Understanding how and why changes propagate during engineering design is critical because most products and systems emerge from predecessors and not through clean sheet design. This paper examines a large data set from industry including 41,500 change requests that were generated during the design of a complex sensor system spanning a period of 8 years. In particular, the networks of connected parent, child, and sibling changes are resolved over time and mapped to 46 subsystem areas of the sensor system. These change networks are then decomposed into one-, two-, and three-node motifs as the fundamental building blocks of change activity. A statistical analysis suggests that only about half (48.2%) of all proposed changes were actually implemented and that some motifs occur much more frequently than others. Furthermore, a set of indices is developed to help classify areas of the system as acceptors or reflectors of change and a normalized change propagation index shows the relative strength of each area on the absorber-multiplier spectrum between −1 and $+1$. Multipliers are good candidates for more focused change management. Another interesting finding is the quantitative confirmation of the “ripple” change pattern previously proposed. Unlike the earlier prediction, however, it was found that the peak of cyclical change activity occurred late in the program driven by rework discovered during systems integration and functional testing.

This study is an overview of network topology metrics and a computational approach to analyzing graph topology via multiple-metric analysis on graph ensembles. The paper cautions against studying single metrics or combining disparate graph ensembles from different domains to extract global patterns. This is because there often exists considerable diversity among graphs that share any given topology metric, patterns vary depending on the underlying graph construction model, and many real data sets are not actual statistical ensembles. As real data examples, we present five airline ensembles, comprising temporal snapshots of networks of similar topology. Wikipedia language networks are shown as an example of a nontemporal ensemble. General patterns in metric correlations, as well as exceptions, are discussed by representing the data sets via hierarchically clustered correlation heat maps. Most topology metrics are not independent and their correlation patterns vary across ensembles. In general, density-related metrics and graph distance-based metrics cluster and the two groups are orthogonal to each other. Metrics based on degree-degree correlations have the highest variance across ensembles and cluster the different data sets on par with principal component analysis. Namely, the degree correlation, the s metric, their elasticities, and the rich club moments appear to be most useful in distinguishing topologies.