Since 1999, Bill Rokos has spearheaded the development of Parsec’s manufacturing operations management (MOM) platform, TrakSYS.
There has long been a delicate dance of risk versus reward when investing in the future of manufacturing, and historically, organizations have leaned toward being risk-averse. When it comes to artificial intelligence (AI)—a sect of technology that stands to optimize operations and markedly offset operating costs—the proverbial sheen that sought to spark action in the hearts of cautious companies has begun to wear off, as the reality of improvement timelines sink in.
This tipping point manufacturers are heading toward is commonly referred to as the Trough of Disillusionment, coined by Gartner to best explain the lackluster sentiment that follows an investment that does not have immediate positive output. Simply put, manufacturers are getting impatient for tangible results, and some of the excitement that once existed about AI’s disruptive potential is now veiled by performance issues, slow adoption rates and pushed project timelines.
So, what’s next in this cycle? How do we adequately make our return on investment while continuing to be a champion for innovation? As technologists, we need to take the lessons we’ve learned from other manufacturing revolutions and apply the same principles. We must take a critical eye to risk versus reward, fact versus fantasy and continue to advocate for the right AI to significantly improve our lives as manufacturers.
Is AI Losing Its Steam?
When AI became readily available to manufacturers, many technologists raced to find the best solution and implement it fast. Last year, a stunning 4% of companies surveyed by the World Economic Forum said they had no plans to implement AI in their operations, while 57% were beginning to experiment across small projects and another 26% were already defining a road map. But, in this flurry of excitement and activity, some technologists across industries may soon begin sensing a trough. What isn’t adding up from initial planning to execution for AI-driven projects?
Let’s break down a few reasons some manufacturers may believe AI is on a negative trajectory with the facts to support that the industry is still on the right path to success.
Cost
Although AI can cut costs by making manufacturing operations more efficient, businesses may have underestimated the cost of training models, hiring new technical experts to kick-start AI-driven projects, upskilling and professional development operations, system overhauls and more. Until AI is democratized across manufacturing operations, its implementation could come with additional or unexpected costs. Finding an AI-optimized manufacturing platform that is readymade for modern operations is paramount for manufacturers who do not want to be disappointed by unexpected costs.
Limitations
In its current stage, AI has reached a plateau because technologists need more data to properly train AI language models. But this doesn’t mean manufacturers will run short of use cases that can optimize their business. Efficiency begets efficiency, so the more manufacturers use AI in their facilities, the more it will learn how their operations work. When manufacturing AI has the proper amount of data, it will offer tailored recommendations on how to improve production planning, predictive maintenance and much more—which, in turn, makes operations smarter.
Sometimes AI will have an immediate, tangible effect on a facility. Other times, its change will be more incremental, perhaps hardly discernible in the short term. But incremental improvements are sustainable improvements, and they compound over time. AI is, as a technology, not done “cooking” yet! Engineers are improving it every day, expanding its capabilities, and furthering its potency. We haven’t come close to harnessing its true potential; this will take time.
Expected Outcomes
AI has incredible benefits for all aspects of manufacturing, from supply chain management and predictive maintenance to improved quality control and better decision-making. AI allows manufacturers to optimize their production by empowering workforces with contextualized, actionable data, optimizing workflows, and improving product quality. However, the true disconnect between AI’s capabilities and its outcomes ultimately lies in the expectations set by technologists. It’s our responsibility to the industry to proceed responsibly and avoid over-promising.
AI’s Next Step
Any major change is going to have its highs and lows. AI’s reputation has been damaged by misinformation on its hallucinations, biased systems and the lack of data that exists to properly train models, to name a few examples. Meanwhile, new breakthroughs are happening every day. Cross-industry adoption augments the availability of trainable data. More capital is being invested in AI’s research and development. Setting the right expectations for your organization and working closely with technical experts will help bridge the gap between promises and expected outcomes.
Finding true thought partners in an organization’s AI journey is paramount to keep the spirit of innovation alive. While we may be 15 to 20 years away from total AI bliss, manufacturers can benefit today from AI-optimized technology that continuously decreases operational costs and increases profits.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
