Interdicting restructuring networks with applications in illicit trafficking

Shortest path network interdiction with asymmetric uncertainty

“…Our model is a variant of the shortest path interdiction problem [19]. The SPI considers a directed network G(N, A), where N is the set of nodes and A is the set of arcs.…”

“…In effect, the interdictor and the evader use different realizations of the uncertainty, which we denote v(I) and v(E). For a given x, we can express the evader's problem as follows: min y v(E) T R T Xy subject to y satisfies (7e), (7h), (7i) v(E) ∈ arg max{v T R T Xy ∶ v ∈ U} or equivalently min y ||R T Xy|| subject to y satisfies (7e), (7h), (7i) (19) , where the equivalence follows from the optimal choice for v(E) given x and y. We can then express the interdictor's problem as a pessimistic bilevel problem: max subject to (7e), (7h), (7i) (22) which is equivalent to our original optimistic formulation (8a-l).…”

“…This extension introduces asymmetric information where the interdictor has different (and likely more accurate) information about the network parameters as compared to the evader. The authors of [20] study a max flow network interdiction problem where arcs can be added to the network after the interdiction decisions have been made. They apply this new model to disrupt illicit drug and human trafficking networks.…”

Operations research and analytics methods to improve equitable access to public services

“…Research related to OR and analytics efforts to reduce homelessness and the supply of sex and labor trafficking victims has only recently received attention and largely focus on three broad areas: the scope of the problem [129,34], frameworks for addressing the crisis [204,131], and appropriate allocation of scarce resources to combat HT and homelessness [40,177]. A second body of OR literature focuses on detection of hidden HT victims [117,109,154], movement patterns of covert traffickers [152,234,9] and long-term intervention approaches to prevent trafficking or retrafficking by increasing the efficiency and effectiveness of public services [8], as well as improving access to supportive and rehabilitative services [24].…”

Tri-level optimization for building supply chain resilience with safety stock against a worst-case disruption

“…Zeng and An (2014) formalize the algorithm for bilevel optimization. This method is applied by Kosmas et al (2023), who present a max flow network interdiction problem where defender is able to introduce new arcs to the network after a disruption, and Cheng, Adulyasak, and Rousseau (2021), who provide a model to determine facility locations and optimal inventory allocation to customers under stochastic customer demand and worst-case facility disruption.…”