithm are significant and validate these methods as robust tools for parameters optimization in engineering systems developments

Due to changes in competitiveness attributes, consequence direct of technology development, new markets and customer profile, it has been required continuous improvement in the tools and methodologies of Quality and Productivity management to maintain the organization's health. These are new requirements for project managers that facing the challengers in the 21st Century in Automotive Industry. The Lean Manufacturing was a great innovation in this area because improve the organizations profitability, which is necessary for they survive in the competitive new scenarios that we are currently confronting. Allied with these new constraint conditions, the Six Sigma Methodology helps the organizations to improve the Value Add within the business process and to avoid wastes, which are responsible for lack of focus, increase of process costs, decrease of customer satisfaction and consequently lack of competitiveness.

The dependence among the components and the data associated with a repair politic, that can consists of the number of available teams and the components repair priority, rarely evaluate the measures of reliability. Thus, to evalute the reliability and availability values with these conditions, this article develops a methodology using the theory of Markovian stochastic processes. A system of ordinary differential equations is obtained using the concepts of component failure and repair rates, constants or variable in the time. Solving it, these values are obtained, allowing the calculation of the MTTF of the analyzed system, for an allowed error.

This paper discusses some aspects of durability tests in laboratory, justifying the necessity of a new procedure for data compression. Attention is paid to dynamic tests, with the presence of inertia forces. The new procedure is introduced, which uses concepts of fatigue due to vibration and signal analysis. The compression procedure is applied to generate a synthetic signal for the experimental stress test of an automotive component.

The reliability optimization of redundant systems aims to improve some design parameters as reliability or cost, using optimization methods, in order to determine the optimum number of redundancies in each component of the system. This work compares some optimization methods (Lagrange Multipliers, Evolution Strategy and Genetic Algorithm) and its objective is to determine which method is more adjusted to the problem.