Exploring Structural Patterns Across Evolved and Designed Systems: A Network Perspective

Abstract

General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/about/ebr-terms Abstract -Our desire to deliver increased functionality while setting tighter operational and regulative boundaries has fueled a recent influx of highly-coupled systems. Nonetheless, our current capacity to successfully deliver them is still in its infancy. Understanding how such Designed systems are structured, along with how they compare to their naturally Evolved counterparts, can play an important role in bettering our capacity to do so. Based on this premise, the structural patterns underlying a wide range of seemingly unrelated systems is uncovered using tools from network science. By doing so, structural patterns emerge and are subsequently used to highlight both similarities and differences between the two classes of systems. With a focus on the Designed class, and assuming that increased structural variety fuels design uncertainty, it is shown that their adherence to statistical normality (i.e. expected vs. encountered patterns and statistical correlations between combinations of such patterns) is rather limited. Insight of this sort has both theoretical (context agnostic approaches are increasingly relevant within the domain of Systems Engineering, yet are rarely used) and practical (transferability of knowledge) implications.

Uncovering the fragility of large-scale engineering projects

“…Despite these differences in size, these networks exhibit shared properties: they are sparse (densities span 10 -3 -10 -5 ), and while their average path lengths are similar to random expectation, they are in general more highly clustered than expected by chance (10 out of 14 networks, Figure S3c,d). As such, activity networks are 'small-world' (Table 1)-a finding in step with prior work on project networks reported in [38,39]. Finally, their local structure, assessed through the variation of number of dependent activities or 'degree' of an activity, is strikingly similar: we observe that in 78% of the Activity networks of all projects (project 1 top-left to project 14 bottom-right).…”

Modelling indirect interactions during failure spreading in a project activity network

“…In terms of applications, our work provides the grounds for a dialogue between researchers in the network science and project management, where hotly-researched, domain challenges (e.g. project complexity evaluation [52][53][54][55][56][57] ) can be treated as network-related problems 31,32 . For example, increased susceptibility to the spreading of failures can be reasonably interpreted as a contributing factor to project complexity.…”

“…In conjunction with the fact that activity networks have a high concentration of local, non-trivial structures 32 and hence, experience pronounced levels of subsequent exposure (see Supplementary Fig. 1) -the likelihood of obtaining misleading results through the use of typical spreading models is high.…”