Advancing Reliability, Maintainability, and Availability Analysis Through a Robust Simulation Environment
Abstract
The field of reliability has become a dominant factor in industry as companies strive for competitive advantages in the market place. Reliability, maintainability, and availability (RM&A) analysis provides insight into system characteristics and overall behavior, describing parameters such as time between failures, mean down time, and cost estimation. For simple systems, analytical methods can calculate exact solutions. Complex systems, however, require different methods for approximating solutions. Simulation is one solution for evaluating system performance and advancing the understanding of RM&A metrics. The first phase of this research focuses on how data requirements for input can affect simulation results. When failure data is sparse, the triangle distribution can replace the Weibull distribution to describe component failure rates. This reduces the complexity of the modeling process while returning similar output analysis. The research results indicate that the availability metric for system performance is insensitive to input distribution specification. The input parameters for the triangle distribution, which replace the true Weibull failure distribution, can be varied as much as 30% with little impact on system availability. With respect to mean time to first failure, it appears that overestimating the triangle mode and underestimating the worst case failure (maximum parameter) provides accurate prediction of system performance. Overall, the triangle distribution offers a valuable alternative to the Weibull distribution when little or no data is available. The next phase of this research uses simulation results as the basis for comparing sparing alternatives when evaluating competing systems. Sparing strategies are compared on the basis of availability and cost. Initial simulation results provide three alternatives for supporting a simple communications network.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 02, 1999
- Accession Number
- ADA366387
Entities
People
- Stephen Paul Chambal
Organizations
- Arizona State University