Artificial Intelligence (AI) in manufacturing shows the greatest positive impact when compared to other industries. With documented success, why, then, are so many manufacturers slow to get on board the AI train? And are those who have committed to AI reaping the full extent of the benefits it can offer?
"According to a 2021 survey, 56 percent of all respondents report AI adoption in at least one function."
(Source: Mckinsey 2021)
AI isn’t a new concept for the manufacturing industry, and its increase in popularity has been steadily growing over the last few years. The pandemic caused a big shift in how we work, and this can partly explain why AI is becoming important to businesses.
what is new is its increased accessibility to smaller companies. This shift may partly explain the reason for the high research but low implementation statistics.
So, what are the proven successes of AI in the manufacturing industry? What is holding the rest of the industry back from making the leap to AI implementation?
Documented areas of success.
Efficiency/Predictive maintenance/workforce productivity.
McKinsey has reported that AI used to monitor and analyze factory machinery reduces machine downtime by up to 50%. This is achieved through the analysis of multiple data points and historical data to forecast possible service requirements to enable maintenance before any machinery breakdowns. By doing this, not only will the downtime per machine be slashed, but the machine’s life span can be extended by up to 40%. McKinsey estimates the global financial savings from predictive maintenance at $0.5-$0.7 trillion.
Industry magazine The Manufacturer recently reported that 92% of senior manufacturing executives see Artificial Intelligence as an essential tool to increase their productivity. According to ciol.com, the combined cost saving of predictive maintenance and the boost to productivity due to the decrease in downtime will result in average profit increases of 39% by 2035.
Logistics network/warehouse optimization/inventory and parts optimization.
The constant assessment of stock and stock location across sites depending on demand is an element of the manufacturing process that has the power to decimate profit margins. Too much stock in the wrong location means money tied up in physical assets and increased logistics costs.
By using AI to predict manufacturing requirements, companies can be agile and respond to market demands. Forbes.com reported that automating inventory optimization improves service levels by 16% while simultaneously increasing inventory turns by 25%.
“Product will become linked to emerging demand, so we’ll never be in a position where things are just ‘out of stock’.”
Greg Kinsey, VP, Hitachi Insight Group
(Source: The Telegraph)
Furthermore, McKinsey stated that in supply-chain management, forecasting demand improves accuracy by up to 20%, with a knock-on effect of a 5% reduction in inventory costs and revenue increases of 3%.
An example of inventory optimization from AI provider ToolsGroup shows how a major provider of climate control solutions improved service levels by 16% while simultaneously increasing inventory turnover by 25% thanks to the integration of AI in their systems. (Source: Microsoft Industry Blog)
The incredible amount of data created by modern manufacturing equipment is beyond our human capability to decipher and draw conclusions from in a timeframe that would prove useful to daily operations. AI can deliver the patterns, warnings, and insights as a constant flow, enabling in-house specialists to decide which areas to refine. As a result, manufacturers using these methods see increases in yield whilst being able to minimize energy consumption.
As with inventory optimization, reducing supply-chain costs by responding to product demand can have significant offer results. According to Louis Columbus, Principal at manufacturing execution software provider IQMS, AI reduces forecasting errors by 50% and lost sales by 65%.
Artificial Intelligence and machine learning can also play a vital role in a company’s sales and marketing departments. Building on the data insights provided by business intelligence software such as Salesforce and sales-i, AI is transforming the way teams work.
High-performing teams are 4.9X more likely to be using AI than underperforming ones.
AI delivers insights to guide sales and marketing teams to focus their efforts for maximum impact, streamlining human processes as well as on the shop floor.
Roadblocks to AI implementation.
With such a large weight of evidence on the benefits of AI in the manufacturing industry, what’s holding companies back?
As with all innovations, those that forge the way tend to be companies with the time, skills, and resources to commit to new methods. Just like when a new product hits the shelves in the B2C environment, early adopters will pay a premium price. Only when mass adoption lowers costs will smaller companies be able to use AI.
Whilst global brands may be upscaling their AI efforts and pushing the boundaries of current technology, many manufacturers are only just starting to implement the data capture technology that makes AI possible. Without a large volume of data points, AI would be like fitting an F1 engine in an old Skoda Riva.
Another stumbling block is the speed of advancements. Technology is moving at such a pace that it outpaces the education of specialists to implement, run, and maintain it. Again, this adds to the cost as employees can demand a premium for their skills.
Manufacturing is often seen as a traditional industry. As such, people who are currently employed doing more manual tasks and physical labor may be worried about the impact on their job stability due to the introduction of artificial intelligence.
“You need to embrace this technology; if you don’t, because you fear that you might lose some jobs, you are going to lose all the jobs, as your company will no longer be competitive. In fact, digital technologies can improve the workplace and quality of work.”
Greg Kinsey, VP, Hitachi Insight Group
(Source: The Telegraph)
A lack of understanding of the system coupled with fear of obsolescence can unnerve a traditional workforce. Gaining the buy-in of your current workforce is pivotal to the success of the overall implementation. This may take some time and education on the ability of AI to amplify human capabilities.
The march of progress is undeniably in the direction of AI for pretty much every industry. The challenge is now for those without global research and development budgets to try to catch up. How? They can use AI to carve out a niche where they can excel.