The aim of this work is to present possible applications of the systems of tobacco traceability for guiding local enforcement against illicit trade.
The proposed three-step strategy relies on a robust regression technique and Local Moran’s I, a local indicator of spatial association, and aims at identifying retail outlets with significantly low sales compared to normal market conditions, which can indicate illegal sales activities. The ability of the method to produce alerts pointing to areas subject to illicit trade is tested on synthetic data in terms of precision and accuracy in different scenarios. Other metrics are also provided.
Our approach performs well under different metrics and across various levels of illicit trade prevalence, achieving a precision of 94% under the main scenario and method parametrisation.
The proposed strategy provides high-quality leads for investigations into geographical areas disproportionately susceptible to illicit trade, potentially unveiling any form of illegal sales, including those involving products that have never entered the legal supply chain. Therefore, it can be a valuable tool for law enforcement agencies to tackle illegal sales activities. The findings of this study support also the argument in favour of expanding tobacco traceability systems downstream to the full length of the supply chain.