Statistical profiling to predict the biosecurity risk presented by non-compliant international passengers

TL;DR

This study evaluates statistical profiling methods to predict non-compliant passengers carrying biosecurity risk material, demonstrating improved screening efficiency over manual and random methods, and suggesting further investigation for border security enhancement.

Contribution

It introduces a case study applying statistical profiling to predict passenger non-compliance, showing its effectiveness over traditional manual screening methods.

Findings

Statistical profiling outperforms manual screening in predicting non-compliance.

Using risk predictions improves the efficiency of screening higher risk passengers.

The method shows promise for enhancing biosecurity interventions at borders.

Abstract

Biosecurity risk material (BRM) presents a clear and significant threat to national and international environmental and economic assets. Intercepting BRM carried by non-compliant international passengers is a key priority of border biosecurity services. Global travel rates are constantly increasing, which complicates this important responsibility, and necessitates judicious intervention. Selection of passengers for intervention is generally performed manually, and the quality of the selection depends on the experience and judgement of the officer making the selection. In this article we report on a case study to assess the predictive ability of statistical profiling methods that predict non-compliance with biosecurity regulations using data obtained from regulatory documents as inputs. We then evaluate the performance arising from using risk predictions to select higher risk passengers for screening. We find that both prediction performance and screening higher risk passengers from regulatory documents are superior to manual and random screening, and recommend that authorities further investigate statistical profiling for efficient intervention of biosecurity risk material on incoming passengers.