Replacement Policy Model In Uncertain Environments

As the speed up of economy, how to improve the system reliability and sta-bility become an urgent need to study and solve. The maintenance costs of system often be much higher than the budget many times, due to poor reli-ability issues. In particular the sudden failure of the system, results in huge losses, and in some cases lead to disastrous consequences. To replacement the possible failure parts can significantly reduce system operating costs and improve system reliability.In this dissertation, both age replacement policy and block replacement policy in uncertainty evironment were discussed. The age-dependent replace-ment policy is one of the most commonly used strategy for system mainte-nance. A fuzzy age replacement model is eatablished, and a optimal renewal cycle exist theorem is also provided.In a practical system the fuzziness and randomness are often mixed up with each other. Thus, both of the two uncertainties should be considered simultaneously. The fuzzy random age-dependent replacement policy and random fuzzy age-dependent replacement policy are also studied here, and the concept of long-run expected cost per unit time is also extended. In order to facilitate the application, at the same time gives a more direct method to determine the existence of the optimal solution.The block replacement plocy is also a commonly used maintenace pol-icy. This paper extends the random block replacement policy to the fuzzy, fuzzy random and random fuzzy case and establishes the uncertainty block replacement models.In order to get the best solution of the proposed model and get the optimal solution, a new fuzzy simulation algorithm to calculate the expected value of fuzzy variables is presented, and based on the simulation technique, SPSA algorithm, PSO algorithm and a new hybride algorithm is employed. At the end of this dissertation, the numerical examples are enumerated, and solved through the proposed algorithm. The results illustrate the effec-tiveness of the models and algorithm.