Optimal Structure of Multi-State Systems With Uncovered Failures

The paper presents a model of multi-state systems in which some groups of elements can be affected by uncovered failures that cause outages of the entire group. It shows that the maximal reliability of such systems can be achieved through a proper balance between two types of task parallelization: parallel task execution with work sharing, and redundant task execution. The paper also suggests a procedure for finding the optimal balance for series-parallel multi-state systems. This procedure is based on a universal generating function technique, and a genetic algorithm. Illustrative examples are presented.