The R&M performance of a system is usually measured in terms of its Reliability (Ro), Availability (Ao), and Maintainability (MTTR and DMC) During the design phase of a system, several choices need to be made, such as number of redundant modules required to achieve a desired performance level (Reliability, Availability etc.), the system configuration or architecture, performance specifications of the chosen modules comprising the system, and system attributes such as weight & power. The maintenance of the system during its operational life requires labor cost associated with maintenance man hours expended during SMAs and cost of parts & materials used during such actions. In addition to cost associated with SMAs, there will invariably be system failures occurring randomly, causing downing events or mission aborts. Under such circumstances UMAs are required to bring the system back to its operational status. UMAs usually cost much more to fix as compared to SMAs. Both actions (UMAs & SMAs) contribute to the DMC of the system. While it is possible to set up a Reliability Block Diagram to estimate the reliability of the system, very often merely estimating the reliability is not enough. A reliability engineer may equally be, or more interested in the availability of the system, the frequency of SMAs and UMAs required for a given configuration to ensure that the system is available; and the direct operating cost of which DMC is an integral part. A given system may also be used for multi-mission purposes requiring different mission times. Adding to the system complexity are uncertainties associated with estimated maintenance man hours required for overhauling, uncertainty in cost of labor, cost & quantity of parts and materials used during maintenance actions. One may be able to decompose the complex problem into various scenarios, model and simulate (or calculate, if possible) one scenario at a time to estimate its performance parameters. However, such an approach can be tedious and time consuming. In this paper we demonstrate a simulation model that can address all of the above mentioned concerns simultaneously. An RBD of the system is first transformed into a simulation model. The MTBF, MTTR and Repair Costs based on historical data are used as inputs to the model, with appropriate statistical distributions respective to these parameters. The scheduled maintenance or replacement intervals (SMI) of various modules are entered as discrete variables into the model. The multi-stage Monte Carlo Simulation is modeled to yield the number of UMAs, Reliability, Availability, MTBMiA and DMC for various SMIs or replacement Intervals. The output results of the simulation are then analyzed using a statistical software package to yield response surfaces and/or contour plots. Such results can then be used to determine the best possible combination of system architecture and maintenance strategy capable of delivering optimal R&M performance of the system under investigation. [ABSTRACT FROM PUBLISHER]