Resilience Assessment of Self-healing Systems with Redundancy

Abstract

There is a need to analyse systems using complex networks modelling. In this study, we use this modelling to peform the resilience analysis of self-healing systems. These systems have the ability to change their topology to recover themselves from a failure or to adapt to new situations. Also, the resilience of a reference system from the REDS repository and of some expansions of it with the addition of edges to the system using an Small-world topology method were analysed. An algorithm that simulates the reconfiguration of the system in case of failure was implemented. The analysis was performed for a variable number of edges' failures. The resilience investigation was done using the fraction of nodes still connected to an source after the system recovery over the total number of nodes in the system. A study of cost-benefit was also performed for one of the tested system expansions. The results show the benefits of adding redundancy to the system, which was expected, and show that the cost-benefit has a non-linear relation with the increase of redundancies in the system.