Structural Complexity Quantification for Engineered Complex Systems and Implications on System Architecture and Design

Abstract

The complexity of today's highly engineered products is rooted in the interwoven architecture defined by its components and their interactions. Quantitative assessment of structural complexity is mandatory for characterization of engineered complex systems. In this paper, we develop a quantitative measure for structural complexity and illustrate its application to a real-world, complex engineered system. It is observed that low topological complexity implies centralized architecture and it increases as we march towards highly distributed architectures. We posit that the development cost increases super-linearly with structural complexity. Empirical evidences from literature and preliminary results from simple experiments strengthen our hypothesis. Preliminary experiments show that the effort increases super-linearly with increasing structural complexity (i.e., exponent, b = 1.69). We further introduce complicatedness as an observer-dependent property that links structural complexity to system level observables like the development cost. We further discuss distribution of complexity across the system architecture and its strategic implications for system development efforts. © 2013 by ASME