Multi-criteria ranking across importance measures for stochastic networks

The identification of the components important to the performance of a system provides insight into decisions such as which components to protect against failure or where to add redundancy to ensure system function. However, there are many perspectives on “importance,” including consideration of purely topological characteristics or consideration of how components enable flow in a network. And even within these broad perspectives, there are different ways to quantify importance. With this in mind, we offer a means to aggregate across different perspectives of importance to provide a more holistic ranking of components. We specifically consider networks with a stochastic nature to their capacities, as the calculation of importance can change based on such variability. We make use of a multi-criteria decision analysis technique under uncertainty, Fuzzy TOPSIS, to aggregate different importance measures given variability in network link capacities. Our proposed approach is illustrated with a case study based on the Colombia highway transportation network.