Michael Lefenfeld

‍

Wood panel manufacturing has more data than ever, but many mills still struggle to turn that data into real-time decisions. The next phase of the industry will be defined by how effectively mills connect data, process, and materials to operate with greater control, precision, and adaptability.

In wood panel manufacturing, economic value is won or lost by the way the mill runs each day. Resin usage, wood utilization, line speed, energy consumption, and product quality all move together. When one shifts, the others follow. The challenge all mills face is the fact that many of the relationships driving their performance outcomes are not fully visible.

That challenge is even more difficult when you consider the primary raw material: wood. Wood is not a uniform input. It changes by source, season, moisture content, fiber density, and chemical composition. Those differences directly affect resin performance, press behavior, throughput, and final board quality. Most mills compensate for that variability through experience, conservative operating practices, and built-in process buffers. That approach has worked for decades, but it also creates hidden cost throughout the system.

Across the industry, mills collect enormous amounts of data. Fiber characteristics, press conditions, moisture levels, and quality measurements are tracked throughout production. But in many plant, that information is fragmented across systems and only reviewed after the fact, rather than being used to guide decisions in real time. The result is a manufacturing model that is data rich, but analytically poor.

The opportunity is not simply to collect more information, but to operationalize it. As mills gain a better understanding of how raw materials, process conditions, and product outcomes interact, they can begin to operate with far greater precision and responsiveness. Increasingly, AI is becoming the enabling layer that makes this possible by helping mills recognize patterns, manage variability, and translate complex process interactions into real-time operational guidance.

The Stability Tax

You see the impact of this most clearly in the press.

Operators understand the inputs. They know the wood characteristics, resin application rates, moisture levels, and press conditions. They also see the finished board. What

remains less visible to them is the manner in which those variables interact dynamically throughout the process itself.

When that visibility is limited, the system is naturally managed with caution. Extra resin is added to ensure bonding across changing conditions. Line speeds are reduced to lower the risk of defects. Buffers are built into the process to maintain consistency. Each decision makes sense individually, but together they create a system that favors stability over performance.

Over time, those choices become structural. Material usage increases beyond what is necessary. Throughput remains below what the system could support. Energy consumption rises as a byproduct of operating conservatively. This is what I call the “Stability Tax.” It is the hidden cost of managing variability without full visibility into the process itself.

Most mills accept this tradeoff as the price of running safely. But as systems become more connected and responsive, an entirely new level of operational control and performance begins to emerge.

Driving by the Rearview Mirror

The second challenge is timing.

In many mills, critical quality measurements such as internal bond strength or panel density are only available after production has moved through the system. By the time a deviation is identified, additional boards have been produced, stacked, and sometimes shipped.

At that point, the decision window has closed. Adjustments can be made for the next run, but not for the one that created the issue. This creates an operating model that is inherently reactive. Data explains what happened rather than helping influence what is happening. Operators are left to rely on experience, buffers, and after-the-fact adjustments to maintain control.

That reliance on operator experience has always been one of the industry’s strengths. The best operators can read the line, recognize subtle changes in fiber behavior, and make adjustments that keep production stable. But that expertise is also unevenly distributed. Not every shift has the same level of experience, and not every operator encounters the same production conditions. When performance depends heavily on who is running the line, variability becomes built into the system itself.

The opportunity is not to replace operator judgment, but to strengthen and scale it. AI and connected process intelligence can help operators identify patterns, respond faster to changing conditions, and make more informed decisions while production is still underway. Feedback loops that currently take hours need to shrink to minutes and eventually seconds. When that happens, data becomes part of how the mill operates in real time rather than simply a record of past performance.

The result is a more consistent process where performance depends less on individual variability and more on system-level control.

Beyond the Molecule

As mills become more capable of understanding and managing variability in real time, the role of raw materials begins to change as well.

Most resin systems today are designed to perform across a broad range of operating conditions. That approach provides reliability, but it also requires building in margin for uncertainty. As mills gain better visibility into their own processes, a more adaptive model begins to emerge.

Instead of relying on static resins formulations designed to cover every possible scenario, resin systems can increasingly be aligned to the specific realities of an individual mill. Wood species, moisture profiles, press configurations, production targets, and operating conditions can all become part of how material performance is designed and optimized.

This is where AI, process intelligence, and material science begin to converge. Mills are no longer optimizing isolated parts of the process independently. They are managing the performance of the system as a whole.

We are already seeing what this looks like in practice. Mills are reducing resin usage by >10% while improving consistency and throughput. Quality variation is narrowing by 30% or more, while throughput and uptime improvements in the range of 3-8% are becoming achievable without sacrificing product performance. At the same time, energy use, waste, and carbon intensity are declining. These are not incremental improvements. They are system-level gains driven by faster decisions, better visibility, and more adaptive operating models.

This is also where the relationship between suppliers and manufacturers begins to evolve. It is no longer just about supplying materials into the process. It is about delivering performance across the entire system.

What Comes Next

Wood panel manufacturing is entering a period where traditional operating models will increasingly struggle to manage the complexity and variability built into the process. The mills that lead the next decade will not simply run today’s systems more efficiently. They will operate differently, with more connected intelligence, faster feedback loops, and greater control over how materials and processes interact in real time.

The foundation for that shift already exists. The data is already being generated. The expertise is already in the mill. The opportunity now is to connect those capabilities in a way that allows the system to become more adaptive, responsive, and precise over time.

At Hexion, this thinking is shaping the way we approach the future of the industry. Our Chemistry-as-a-Service™ strategy is built around the idea that resins, process intelligence, and AI can no longer operate independently. They need to function as part of a connected system focused on delivering measurable operational performance.

That means helping mills move beyond static resins formulations and reactive operating models toward more adaptive systems that continuously learn, improve, and respond to changing conditions in real time. It also means creating a closer partnerships with our customers, where success is measured not simply by the products we delivered, but by the performance achieved together.

That is the direction the industry is moving. And increasingly, it is where the next generation of mill performance gains will come from.

‍