Language selection

Search


How automation is used to account for imported commercial goods in Canada

Find out how the Canada Border Services Agency (CBSA) Assessment and Revenue Management (CARM) system is using rules-based automation, or algorithms, to assess duties and taxes and to identify issues that may arise when commercial goods are imported into Canada.

On this page

Updating how we do business

The CARM system was introduced to replace outdated systems used to collect duties and taxes on commercial goods imported into Canada, such as the Customs Commercial System.

Using algorithms, the CARM system is able to automate processes to more accurately:

More specifically, the automation in the CARM system helps with:

Officers still make the decisions

The CARM system is considered only partially automated because CBSA officers continue to review transactions and remain responsible for final decision-making.

The system helps officers review and confirm which transactions are compliant after the commercial goods have been imported into Canada. Officers can:

If a business doesn’t agree with a decision, it can submit an appeal using the recourse mechanisms available to them.

Collecting personal information

The CBSA only discloses or collects commercial and personal information in accordance with the Privacy Act and the Customs Act and its associated regulations.

CARM captures personal information only to the extent that is required to register a profile and business account in the CARM Client Portal (CCP). The information is used to establish that the business meets specific requirements.

This information is stored in the CARM system and may be used for:

Results of the impact assessment

The CBSA has assessed the public impacts of deploying this automated system as per Treasury Board of Canada Secretariat's Directive on Automated Decision-Making. To read the full document go to: Algorithmic Impact Assessment - CBSA Assessment and Revenue Management (CARM) - Open Government Portal

Page details

Date modified: