The MEND project is focused on creating and validating network interdiction models with powerful insights into how to best disrupt and dismantle, as opposed to simply displace, sex trafficking operations that account for: (a) how the network reacts to the disruptions, (b) how this reaction impacts victims, and (c) how such disruptions impact the forced criminality of victims. The MEND project created a transdisciplinary, survivor-informed research process to build operations research models related to sex trafficking networks that integrates qualitative and quantitative research methods. This process has resulted in some of the first operations research models of sex trafficking created through co-production of knowledge with survivor-leaders and other lived experience experts.
The MEND Project is supported by the National Science Foundation through grant number 2039584. The views and conclusions presented here are those of the project team and should not be interpreted as reflecting the views of the National Science Foundation.
Research Overview
Computational models of sex trafficking are abstracting traumatic experiences of victims. Our team recognizes that we cannot possibly capture all the complexities of the lived experiences of victims. We seek to build operations research models to help disrupt sex trafficking in ethical, responsible manners.
We feel that domain experts outside of operations research and lived experience experts from outside of academia are crucial to accomplish this goal. We implemented an operations research modeling process where these experts inform conceptualizations, assumptions, and data.
We are building the some of the first operations research models that are being co-produced with lived experience experts and believe that our inclusive research process is capturing unintended consequences of well-meaning interventions.
Operations Research Modeling Process for Sex Trafficking: Courtesy of Sharkey et al. (2021)
Human Trafficking Data Pyramid: Courtesy of Kosmas et al. (2024a)
In order to validate network interdiction models, we need realistic network data. Our team combined case file analysis, interviews with subject matter experts, and working sessions with our survivor-centered advisory group to access the hidden data level of the human trafficking data pyramid.
The results of accessing the hidden data allowed us to create a network generator that produces realistic sex trafficking networks. The network generator introduces the idea of 'pods' of victims - who are all connected to one another but not connected to other victims of the trafficker - and captures how a "main girl" influences the number of victims in the network.
This data helps to test network interdiction models and analytics in a way that is grounded in the reality of certain types of trafficking operations.
Realistic Network Example: Courtesy of Kosmas et al. (2024a)
Measuring the Importance of Capturing Network Reactions: Courtesy of Kosmas et al. (2024a)
Our modeling approaches have helped to better understand how the trafficking network reacts to disruptions by restructuring its operations. These reactions include recruiting new victims. Our analysis shows the importance of coupling efforts to remove victims from their trafficking environments while simultaneously disrupting the ability of the traffickers to recruit new victims (Kosmas et al., 2024b).
Sex trafficking victims are often required to participate in illegal, revenue-generating activities beyond commercial sex. Our team has created and analyzed an operational model of a sex trafficking network with additional forced criminality. The idea for this inquiry arose from our survivor-centered research partners and has led to the first network model of sex trafficking co-produced with lived experience experts.
Forced Criminality Network Model: Courtesy of Clark et al. (2023)
Note: Presented in reverse chronological order.
Yow, K. (2024). A Transdisciplinary, Survivor-Centered Approach for Creating a System Dynamics Model of Closed-Buyer Sex Trafficking Networks. Master’s Thesis, Department of Industrial Engineering, Clemson University.
Kosmas, D., Sharkey, T. C., Mitchell, J. E., Maass, K. L., & Martin, L. (2024b). Multi-period max flow network interdiction with restructuring for disrupting domestic sex trafficking networks. Annals of Operations Research, 335(2), 797-860.
Kosmas, D., Melander, C., Singerhouse, E., Martin, L., Barrick, K., Maass, K.L., & Sharkey, T.C. (2024a). A transdisciplinary approach for generating synthetic but realistic domestic sex trafficking networks. IISE Transactions, 56(3), 340-354.
Clark, M., Sharkey, T.C., Ayler, T., Forliti, T., Friedman, J., Mariotti, M., Nelson, C., & Martin, L. (2023). Modeling disruptions for sex trafficking networks with multiple forced illegal activities. Forthcoming in the Journal of Human Trafficking.
Barrick, K., Maass, K.L., Sharkey, T.C., Song, Y., & Martin, L. (2023). Expanding our understanding of traffickers and their operations: A review of the literature and path forward. Forthcoming in Trauma, Violence, & Abuse.
Diaz, M. (2022). Generating Sex Trafficking Networks from Text Documents. Master’s Thesis, Department of Industrial Engineering, Clemson University.
Clark, M.T. (2022). Interdiction Models to Disrupt the Operations of Sex Trafficking and Other Forced Illicit Labor Networks. Master’s Thesis, Department of Industrial Engineering, Clemson University.
Sharkey, T.C., Barrick, K., Farrell, A., Maass, K.L., Martin, L. and Song, Y. (2021) Better together: A transdisciplinary approach to disrupt human trafficking. ISE Magazine, 51(11), 34–39.