Opinion: AI set for post-pandemic uptake

The manufacturing industry is no stranger to artificial intelligence (AI), digital enablers and disruptive technologies. But while smart technologies have been on the cards for some time now, many manufacturers are yet to fully commit to digitalisation. According to Gartner, only 21 per cent of companies in the industry have active AI initiatives in production.

AI and its role in keeping engineering moving during lockdown

Opinion: Engineer the new normal with AI and automation

That said, recent research by Google Cloud found that 76 per cent of senior manufacturing executives have turned to digital enablers such as data and analytics, cloud and AI in response to the pandemic. The promise of assistance with business continuity, increased employee productivity and more efficient operations are becoming too great to ignore. So, what do manufacturing executives’ priorities look like beyond the pandemic?

What’s next: Further AI adoption across the industry 

The pace of AI adoption in manufacturing has so far been slow, but it is accelerating. Google Cloud’s research found that two-thirds of UK manufacturers (66 per cent) are increasingly reliant on AI to assist in day-to-day operations, a quarter of whom (25 per cent) already allocate half or more of their overall IT spend towards AI. This is seen across all sub-sectors of the industry, though automotive or original equipment manufacturers (OEMs), automotive suppliers, and heavy machinery are using it the most. For example, Google Cloud recently partnered with Ford to deliver AI, machine learning (ML) and data analytics tools to the automaker's factories. Tools like vision AI – which derives insights from images with prediction accuracy – help Ford boost employee training and ensure plant equipment performs to an even higher standard.

Cloud computing – which is important for AI use – has high adoption rates among manufacturers too. Almost all manufacturing executives (83 per cent) already have some sort of cloud strategy, regardless of region or sub-sector. For example, industrial automation and software leader Siemens recently announced it will integrate Google Cloud’s data cloud and AI/ML technologies into its factory automation solutions to help manufacturers across the industry innovate for the future. Whereas previously AI projects have been deployed across the industry in silos, this partnership will simplify AI deployment and help manufacturing companies leverage smart transport and handling systems, supply chain optimisation and autonomous production, on a global scale.

Real world example: Uses of AI in manufacturing

One of AI’s great strengths is helping solve real-world problems, and the pandemic has thrown the manufacturing industry a whole host of challenges over the past year or so. As a result, we’re seeing more and more AI use cases across the industry as manufacturers navigate the current climate and prepare for the future.

One area with big potential to benefit from AI is quality control, which was called out by 39 per cent of surveyed manufacturers in our research. AI can be used for visual inspection of finished products and help manufacturers achieve higher quality checks, both at speed and at a lower cost. Using AI vision means production line workers can spend less time on repetitive product inspections and can instead focus on more complex tasks, such as root cause analysis.

AI can also be applied to supply chain optimisation, powering connected factories or assisting with predictive maintenance. In the area of supply chain optimisation, Google Cloud found that manufacturers tap into AI and ML for supply chain management (36 per cent), risk management (36 per cent), and inventory management (34 per cent).

AI can interpret huge datasets of internal and external data, leverage insights and predict potential disruptions (such as weather, traffic, transportation cost, and competitive and raw material pricing). By removing the guesswork and improving foresight and agility, operational teams can make informed decisions. When it comes to inventory management, custom ML models can predict when machines might cause unscheduled downtime and how this might impact production schedules.

A look ahead: What the future holds for AI in manufacturing 

The key to AI adoption in manufacturing rests in its ease of deployment and use. Although skills, shortages and cost are generally cited as barriers to AI adoption, Google Cloud found these challenges not to be particularly significant. According to its report, just a quarter (23 per cent) of manufacturers believe they don’t have the talent to properly leverage AI, while even less (21 per cent) recorded cost as a roadblock. Where manufacturers seem to be drawing a blank is having the right technology platform and tools in place to manage an AI strategy equipped for the industry. As AI adoption continues to grow, manufacturing executives will need to ensure they’re focussing their efforts in the right places.

AI is already shaking up manufacturing, but as we move beyond the pandemic we’ll see its uptake continue to grow. Now, more than ever, we will see AI — and other digital enablers and disruptive technologies — offer the industry the tools it needs to accelerate.

John Abel, technical director, Office of the CTO at Google Cloud