Optimization of System Reliability

As described in Chap. 5, reliability of a multicomponent system depends on the component reliability, number of redundant components, and the arrangement of the components. With everything else remaining fixed, the system reliability gets higher as the number of redundant components increases. However, noth­ing is free—achieving higher system reliability has to be paid for by a higher price for the system. In general, practically all systems exhibit a behavior of diminishing rate of return as the number of redundant components increases (see Example 5.2). In the context in system reliability, this behavior is describ- able by a strictly concave function relation between the system reliability and number of redundant components (or total system cost). A relevant issue is how much one is willing to invest before the improvement in system reliability is no longer economically justifiable. The problem becomes even more challenging when there are several components with different reliability levels competing for the limited resources, such as budget and space. How would an engineer decide the best allocation of resources to achieve the highest system reliability possible? This section describes several examples showing how system reliabil­ity can be optimized by various techniques introduced in Sec. 8.1.