AI used to reduce time resolving air traveller complaints

- Ottawa, Ontario

Members of the Data Analytics Centre team discuss the new project with the Canadian Transportation Agency.

Members of the Data Analytics Centre team discuss the new project with the Canadian Transportation Agency.

The NRC’s Data Analytics Centre developed digital solutions to help resolve complaints backlog.

Did you fly on a Canadian airline recently and have a complaint?

The Canadian Transportation Agency (CTA) receives complaints from Canadians about airlines. In recent years, the volume of incoming complaints has skyrocketed, leading to a backlog. Data experts at the NRC have developed a solution that uses artificial intelligence (AI) to process complaints faster—helping Canadian travellers get their complaints resolved more quickly.

The CTA has been able to reduce the time it takes to review complaints from Canadian air travellers thanks to an AI solution developed for them by the NRC’S Data Analytics Centre (DAC), part of the Digital Technologies Research Centre.

The team of data experts at the DAC are uniquely positioned to work with private‑ and public‑sector organizations to help them extract the most strategic value from their data.

Automating a manual process

Resolving a complaint is a multi-step process. “Part of the problem,” says Rachael Donovan, Manager of Compliance Strategy and Analysis at the Canadian Transportation Agency, “is that the first part of that process was largely manual work.”

Before a complaint can be placed in the hands of a CTA staff member for initial review, 2 things are needed. There must be proof of purchase, meaning a ticket number or itinerary number, and there must be proof of correspondence—evidence that the passenger has attempted to resolve the complaint with the airline. The first step in the process is to confirm this documentation was included among the numerous documents that can make up a complaint. Done manually, this check can be difficult and often require review of the original file multiple times.

The CTA approached the DAC to develop a way to automate this first step. The solution involves using AI to comb through those documents, determine exactly where the proof is and catalogue the information so it can be easily found.

“It’s the difference between flipping through an entire book and having someone tell you that what you're looking for is on page 20,” says the NRC project lead, Tyson Mitchell.

AI speeds up decision making

The solution Mitchell developed speeds up that first step in the complaint resolution process. It has also sped up the entire investigation because officers know the information is complete and don’t have to waste precious time looking for the proof of purchase and correspondence or going back to the complainant for more information. The bonus? It took only about 8 weeks to go from idea to full implementation.

The CTA has been piloting the new system, with impressive results. “The time to complete the review of incoming complaint documentation has been reduced by half, from 10 to 30 minutes when done manually to 5 to 15 minutes, and with greater than 90% accuracy,” says Michelle Greenshields, Director General of the Dispute Resolution Branch at the Canadian Transportation Agency.

The knowledge and positive experience gained through this project has helped the CTA implement improvements and digitize the complaint intake form on its website. This is one of what Donovan calls “related efficiencies.” The CTA expects to find more opportunities for efficiencies by leveraging technology and data as it moves toward a simpler, clearer, faster and more cost-effective process that will ensure timely complaint resolution and access to justice for Canadians.

Success built on strong relationships

The Canadian Transportation Agency has been working with the Data Analytics Centre since 2018. The CTA first sought the DAC ’s assistance to assist help it become a more data-driven organization. “The length of the relationship is part of the reason this project has been so successful,” says Stéphane Tremblay, Team Leader at the DAC . He describes his team’s mission as “transferring knowledge to our brother and sister departments in government.”

This long-standing relationship is critical to the success of the Data Analytics Centre. “The longer we do business with an organization, the deeper we understand their needs,” says Tremblay.

Donovan adds, “This did not happen by accident.” When the CTA decided to make modernizing its processes a priority and to increase its data analysis capacity, other government departments steered her to the Centre.

“Stéphane and his team have been really supportive. They worked with us to create tools and processes to make our work as efficient as possible,” she says. “The relationship we have with the Data Analytics Centre means that our senior executives are fully supportive of our work.”

According to Greenshields, the project was an operational and cultural success. It also serves as an example of a successful introduction of technology and a collaborative effort. The project demonstrated to all Canadian Transportation Agency staff what the Centre strives for with every client—the promise of technology to improve work.

Contact us

Media Relations, National Research Council of Canada
1-855-282-1637 (toll-free in Canada only)
1-613-991-1431 (elsewhere in North America)
001-613-991-1431 (international)
Follow us on X: @NRC_CNRC