Whether or not you’ve been to a tech-oriented manufacturing conference lately, you’re no doubt hearing some buzz about artificial intelligence (AI), machine learning, deep learning, edge computing, and the industrial Internet of Things (IIoT). There’s no denying that the growing potential of these technologies is exciting. Not relentlessly-attacking-your-manufacturing-challenges-like-the-terminator-without-any-human-intervention exciting (at least for now), but there’s little doubt these technologies will have a transformative effect on operations moving forward.
At a fundamental level, the trends with these technologies revolve around gathering and using data, or big data, from across the IT infrastructure to make better decisions and/or drive more efficient operations. But, as with anything IT related, realizing the true value of these technologies will take a lot of careful planning and work. This blog post explores why and provides some key considerations for getting on the right track for taking advantage of AI, machine learning, and other data-driven technologies in manufacturing operations.
While some advanced manufacturers have already incorporated technologies such as the IIoT and AI into their operations, most are still wrestling with more basic challenges around how to more effectively sort through and use all of the data that is now available to them. And, as everyone who works in the field knows, that’s easier said than done. In past blogs, we’ve discussed what we call “tool traps” and why emphasizing the right knowledge, metrics, and processes is just as essential as technology. And it should come as no surprise that this holds true with technologies such as machine learning and AI. After all, there is only so much your business can do at one time, and whatever that is should be laser focused on key business goals.
In complex operations, the combination of the IIoT and machine learning has demonstrated tremendous potential. For example, one manufacturer used hundreds of sensors tied to machine learning and AI to measure a host of variables on gas turbines and automatically optimize turbine combustion levels based on weather and other factors.
One of the most exciting things about a combination of IoT and machine learning solutions is their potential to uncover unknown problems and solutions. But let’s not forget the technical challenges as well as the expense that go into gathering and crunching all of the data that could be captured. Sure, it’s gotten easier and less expensive to collect and analyze millions of data points a day, but that doesn’t mean it’s inexpensive or necessarily in the best interest of your business. After all, getting anything useful from the data may involve creating specialized algorithms and thus require expensive and hard-to-find skill sets.
That’s why it’s important to remember that just because more and more granular data is available doesn’t mean that you need it to run your business. We’ve seen plenty of companies that were struggling to monitor a very large number of KPIs quickly start doing much better once they began emphasizing the critical few that were important to near- and longer-term goals. While there may be good reason to add KPIs or metrics over time, it’s important to do it in such a way that they don’t just become noise or expensive art projects that no one has the time to try to make sense of.
One of the key takeaways from the recent Gartner Symposium/ITxpo in Orlando was the importance of flexible, unified leadership in enabling transformation into a truly digitally-driven business. In other words, the close alignment of business and technical leadership in pursuit of key business goals and objectives is critical to success. Only when both sides of the business are aligned can you ensure the right technologies are put into place to solve the challenges at hand and that you are able to map out a viable plan for continuous innovation. And one of the keys to success is continually asking, “What are we trying to accomplish, and where are we trying to go?” Then you can adjust the map of your company’s digital future accordingly.
Given the current trends in manufacturing, it is important to ponder how you could leverage technologies like machine learning, sensors, and edge computing in your business. But, unless you are able to consistently execute against critical objectives and realize performance goals, a jump to machine learning, the IIoT, or AI capabilities can wait. To get on the right path to becoming a digitally-driven business, you need a solution in place that gives you a clear window into real-time performance and past performance trends.
But don’t forget that technology is only part of the equation. A proven management approach and the process rigor to execute effectively against objectives and performance goals are also critical. In other words, once you are able to harness and use your data effectively on a day-to-day basis, you will be in a much better position to implement more advanced technologies in a way that makes sense for your operations and ultimately see greater benefits.
Ken Koenemann is VP of Supply Chain and Technology at Dploy Solutions, a TBM Consulting Group company.