The agency on Thursday released its strategy for 2020-2025 in which it highlighted predictive analytics as a potential tool for trade enforcement.
Customs and Border Protection (CBP) wants to use more data and analytics to strengthen trade enforcement and compliance, the agency said in its 2020-2025 strategy document released Thursday.
“CBP will take an enterprise-wide approach to data analytics, using tools and capabilities that empower personnel in various roles to access and explore scenarios, potential outcomes and relationships that exist within available data and ensure program sustainment through data governance,” the document states.
The agency plans to “collect and connect quality data,” including intelligence and risk assessments, to provide predictive analytics supporting an “actionable common operating picture” that ensures that trade personnel, as well as agents and officers, have “relevant quality information” to conduct trade enforcement activities.
The document release comes after CBP Executive Assistant Commissioner for Trade Brenda Smith during a conference on April 18 said her agency is looking at potential trade uses of descriptive, predictive and prescriptive analytics.
CBP is looking at whether descriptive analytics — which tells what’s already happened — potentially could help the agency identify trade abnormalities such as low-value sales, goods shipped in odd quantities and port movements that don’t make sense from a price perspective, Smith said at the National Customs Brokers and Forwarders Association of America (NCBFAA) conference in San Antonio.
Predictive analytics can discern what’s likely to happen in the future based on past data, she said.
These analytics potentially could help the agency identify evasion of new antidumping duty orders by exposing shipments that come from countries with no previous production facilities for the finished product, Smith said.
Prescriptive analytics can examine problem sets and identify options for action, she said.
CBP is looking at leveraging the technology to expedite its rulings process, specifically to speed up the time it takes to analyze information relevant to things like tariff classification, though the agency generally meets its 30-day benchmark for issuing rulings after receiving a ruling request, Smith noted.
“While we’re not going to fully trust the machine, we do think that having analytics in place could help us make some of our workload processes faster and more efficient,” she said. “So we’re looking to invest in those.”