Risk Propagation Simulator for Assessing Urban risk

Munasinghe, Dayan

Abstract

The Intergovernmental Panel on Climate Change (IPCC) report on "Climate Change and
Land" (2020) underscores how climate change intensifies disaster risks such as floods,
landslides, heatwaves, wildfires, and storm surges, significantly impacting economies,
infrastructure, and food and water security. In response, robust approaches to
understanding and mitigating local risks from climate change are essential. Sound risk
assessment methods enable stakeholders to pinpoint and implement targeted interventions,
thereby preventing the erosion of hard-earned economic gains and enhancing climate
resilience. This proactive approach is endorsed by reports from the IPCC, UNDRR, World
Bank, and OECD, highlighting the critical importance of early interventions in risk reduction
for safeguarding economic progress and fostering resilient, sustainable communities.
There have been numerous efforts to develop frameworks for assessing risks, such as the
ISO 31000 risk management framework and Sendai Framework for Disaster Risk Reduction.
These frameworks utilise various quantitative and qualitative methods. However, these
frameworks often fail to address the complex interdependencies of risks, mainly how risks
propagate and intensify due to the risk behaviour and perception of a local community. As a
result, the cascading impacts due to socio-technical interdependencies remain
underexplored due to their complexity and the extensive data requirements.
Hence, considering the community risk principles, there is a critical need to develop a risk
assessment model that captures these complex interdependencies and cascading risks. This
research, therefore, investigates a risk assessment model that deploys the system dynamics
techniques to identify and model how various community risk principles interact with each
other to influence the overall risk state of a given community. The resulting model allows the
decision-makers to conduct "what-if" scenarios, using the proposed risk model, to explore
how local climate risks can be reduced by addressing various community characteristics such
as risk perception, risk knowledge, risk understanding, risk communication to reduce overall
local risks and hence build community resilience.
The research deploys the design science approach to develop the interactive risk model
following the guiding seven design science principles. The development of the risk model
xvii
commenced with an exhaustive literature survey to identify common risk elements utilised
in risk assessment models, including those pertinent to communities. Subsequently, another
comprehensive literature review was undertaken to pinpoint the principles of risk
propagation and methodologies for their modelling. These risk elements were carefully
analysed, and their interconnections were delineated and modelled using System Dynamics
(SD). The SD model aimed to encapsulate socio-technical risk interdependencies, thereby
establishing a comprehensive risk model. This holistic modelling approach not only
facilitated the identification of multiple risk reduction pathways crucial for stakeholders
collaborating with communities to mitigate local risks but also enabled sensitivity analysis to
comprehend the influence of each risk variable on the overall risk status of a given
community.
The System Dynamics (SD) model developed in this study offers a tool for unravelling the
intricate interdependencies among diverse risk elements and their propagation dynamics.
This research uncovered fourteen distinct risk reduction pathways through system analysis
and outlined four pivotal risk propagation principles. Leveraging the capabilities of AnyLogic
software, this research operationalised the SD model to simulate risk propagation within a
local community and explore various "what-if" scenarios. The SD approach was found to
foster a nuanced understanding of how individual risk elements interact with each other and
offer a tool to evaluate the potential cascading impact of risks, providing invaluable insights
to the decision-makers to devise effective risk mitigation strategies. Before validating the SD
model with the users, the representation of the causal loop diagrams was verified with the
support of two SD experts. The mathematical constructs used to model the SD model for
each loop were then verified by two mathematicians. This verification ensured the validity of
the SD model for simulating the cascading impact of risks.
The evaluation of the SD model was conducted with twelve experts using two proposed
development programs in Kalutara: one focused on risk communication improvement and
the other on drainage improvement to fortify resistance. User evaluation was carried out to
assess the ability of the System Dynamics model to aid decision-makers in comprehending
the impact of their interventions on community resilience.
xviii
All experts agreed that the model effectively captures the relationships between key risk
principles. Additionally, they found the model valuable for exploring how various
interventions impact community resilience over time. The model's utility in the planning
process was widely recognized, as it helped stakeholders understand how different
interventions contribute to building resilient environments. Based on stakeholder feedback,
future improvements have been suggested, and the researcher plans to extend the study by
addressing these emerging requirements.

• Thesis Type: Thesis
• Online Publication Date: Mar 27, 2025
• Deposit Date: Mar 17, 2025
• Publicly Available Date: Apr 28, 2025
• Award Date: Mar 27, 2025