Abstract
Enhancement of the survivability of a system can be achieved by separating its elements. Since different elements can play different roles in fulfilling the system mission, the way in which they are separated strongly affects the achieved level of the mission survivability. Thus, the optimal separation problem arises. This paper formulates a new problem of optimal element separation in non-repairable phased-mission systems (PMSs) where the mission consists of multiple, consecutive and non-overlapping phases of operation. An accurate survivability analysis of a PMS must consider the statistical dependence of element states across phases as well as dynamics in system structure function and element failure behaviour. We suggest a method for finding the separation of the system elements that provides a maximal possible level of the PMS survivability while satisfying a certain separation cost constraint. A backward recursion algorithm is applied for determining the survivability of a vulnerable PMS. A genetic algorithm is used as an optimization tool in solving the newly formulated separation problem for PMSs.