Building them up, breaking them down: Topology, vendor selection patterns, and a digital drug market’s robustness to disruption

Criminal network security: An agent‐based approach to evaluating network resilience*

“…The overarching objective of this study was to use simulation methods to evaluate how criminal networks respond to disruptions. Although prior research has been aimed at evaluating the immediate structural effects of disruption (Bouchard, ; Bright, ; Duxbury & Haynie, ; Wood, ), very few scholars have considered the pace of network recovery after disruption (for exceptions, see Duijn et al., ; Mellon, Yoder, & Evans, ). From a crime control perspective, we find that criminal networks often take a long time to regain structural characteristics after disruption.…”

“…An important justification for studying criminal networks is to generate knowledge that can inform criminal intelligence (Kennedy, ; Malm & Bichler, ; McGloin, ; Sparrow, ). Yet, in many of the network‐based crime policy recommendations, degree centrality is assumed to be the best indicator of criminal network structural embeddedness (Bouchard, ; Duxbury & Haynie, ; McGloin, ). Our simulation results provide two insights that may help law enforcement more effectively disrupt criminal networks.…”

“…Various targeting strategies can affect networks in several ways. In prior studies, researchers have examined multiple consequences of network disruption, and so we consider three possible outcomes to evaluate network damage, all of which have been explored in prior research (Borgatti, ; Duijn et al., ; Duxbury & Haynie, ; Wood, ). Our first outcome is the mean geodesic distance in the network.…”

“…It is plausible to expect that some security-oriented networks take longer to recover from disruption compared with efficiency-oriented networks. First, when criminal networks prioritize security, they face substantial pressure to forge only highly trusted connections (Baker & Faulkner, 1993;Bright et al, 2018;Duxbury & Haynie, 2018;Gambetta, 2009;Smith & Papachristos, 2016) and to structure the network to minimize actors' ability to incriminate one another (Krebs, 2002;Morselli et al, 2007). Security-oriented criminal networks also have a tendency to protect the most important actors in the network, typically by limiting their visibility and minimizing their connections to other actors in the network (Breiger et al, 2014;Morselli, 2009).…”

“…The security–efficiency trade‐off has motivated much research on criminal network disruption, and in general, findings support the idea of a trade‐off (Bouchard, ; Bright, ; Duijn, Kashirin, & Sloot, ; Malm & Bichler, ; McGloin, ; Mellon, Yoder, & Evans, ; Wood, ). Yet, empirical and theoretical considerations of network resilience have almost exclusively been focused on network robustness without considering network recovery after disruption (Bright, ; Duxbury & Haynie, ; Malm & Bichler, ; Wood, ). Although important, a focus on robustness paints an incomplete picture of network resilience.…”