Previous Article: Mirage: Prioritizing the right digital transformation project, not so simple!
Next Article: Artificial Intelligence in Our Homes, but Not in Our Businesses...

May 25, 2021

Increase the efficiency of your quality control with artificial intelligence

by Annie O'Farrell Marketing expert

Artificial intelligence

Hugo D’Astous is the general director at Patates Dolbec. Leader in Eastern Canada, the company produces, packages, and distributes over 100 million pounds of potatoes each year. They recently implemented artificial intelligence into their production line, and the results have blown them away. To manufacturers who claim they can’t improve their equipment, here’s a prime example of the opposite.

According to Mr. d’Astous, the initial idea was not to start a 4.0 project to simply follow the trend but rather to “automate, simplify processes, help operators, and computerize everything that could be computerized so that people could perform value-added tasks.” 

So one of the digital transformation projects was to improve the efficiency of the optical sorting machine. Although the company had scoured the globe to find the best technology that would meet its needs and sort potatoes efficiently, the machine still had a 30% error rate. This has forced employees to manually sort a portion of the potatoes, resulting in the loss of a significant amount of healthy potatoes and associated revenue. In an industry with a persistent labor shortage, it was crucial to find a more efficient solution, hence the idea of improving the machine’s performance with AI.

Take one step at a time

According to Mr. d’Astous, it is important to approach this type of project phase by phase. You must plan the whole project well and then focus on one step at a time. In the case of the production line automation project, he asked himself, “Do I have the confidence, and am I willing to gamble an initial $200,000? If so, then let’s go ahead and plan it properly.” Thus, phase after phase, the team set new objectives, and the project progressed to finally achieve results that exceeded expectations. On the other hand, he reminds us that risks are always associated with innovation projects and that we must be ready to assume them when we decide to embark on such a project.

Currently, many manufacturers in Quebec are falling behind technologically as SMEs have difficulty taking the leap and investing in their innovation projects.

take full advantage of the subsidy programs

Therefore, Mr. d’Astous suggests that companies take full advantage of the subsidy programs available to them. “If AI is available in the potato industry, the rest of Quebec has to wake up!” Patates Dolbec admits that these subsidies allowed it to move forward with the AI project, without which the size of the company would not have allowed it to implement such technology. Over the years, the organization has used several financial levers, including the INVEST-AI program for its sorting process automation project.

To conclude, Mr. d’Astous recommends that companies embarking on a digital transformation project take the time to fully understand and audit the potential of the AI team they wish to work with. “It’s important to choose the right partner because artificial intelligence happens through the expertise of people.” Mr. d’Astous admits that his past experiences with technology suppliers have led him to see that “many are there to bill only, and eventually everything ends up costing a fortune.” This inevitably led to a fear of not achieving the expected results within the established budget, hence the importance of paying attention to choosing the right partner. “A project has a beginning, and the end is not always known with precision. So you have to make sure that in the difficult moments, you are with people with whom you are going to be hand in hand in the project.”

In the end, the implementation of AI in the quality control process will have allowed Patates Dolbec to gain no less than 25% in efficiency as the error rate went from 30% to 5% in just a few months. The power of technology in 2021 is simply under estimated by the vast majority of manufacturers. Let’s stop making humans do what machines and technology are much better at.