Home » Technology » AI and Manufacturing 4.0 — The Light Blue Revolution
Big Data and AI

AI and Manufacturing 4.0 — The Light Blue Revolution

A robotic arm at work
A robotic arm at work
Ulrike Bahr-Gedalia, Canadian Chamber of Commerce_

Ulrike Bahr-Gedalia

Senior Director, Digital Economy, Technology & Innovation, Canadian Chamber of Commerce

Mehdi Merai, Dataperformers

Mehdi Merai

CEO, Dataperformers

As the manufacturing sector is increasingly moving from blue collar to white collar and embracing the light blue realm, the wealth and magnitude of where the next generation of technology, transformation, people, and opportunities could take us is exponential.

In an interview with Ulrike Bahr-Gedalia, Senior Director, Digital Economy, Technology & Innovation at the Canadian Chamber of Commerce, Mehdi Merai, a member of the Chamber of Commerce and CEO of Dataperformers, explains what artificial intelligence (AI) contributes to manufacturing, what some of these opportunities might be, and how the future could look.


We keep hearing about Industry 4.0. What is it?

The Fourth Industrial Revolution, or Industry 4.0, is all these new technologies like the Internet of Things (IoT), AI, and 3D printing, that when combined can create new ways to do manufacturing, also known as smart manufacturing.

What’s big data and how can it influence smart manufacturing?

Every day, we generate data when performing tasks like searching the web, checking our Facebook, or using a GPS. This phenomenon of generating, in an exponential way, more and more data is what constitutes big data. Likewise, manufacturers are generating lots of data through the use of their equipment (IoT), quality control processes (images), or through their maintenance logs.

What are some smart manufacturing use cases? Increased technology adoption across Canada in all sectors and sizes of business is crucial, and we’d definitely like to see more of it!

We have to understand that AI has applications across the entire manufacturing value chain. Two of the most mature applications are on preventive maintenance and quality checks. Siemens is currently using sensors on old industrial motors to detect subtle changes in vibration or energy use to predict near failure — and so it triggers maintenance before such failures can occur.

At Dataperformers, we worked with Toyota Aisin Seiki to help them detect welding anomalies in real time. Molding defects are hard to catch as the discrepancies can be really fine and image quality and luminosity are often poor. Nevertheless, using techniques like deep learning, we developed an AI model that performed with high accuracy. Following that success, we developed a solution, Macula AI, which was a platform for any kind of defect detection using images or videos.

As often is the case with adoption and transformation, change doesn’t come without challenges. What are some issues that could arise?

This revolution presents challenges for different stakeholders:

Policymakers: The skills and competencies needed today in the manufacturing industry aren’t the ones that will be needed in the future. Jobs will disappear, new ones will be created, and others will evolve. To keep the manufacturing sector competitive in Canada, we need to attract and prepare youth today.

Society: Those technologies will bring new situations that will need new legislation, like the new law on data privacy in preparation. In manufacturing this will be the same, with new standards and regulations.

Manufacturers: This technology enables new opportunities. Manufacturer business models will have to evolve to stay competitive by taking the best of what those technologies enable and become more agile.

There always seems to be a storyline. What does the AI journey in manufacturing look like?

The prerequisite is to create a general database, with no specific purpose, that collects all the different data generated. We call it a “data lake.” Manufacturer AI journeys start by identifying problems easily solved by AI and that would have a high impact for the company. Because of the lack of AI specialists and the cost of their expertise, the manufacturer hires an AI company to undertake its pilot project. This pilot is generally the demonstration that convinces the chief experience officer to start a company transition to AI. Then the manufacturer generalizes the use of AI by rethinking their process, hiring an AI team, and helping employees adapt. While this is the final step for the initial journey, it means a future full of potential for the company at large.

Next article