May 17, 2022
Box plants are constantly looking for more efficient ways to maintain and increase output as the corrugated industry continues its trajectory of rapid growth and increased demand due to the rise of e-commerce. Increasing uptime through longer operational hours and other traditional methods can help in the short term, but investing in a smart long-term solution is critical for improving productivity and profitability.
Since the invention of corrugated box manufacturing machines nearly 150 years ago, optimizing machine performance has been heavily reliant on the experience, guesswork, and intuition of its operators. Now, predictive analytics via machine learning can provide predictions of when—and how—corrugated equipment may fail. The promise of advance notice of downtime is empowering plant leadership to make more informed decisions to improve the reliability and efficiency of their fleet.
While implementing an industrial internet of things (IIoT) solution, such as machine learning, is not a new idea in the greater manufacturing industry, these technologies are now gaining market adaption and adoption in the corrugated converting industry. Early adopters and leaders in the corrugated industry are already running pilot programs to see the true capacity, resiliency, and potential return of incorporating machine learning into their fleet. These experiments are allowing box plants to begin the process of optimizing their operation as a whole, feeding machine-learning algorithms with the necessary data and feedback one machine at a time.
Machine-learning technology is dynamic, evolving to provide flexibility for how box plant managers can incorporate IIoT into their production line without overhauling or drastically halting their operation. Machine-agnostic tools can collect vital information from corrugated converting equipment through sensors that don’t disrupt an operation and don’t void warranties on older machinery. Data is then collected, analyzed, and presented through a digital dashboard, allowing anybody to view actionable insights, optimize maintenance, and reduce downtime—no matter what brand of machine equipment is in use at their plants.
The latest advancements in machine learning include advanced insights and optimization, providing analytics that predict when and how corrugated equipment will fail. Whether the machine can be fixed before the failure or production is rerouted to another machine, predictive analytics allows plant management to make better, more informed decisions to improve the overall uptime and efficiency of their fleet.
Machine learning solutions can also provide additional value by helping box plants take advantage of planned downtime. The benefits of planned downtime open up opportunities for plant managers to order parts in advance, schedule maintenance resources, and shift production to alternate machines. Being prepared for these instances, rather than racing for resources, parts, and labor during unplanned downtime, is yet another avenue that machine-learning tools provide to limit lost time and improve efficiency in an operation.
The beauty of machine learning, especially in such a rapidly growing industry, is that wider adoption and collection of data can yield more accurate predictions. Through the progressive nature of machine-learning algorithms, box plants will be able to plan around when and how their converting equipment will fail. Understanding and proactively solving them—thanks to machine learning—are revolutionizing how box plants understand, predict, and maintain their converting equipment.
Matthew Miller is technology director at Helios, with expertise in bringing products, processes, and technology to life. An expert at aligning technology solutions with customer needs, he has successfully led teams to build products that have led to significant commercial growth and increased business productivity.
