Browsing by Author "Span, Martin "Trae", author"
Now showing 1 - 1 of 1
- Results Per Page
- Sort Options
Item Open Access Eliciting cybersecurity goals for cyber-physical system conceptual design(Colorado State University. Libraries, 2025) Span, Martin "Trae", author; Daily, Jeremy, advisor; Bradley, Thomas, committee member; Simske, Steve, committee member; Wise, Dan, committee memberThis research contributes to the systems engineering body of knowledge by advancing security by design for Cyber-Physical Systems (CPS). It leverages Systems Thinking and Model-Based Systems Engineering (MBSE) methodologies to address both organizational and technical challenges in early-stage secure system development. The research is structured around two primary themes: (1) What recommendations can improve CPS Design Teams with respect to security? and (2) Proving secure system design be improved through early system security goal elicitation. To address the first research question, a systematic analysis utilizing Systems Thinking tools, such as iceberg models, causal loop diagrams, and system modeling, is conducted. These analyses identify the root causes of weak security design within CPS development teams, revealing systemic organizational challenges, ineffective mental models, and gaps in team member knowledge skills and abilities. The research presents targeted recommendations to enhance security considerations within design teams by implementing Systems Thinking principles, refining organizational structures, and prioritizing security training. However, findings indicate that training alone is insufficient for achieving secure CPS design, necessitating a more structured approach to security design consideration elicitation in early system development. The second research question is answered with the development of Eliciting Goals for Requirement Engineering of Secure Systems (EGRESS), a novel methodology designed to facilitate system security goal elicitation during the conceptual design phase of CPS. By addressing a critical gap in current systems engineering practices, EGRESS provides a structured and traceable approach to defining security goals before an architecture is established. This method incorporates best practices from Systems Thinking, loss-driven engineering analysis, and MBSE to ensure security is foundational in CPS design rather than an afterthought. Furthermore, the research evaluates the applicability of the Risk Analysis and Assessment Modeling Language (RAAML) standard for cybersecurity and proposes refinements to enhance its utility for security analysis in CPS design. The key contribution of this work utilizes Popper's falsification principle to evaluate the hypothesis that secure system design can be improved through early security goal elicitation. Given the lack of long-term operational data proving increased security over a system's lifecycle, falsification serves as a rigorous alternative by testing for refutation rather than statistical verification. The research demonstrates that EGRESS cannot be falsified, supporting its validity in improving secure system design. This claim is further reinforced through peer-reviewed evaluations and expert discussions within the system engineering and security communities, where, through publication, the methodology's utility was recognized and endorsed. Beyond methodology development, this research contributes to the broader systems engineering body of knowledge by addressing the distinction between requirements and security-focused system goals. It also explores the balance between common and custom SysML profiles to improve security goal elicitation. These contributions collectively support the advancement of more resilient and secure CPS architectures, aligning with the broader vision of integrating security as a fundamental design consideration alongside functionality and safety.