Complexity Quantification in System-of-Systems Architectures:
Wildfire Response Applications
Name
cadd-tcadd-masters-aeroastro-2026-thesis.pdf
Size
36.24 MB
Format
Adobe PDF
Checksum (MD5)
50a9ca12021833d4dae8bf0a484c0f14
Author(s)
Cadd, Terelle A.
Advisor(s)
de Weck, Olivier L.
Date Issued
May 2026
Publisher
Massachusetts Institute of Technology
Abstract
Modern system-of-systems architectures involve tightly coupled interactions across physical, functional, and organizational domains, yet their structural complexity is often introduced without a clear means of quantification or direct linkage to system performance. This limitation makes it difficult to evaluate how architectural decisions influence system effectiveness in large-scale, multi-domain environments. In this work, a generalized framework is developed for representing and analyzing system-of-systems architectures using a Multi-Domain Matrix (MDM), combined with a unified formulation for structural complexity and performance evaluation. The framework extends existing graph-based complexity formulations by incorporating typed multi-domain interfaces and enabling exploration of architectural trade spaces. The approach is applied to wildfire response systems and evaluated using a coupled MATLAB, Systems Tool Kit (STK), and geospatial modeling environment. Across more than 45,000 candidate architectures, multiple wildfire response contexts are evaluated, including the Camp Fire incident in Butte County as a high-resource case, the Shelburne Fire representing a multi-organizational international coordination scenario, and the Etosha National Park Fire as a resource-constrained environment. Performance is assessed using resolution time, burn burden, cost, and CO₂ emissions, with results demonstrating substantial variation in burned area, estimated CO₂ output, suppression resource requirements ranging from tens to hundreds of assets, and containment times spanning days to weeks depending on architectural configuration. Within this framework, structural complexity is directly coupled with performance evaluation, allowing complexity to be interpreted in terms of its impact on system outcomes. The results show that performance depends strongly on architectural structure and coordination, not solely on resource availability, with increased complexity often introducing diminishing returns due to coordination overhead. In many cases, near-optimal performance is achieved within lower-complexity regions of the design space, highlighting the importance of efficient system configuration. By linking complexity to performance, the framework also serves as a diagnostic tool for identifying inefficiencies and coordination challenges that are not readily observable through performance metrics alone, supporting more informed design and decision-making in complex engineering applications.
MIT Department
Massachusetts Institute of Technology. Department of Aeronautics and Astronautics
Terms of Use
In Copyright - Educational Use Permitted
Copyright retained by author(s)
Persistent DSpace Link
