Spatial agent-based modelling of smuggling dynamics in the Niger-Libya corridor

萨赫勒地区几个世纪以来的走私网络对该地区的政治经济造成了毁灭性的影响。这些犯罪网络在萨赫勒国家漏洞多端的边界之间活动，为武器、毒品和人员的流动提供了便利。最近欧盟支持的主要针对利比亚和尼日尔的干预措施减少了犯罪活动，从2017年从西非流入欧洲的非正规移民大幅减少就可以看出。然而，在一个走私业务已融入社会结构的地区，禁止走私的不利影响尚未减轻。此外，由于移民危机的基本条件和驱动因素在来源国仍然存在，犯罪网络不太可能看到走私需求减少。加上政府支持的拦截措施越来越多，这种需求迫使走私者使用更危险、不受监控的路线，并采取创新策略。因此，目前的封锁措施很可能只会在短期内有效。因此，迫切需要从经验上了解如何最好地处理萨赫勒地区的走私动态。本研究将重建和模拟在尼日尔-利比亚走廊运作的移民走私网络的结构、活动和行为。利用空间分析和基于代理的模型，它将调查走私网络如何应对反走私措施。它将特别检查战术和路线的变化。基于主体的模型是一类计算模型，它模拟自主主体的行为和相互作用，以重新创建人工环境和预测复杂现象的出现。考虑到数据的缺乏，一个基于主体的模型将允许测试干预措施及其对走私网络的影响，为有组织犯罪领域的学术文献以及欧洲和非洲的决策者提供一个关键的实证研究。该模型将根据实地通过对走私者和移徙者的访谈和调查以及卫星图像和其他关于移徙的二手数据来源收集的数据进行校准。

Centuries-old smuggling networks in the Sahel have had devastating effects on the political economy of the region. Operating between the porous borders of Sahelian countries, these criminal networks have facilitated the flows of weapons, drugs, and humans. Recent EU-backed interventions targeting primarily Libya and Niger have reduced criminal operations, as seen by the significant fall in irregular migrant flows from West Africa to Europe in 2017. However, the adverse effects of smuggling interdictions in a region where the smuggling business is embedded in the social fabric have not been mitigated. Moreover, as the underlying conditions and drivers for the migrant crisis remain in place in source countries, criminal networks are unlikely to see demand for smuggling decrease. Combined with increased government-backed interdiction measures, this demand pushes smugglers to use more dangerous, unmonitored routes and adopt innovative tactics. The current interdiction measures will therefore most likely only be effective in the short term. Hence, there is a crucial need to empirically understand how to best tackle smuggling dynamics in the Sahel.This research will reconstruct and model the structure, activities and behaviours of migrant smuggling networks operating in the Niger-Libya corridor. Using spatial analysis and agent-based modelling, it will investigate how smuggling networks respond to counter-smuggling measures. It will especially examine changes in tactics and routes. An agent-based model is a class of computational models that simulates the actions and interactions of autonomous agents with a view of re-creating an artificial environment and predicting the appearance of complex phenomena. Given the scarcity in data, an agent-based model will allow for the testing of interventions and their effects on smuggling networks, providing a key empirical study both for the academic literature in the field of organised crime and for European and African policy-makers. The model will be calibrated with data collected in the field through interviews and surveys with smugglers and migrants, as well as satellite imagery and other secondary sources of data on migration.