Theoretical And Applied Research On The Preventive Maintenance Strategy Of Degradation System

The optimal replacement strategy of repairable system is a hot research issue in the field of reliability maintenance. The degradation is a complicated process to rely on many environmental factors. Increasing with age, in real life, most systems have different degrees of aging, wear and fatigue. The life time of the parts decrease after repair, and the repair time of the units increases after fault gradually, finally the parts cannot be work or repaired, this process was described as Monotone Geometric Process by Lam Yeh[1,2]. The optimal replacement policy is designed to determine an efficient maintenance plan for a system. Maintenance too much or insufficient, system maybe have more and more loss. However, if the maintenance frequency is too high, the reliability will be increased, but maintenance cost will be increase. Therefore, it is very necessary for system to determine a reliability maximum or maintenance cost minimum.Model I:Considering the system of single repairman vacation, the two optimal replacement strategy models are established. Using correlation theory, we derive to the corresponding availability and average cost. Maximizing availability and minimizing average cost rate, so as to obtain the optimal replacement policy. According to the actual requirements, we determine the corresponding weight ratio, and establish the tradeoff optimization model of availability and cost rate.Model Ⅱ:We research the two parameters replacement strategy of series repairable system. For unit 1, we use to maximize availability. As to part 2, we put forward a new standard to determine the optimal replacement strategy of the component 2. Firstly, we set up the model of λ2n= EX2n≤ a0 and μ2n= EY2n≥ b0. Secondly, on the basis of this, we build the fuzzy decision model of two factors of components 2. With the application of Gathering Statistics Iterative Method, we determine their weight. According to The Classic Comprehensive Judgment Decision Theory, we identify replacement policy N*2= A(u1)N*21+A(u2)N*22 of the component 2. Thus, we get replacement policy of the series system of two parameter.Finally, the model is simulated to verify the validity and feasibility of the results by MATLAB software theoretically, and the theoretical value and practical significance of the results are discussed.