Survivability maximization for vulnerable multi-state systems with bridge topology

The multi-state system consisting of elements combined into a bridge structure is considered. Both the impact of external factors and internal causes can prevent the system from performing its task. When the system consists of elements with different reliability and performance (productivity or capacity) characteristics, its survivability, defined as the ability to meet a demand, strongly depends on the way the elements are distributed between components as a result of separation. In this paper we consider the problem of how to separate the elements of the same functionality between two parallel bridge components in order to achieve a maximal possible level of system survivability. The procedure developed to solve this problem is based on the use of a universal moment generating function (UMGF) for fast evaluation of multi-state system reliability and a genetic algorithm for optimization. Basic UMGF technique operators are developed to evaluate survivability of bridge structures for two different types of systems: capacity based and processing speed based. Examples of the optimal transformation of series–parallel structures into bridge structures are presented.