Thesis

A multi-agent approach to support decision making uncertainty : optimising climate adaptation and resilience measures for the finfish aquaculture sector in Scotland

Downloadable Content

Download PDF
Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2025
Thesis identifier
  • T17238
Person Identifier (Local)
  • 201961716
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • Understanding and decision-making in complex systems can be significantly enhanced by decisionsupport tools that elicit and present data in an informative, accurate, applicable, and strategic manner (Wagener et al., 2022). Despite the widespread adoption of such tools, a gap persists: no existing tool effectively integrates multiple stakeholder perspectives, quantifies their values, and simultaneously addresses uncertainties in decision-making processes. This PhD research addresses this gap by developing a novel methodology and decision-support tool, SEAD (Stakeholder interaction, Elicitation, Analysis, Dialogue). By leveraging Multi-Agent Influence Diagrams (MAIDs) derived from Bayesian Networks, the SEAD methodology facilitates structured, quantified, and participatory decision-making processes. The innovation of SEAD lies in its adaptability across multistakeholder scenarios, particularly those characterized by significant uncertainty, such as climate change adaptation. The thesis is organized to systematically address the research objectives. Chapter 2 reviews existing decision-making methodologies and identifies critical gaps, establishing the need for a tool like SEAD. Chapter 3 details the conceptualization and development of the SEAD methodology, emphasizing its novel integration of MAIDs for scenario evaluation and stakeholder engagement. Chapter 4 investigates climate change vulnerabilities within the freshwater aquaculture sector, providing the contextual foundation for SEAD’s application. Chapters 5 through 8 present the case study application of SEAD to the Scottish freshwater finfish aquaculture industry, demonstrating its capacity to optimize decision-making under climate pressures and to identify adaptation and resilience measures. Finally, Chapter 9 explores broader adaptation and mitigation pathways, extending SEAD’s implications beyond the case study. The freshwater aquaculture sector was chosen as a case study due to its economic importance and vulnerability to climate change. Salmon is Scotland’s leading food export, with exports worth £578m in 2022 (Salmon Scotland, 2023). The industry, comprising 108 freshwater aquaculture sites (Aquaculture Scotland, 2024), faces significant risks from climate change, including water shortages and temperature fluctuations. Adaptation decisions require input from diverse stakeholders— including insurers, regulators, and industry bodies—each with distinct priorities and risk perceptions. Such complexity often hampers timely and effective decision-making. The SEAD method addresses this challenge by enabling real-time, participatory assessments of stakeholder values and priorities against customizable climate risk scenarios. In the case study, the tool engaged three key stakeholder groups from the aquaculture sector: insurance providers, regulatory bodies, and industry representatives. These stakeholders used SEAD to evaluate risks, explore cooperative pathways, and identify priority areas for intervention. The process not only facilitated consensus-building but also highlighted specific adaptation and mitigation measures. Stakeholders found SEAD to be both innovative and practical, praising its ability to provide structured and actionable insights. However, feedback indicated that future development should streamline the methodology and enhance the usability of the tool’s interface. The next phase of this research involves applying SEAD to specific sites to optimise climate resilience interventions and assess the tool’s effectiveness. Further studies will explore variability within stakeholder groups by engaging different participants and extend the SEAD methodology to other sectors and applications, broadening its utility in addressing complex, uncertain decision-making scenarios.
Advisor / supervisor
  • Roberts, Jennifer
  • Quigley, John
Resource Type
Note
  • This thesis was previously held under moratorium between 26 February 2025 and 26 February 2026.
DOI

Relations

Items

Thumbnail Title Date Uploaded Visibility Actions
file details PDF of thesis T17238 2025-04-15 Public Download