Every year, hundreds of billions of packaged merchandises are imported to Canada. Among these, certain types of packaging material, especially the WPM, could pose significant risks to Canada's environmental health, as they could act as a pathway for foreign plant pests, diseases or invasive species. To mitigate these risks, the CFIA has developed regulations, and the CBSA conducts related oversight activities to verify the compliance of packages imported into the country.
To improve the efficiency and effectiveness of the existing oversight activities, we aimed to develop a decision-support tool based on machine learning algorithms using historical inspection data (wood packages shipped to Canada from January 1, 2009, to March 31, 2018) that can be used after shipment as a pre-arrival assessment tool to predict which wood packages are at a given risk category (elevated or minimal risk) with regards to compliance standards within their arrival at a port of entry in Canada.
