Bottom-up models analyze risk in individual processes. Therefore, they are relatively complex (especially when processes are aggregated) and can differentiate between HFLS and LFHS events. Bottom-up approaches are very data intensive in relation to top-down models, which use aggregate data.
