Abstract
Evaluating the system reliability of a stochastic network is an important topic in the planning, designing and control of systems. It is always desirable to minimise the resource consumption, e.g., the total cost, under the network reliability constraint in a real-life network problem. In this study, an intuitive Monte Carlo simulation (MCS) was first developed to find the estimated reliability w.r.t some specific combinations of the node reliability. Then, the response surface methodology (RSM) with the Box-Behnken design (BBD) was implemented to obtain the reliability function. Next, the proposed problem was modelled and solved by nonlinear programming. One example is given to illustrate the proposed MCS-RSM approach.