2021-09-29

Why AI is a key building block of Industry 4.0

by Vawn Himmelsbach, a technology and business writer
Montreal, Quebec - September 28, 2021
Manufacturing

Manufacturing is all about uptime. A disruption in your factory line could reduce output and cause delays, while a complete shutdown could end up costing millions of dollars. That’s why predictive maintenance is a key focus in the manufacturing sector, especially in high-speed, high-frequency production lines.

Share
“Another key concern for CEOs, and one that can have dramatic impacts, is what I call cost of poor quality, which means assessing the quality control of your assembly line or your manufacturing facility,” says Clement Bourgogne, VP Strategic Programs at Scale AI, Canada’s artificial intelligence supercluster.

We’ve entered the era of Industry 4.0, which integrates the Internet of Things, data analytics, automation, machine learning and artificial intelligence to optimize manufacturing processes and increase agility so, for example, a manufacturer could quickly adjust production to changing market demands.

But deploying AI takes resources, human capital and a willingness to wait for the technology to ‘learn’ from big data to see a return on investment. A good starting point for incorporating AI into your operations is predictive maintenance, along with product quality inspection and demand planning.

“AI enables manufacturers to optimize cost in each factory through predictive maintenance and better planning,” according to Boston Consulting Group’s recent study, The Rise of the AI-Powered Company in the Postcrisis World. “It also allows them to operate a larger number of small, efficient facilities nearer to customers — rather than a few massive factories in low-wage nations — by deploying advanced manufacturing technologies such as 3D printing and autonomous robots that require few workers.”

AI ‘learns’ when equipment is about to fail, so repairs, replacement or maintenance can be taken care of before that happens — reducing downtime and the costs associated with equipment failure. But it can also help to quickly resolve quality assurance issues (and save money).

For example, in airplane manufacturing, when there’s a defect, typically a skilled operator with years of experience will determine what the problem is and which department can fix it. “That’s what AI can reproduce — it can absorb 30 years of data and determine the probability that this is an engineering problem, therefore we should feed it first to the engineering team,” says Bourgogne. “So it cuts down the time it takes to solve quality assurance and quality control.”

AI’s ability to learn from data can also be used for demand planning. Scale AI is currently working with Bombardier to develop AI models to better predict demand for spare parts for aircraft maintenance, which will reduce the risk of supply shocks — ultimately benefitting all stakeholders in the supply chain.

Another area where the technology is expected to play a major role in manufacturing is through AI-powered camera systems that automate visual quality inspection — replacing human inspection methods — to ensure higher product quality. While such a system could identify any major defects, it could also be used to identify smaller defects in real time that don’t require shutting down production but could impact overall product quality.

For example, maybe there’s too much ink in the printer and therefore the product label is too yellow, or a misalignment on the production line causes scratches on the product packaging. These types of issues can be easily fixed without disruption or downtime to ensure consistent product quality.

“Quality control in any sort of manufacturing setting is quite high and quite important. And AI is well suited for that, Predictive maintenance and quality control are hot topics in manufacturing, but there are many others — such as buffer inventory and warehouse management — that are applicable as well.

Clement Bourgogne, VP Strategic Programs at Scale AI, Canada’s artificial intelligence supercluster.

For example, CAE Inc., a Canadian manufacturer of simulation and modelling technologies for airlines, aircraft manufacturers, healthcare specialists and defense customers, is working with Scale AI to optimize its manufacturing processes, which will reduce production cycle times, minimize inventory and improve product time-to-market.

On the other hand, some manufacturers are choosing to take a wait-and-see approach. But Bourgogne cautions CEOs against this. “You could say this is like any other technology where I will let the early adopters take the risk and fail, and then I do it five or 10 years later and benefit from all their learnings — unless you consider the fact that every day the AI model is trained on new data and becomes more and more accurate,” he says.

“So you can certainly choose not to invest, It’s just that you’ll be much less competitive than those who leverage their data and digital tools over the foreseeable future. This is not something that’s going to be easy to catch up on in the next few years.”

Clement Bourgogne, VP Strategic Programs at Scale AI, Canada’s artificial intelligence supercluster.

While AI does require a long-term commitment, it will ultimately build resilience into production lines and supply chains, so manufacturers can reduce risk (and costs) while gaining new flexibility to meet customer demands. As the Boston Consulting Group study points out, strong leadership commitment is key to successful transformations and “now is the perfect time to take bold, transformative action” to prepare for the post-pandemic world.

-30-

Made possible through the
financial support of
Gouvernement du Québec
Gouvernement du Canada