PEER REVIEW
20
th
Australian International Aerospace Congress, 27-28 February 2023, Melbourne
Handling interconnected cascading Risks: an interoperable
holistic Framework
John Wondoh
1,2,*
, Karamjit Kaur
1,2
, Matt Selway
1,2
, Mansi Patel
1,2
, Andrew McRae
3
, Georg Grossmann
1,2
, Markus Stumptner
1,2
, Don Sands
3
and Alan Johnston
4
1
Industrial AI Research Centre, University of South Australia STEM, Australia
2
Future Energy Exports Cooperative Research Centre, Perth, Australia
3
Synengco Pty Ltd, Queensland, Australia
4
MIMOSA, Tuscaloosa, Alabama, USA
1
firstname.lastname@unisa.edu.au
3
firstname.lastname@synengco.com
4
atjohn@mimosa.org
Abstract
In critical industries such as Energy or Defense that consists of complex interconnected systems,
risks generally do not occur in isolation and are dynamic in nature. Traditional risk management
frameworks do not possess the capability to reflect the cumulative effect of dynamic
interconnected risks. Whilst the individual risks arising at component level from an asset health
monitoring may not be regarded as significant based on their likelihood and impact, the
assessment may be completely different when this analysis is aggregated and cascaded to the
system level by assessing the connected components. The inability to assess the cascading
impact of risk events within and across the system boundary as well as across the different
levels and types of risks is hampering decision makers to produce effective decisions that can
control the expected risks and identify emergent risks.
Within a complex system, multiple analytical models and techniques can be employed
to perform analysis on a component for different contexts such as reliability, efficiency and
safety. For example, health and monitoring systems generate risk events based on an asset’s
health which are assessed by risk models for their impact. The assessments produced by
individual risk models need to be shared and propagated across siloed systems and multiple
system levels in the hierarchy in an interoperable manner. We provide details of a framework
which can enable organisations to attain a holistic view of the system, by sharing the outputs
produced by various risk models and analysis methods in a standard format that can be used by
other risk models and systems. An extra layer of assessment can thus be performed on top of
existing independent analysis, that can empower organizations to have a more coherent and
comprehensive view of their asset status. This is essential to avoid potentially major oversights
across the systems engineering lifecycle.
Keywords: cascading risk, interconnected risk, interoperable analytics, risk model.
Introduction
Risk management is necessary for the reliable operation of large-scale critical infrastructure,
including energy and defense infrastructure and ecosystems. These infrastructures are
characterised by their complex structures, which typically consist of an aggregation of systems
operating within and across organisational units [1]. In addition, an engineering infrastructure
may interact with external systems by contracting their ability to satisfy operational goals.
Therefore, critical engineering infrastructures must be robust and uninterruptedly provide their
functions.
Due to their complex structures and interactions, critical infrastructures are prone to disruptive
failures resulting from cascading failures if risks are not detected early, their impact evaluated
and mitigated before their actual occurrence [2,3]. The operational independence property of
