Risk analysis is a very broad field, utilizing a variety of quantitative approaches. In the current context, however, we are primarily concerned with risk analysis of complex engineered systems (e.g., nuclear power plants, infrastructure such as dams, and space and defense systems) that are composed of highly-reliable and frequently redundant components, which in most cases are required to have an extremely low risk of a catastrophic failure.
The conventional approach to risk analysis for such systems focuses on the analysis of initiating events and subsequent event sequences that could lead to failures, and on enumerating and calculating the probabilities of different outcomes through tree-based analytical procedures (event trees/fault trees). Stamatelatos et al, (2011) and Vesely et al. (2002) provide good descriptions of these approaches.
For many types of systems (e.g., nuclear power plant probabilistic risk assessments), these approaches work well. However, systems that are highly dynamic or have significant process variability can be very difficult to model realistically using event tree/fault tree approaches, and they require a tremendous amount of preprocessing effort.
As a result, an approach like GoldSim's that facilitates explicit representation of dynamics and variability potentially provides a powerful complement to existing methods.