Industrial AI gives people 'superpowers' in advanced manufacturing. Here's how

In the advanced manufacturing sector, innovative leaders adopting industrial artificial intelligence (AI) empower their human workers with “superhuman” capabilities. Human labour forces equipped with self-learning and self-adapting technology are achieving unprecedented production yield and quality levels while simultaneously accelerating toward sustainability and safety goals.

AI has the potential to impact advanced manufacturing significantly. The report, Unlocking Value from Artificial Intelligence in Manufacturing, finds that AI could boost gross domestic product by 2% per year.

While many fear that AI offers a zero-sum tradeoff to traditional labour force alternatives, AI is actually forecasted to create millions of new jobs. The World Economic Forum’s Future of Jobs Report 2023 states that new roles like AI and machine learning specialists, data analysts and scientists, and digital transformation specialists are among the fastest-growing roles.

While job growth is a net positive, labour shortages and ageing populations will continue to constrain the world’s workforce over the next decade. According to the eighth annual State of Smart Manufacturing Report, 46% of manufacturers lack skilled workers to outpace the competition.

Contrary to concerns about job replacement, industrial AI is not displacing humans; it’s augmenting our abilities by simplifying problem solving. In traditional manufacturing, countless hours are spent optimizing parameters to find the best recipe for a given product at scale. During production, operators are often glued to machine stations, constantly manipulating parameters to keep the process under control and reacting to deviations in product quality.

With AI, recipes can be automatically refined in a virtual environment before production, and AI-native control algorithms can handle routine manipulations in the equipment. This capability frees the operator to “manage out” across a broader array of processes and focus more on strategic problem-solving rather than tactical manipulation.

This fusion will continue to shape the future of manufacturing in ways we’re only beginning to imagine. Essential to success, however, is identifying the most impactful use cases and experiences people have with the technology. There are numerous examples of AI’s potential to empower people with elevated levels of productivity and sustainability – from training to operations management.

Training effectiveness receives a power boost when using augmentation technologies. According to the New Narrative, augmentation technologies can offer gains in training effectiveness by up to 80% compared to in-person training. And these technologies can be accessed in real-time. Much like an online search for how-to videos or instructions that can support a minor repair at home, operators can leverage AI to support equipment diagnostics and repairs with on-demand access to user manuals, maintenance procedures and videos. Knowledge transfer becomes instantaneous.

Fundamentally, AI brings next-level computational powers to simplify complex, multi-variable optimization problems. Typically, manufacturers harness this power with two predominant strategies:

These advanced problem-solving techniques apply at the operations management layer as they do on the plant floor. Operations management requires complex tradeoffs between many economic objectives. For example, one significant challenge that advanced manufacturing faces is achieving net zero within operations. If seeking to reduce the energy of an energy-intensive machine while also maintaining necessary throughput, there starts to become more than one objective function to optimize simultaneously.

Now consider when programmable logic is insufficient to tackle the most complex automation problems. Tesla recently introduced the Full Self Driving Version 12 architecture, eliminating over 300,000 lines of programmable logic in exchange for a neural network that directly mapped sensory input to control outputs. Tesla realized the number of decisions in a real-world driving environment is so numerous that they couldn’t anticipate various outcomes in a traditional programming paradigm. Tesla’s novel control architecture proved revolutionary in its performance and accelerated its vision of autonomy.

Industrial companies are experimenting with similar architectures for the future of automation – for problems that are too complex for code, you can derive a control policy out of historical data by algorithmically discovering which control actions have yielded positive or negative outcomes under various circumstances. AI can enable self-driving production systems just as it has enabled self-driving vehicles.

Industrial AI is helping the world’s largest manufacturers evolve from traditional automated systems to autonomous systems capable of learning and adapting to their environment.

Here’s what we can infer about the long-term impact of autonomy on the global advanced manufacturing economy:

The fusion of human ingenuity and AI is shaping the future of manufacturing. And most importantly, manufacturing is being shaped by and for people. Leveraging the best of technology and building superior experiences will empower people to do extraordinary things. By embracing AI as an amplifier, we can unlock the full potential of human creativity and drive an era of unprecedented progress.