How to leverage AI for real-time, actionable insights in paper manufacturing

In the always-on world of paper manufacturing, real-time process control is not just a luxury—it’s a necessity. As market demands fluctuate and raw material quality varies, paper manufacturers must maintain precise control over their production processes to ensure consistent quality, minimize waste, and maximize efficiency. 

Traditional control methods have served the industry well, but these strategies are nearing the end of their usefulness as the market becomes increasingly competitive. Integrating Artificial Intelligence (AI) into process control systems promises a quantum leap forward in terms of operational capabilities and outcomes.

AI enhances traditional control methods by providing teams with deeper insights, faster responses, and more accurate predictions. Unlike conventional systems that rely on predetermined rules and thresholds, AI-powered systems can learn from historical data, adapt to changing conditions, and even anticipate potential issues before they occur. This level of intelligent control enables paper manufacturers to achieve unprecedented levels of efficiency.

AI can help your teams get greater value from your production data by uncovering actionable insights. Let’s take a closer look at some specific ways that AI can build upon your team’s existing knowledge and drive better outcomes.

Stabilize production with advanced centerlining

Centerlining, a technique used to identify and maintain optimal operating conditions, has long been a staple of efficient paper production. The goal of centerlining is to get each aspect of production to operate at maximum efficiency while maintaining quality standards. In theory, if you can stabilize each asset and process from start to finish, you can consistently produce high-quality products with minimum variability.

Traditional centerlining often relies on static set points determined through historical analysis and intrinsic operator knowledge. Early on, this approach can improve processes that are highly erratic or unpredictable. We’ve seen numerous manufacturers leverage shop floor visualizations and digital parameter storage for improved operator compliance and stabilized production outcomes.

Relying on “assumptions” about the best thing to do or manually adding updated data into new analyses are major bottlenecks to achieving optimal performance, though. Over time, the effort-to-impact ratio of these traditional centerlining practices increases: teams are putting in exorbitant resources and hours but not reaping the same benefits they were early on in their centerlining strategy. 

Adopting AI into your company’s centerlining strategy can uncover a new level of sophistication and effectiveness. For example, AI-driven centerlining continuously analyzes vast amounts of process data in real-time and can recommend dynamic set points to operations teams based on current conditions. This approach ensures that production always operates at peak efficiency, even as conditions change throughout the day or between production runs.

Use case: International Paper uses AI to increase Golden Runs

International Paper tracks optimal production cycles, known as “golden runs,” to gauge process stability. A golden run is a particularly good set of operating parameters that generate high-quality products for one grade of paper. If teams can achieve golden runs repeatedly, they know that they are consistently compliant to the right production parameters for a given product.

Braincube’s tools make it easier for teams to look at prior runs, choose the runs with the best conditions, and provide operators with guidance about how to set up the next run. For example, using Studio, individuals can automate a report on any batch run to select the grade and speed range they’d like to implement. The app populates with the running conditions of key input variables from the most recent “golden batch.” In this sense, the dashboard recreates a rule/production standard in real-time based on the most current data, giving operators guidance as they set up for the next grade run.

However, it doesn’t mean that teams can repeatedly run the same parameters and expect the same results. Like any paper manufacturer, International Paper’s operations teams must continuously adjust for set-point drift, grade changes, asset performance, and other variables.

“If you just go back to the last settings you ran, after about six or seven runs you realize you’ve drifted off into some poor setting,” said Andrew Jones, Senior Principal Engineer at International Paper. “No one really notices it happening, but it happens. With Braincube, teams can evaluate what happened and ask themselves ‘How do we run this next grade best?’”

With Braincube, teams can evaluate what happened and ask themselves “How do we run this next grade best?

Andrew Jones, Senior Principal Engineer at International Paper

Pulling the most recent “golden” settings into their shop floor dashboards means operations teams can account for process drift. They aren’t simply relying on what worked many months or years ago: they are continuously using the most up-to-date settings that led to good performance

The team leverages AI to uncover production trends and identify new best practices that could be implemented as new standards. AI can analyze historical production data to establish a baseline of what to expect from key processes under various conditions. However, the baseline isn’t static; it evolves as the AI continues learning from each production run. The system can then use this continuously updated model to predict how changes in input variables will affect the final product. Jones likens this process to incorporating live AI recommendations into tried-and-true methods based on historical data. 

In other words, AI bridges the gaps between what’s happening now and what has worked well previously, providing recommendations that improve centerlining and process stability outcomes across end-to-end production.  

Improve outcomes with advanced real-time alerts and recommendations

One of the most powerful applications of AI in paper manufacturing is in the realm of real-time monitoring and adjustment. AI systems can simultaneously track hundreds of process variables, identifying subtle patterns and correlations that are impossible for human operators to monitor and detect.

This is particularly valuable for alerting operations teams about pressing or concerning situations taking place on the shop floor. Paper manufacturing facilities are behemoths: your workforce cannot possibly be everywhere at once. This is why most paper manufacturers have already adopted an alerting strategy that pings operations teams when something is amiss. Alerts provide operators with visibility into what’s happening on the shop floor with greater ease. 

Sometimes, though, traditional alerting systems can backfire. They often generate a flood of notifications for every minor infraction occurring on the line. These systems typically lack the nuance to identify the difference between a small error and a situation that could lead to a paper break. 

AI-powered alerting systems, on the other hand, can be configured to generate intelligent alerts. These alerts are contextual and prioritized, removing the guesswork regarding which situations are mission-critical and which ones can temporarily move to the backburner. Teams can also use AI to find optimal setting ranges and set smart alerts for when parameters drift outside these ranges. 

Moreover, AI-powered alerting systems can learn from historical data and predicted outcomes to go beyond simply informing operators about the issue: they can also provide recommendations for corrective actions. Process experts and engineers can validate these recommendations, push them to the shop floor, and incorporate these updated recommendations into future corrective action workflows.

Use case: reducing costly downtimes with real-time alerting

A global paper manufacturer faced significant challenges with costly downtimes due to a lack of live data insights. Operators could only react to issues one at a time after they occurred, and identifying root causes was time-consuming. 

To address these problems, the company implemented a proactive troubleshooting strategy using AI and an advanced analytics platform. They worked with their Braincube team to connect all relevant IT/OT data sources into a centralized repository. This integration allowed for live condition monitoring and advanced alerting capabilities across the entire production stream. 

Once contextualized data was centralized and made accessible to all teams, supervisors set up a custom alerting strategy. This new alerting strategy leveraged AI recommendations and prioritization, enabling operators to receive immediate and relevant notifications when critical conditions arose.

Braincube’s live dashboards and analytics also provided operators and engineers continuous access to a suite of analytical tools, including control cards, one-hour trends, and asset metrics. These real-time data visualizations empowered teams to quickly identify the root causes of production drifts and troubleshoot issues before they escalated into costly downtimes.

By shifting from a reactive to a proactive troubleshooting strategy, the paper manufacturer reduced unplanned downtime by 30% and saw a 10% increase in production output. This data-driven approach enhanced production stability while fostering a more efficient and responsive operational culture.

Operate at optimal capacity by improving the controllability of specific variables

In paper manufacturing, several key variables (such as tensile strength, bulk, burst strength, and porosity) must be carefully controlled to meet different product specifications. AI systems excel at optimizing these variables, often achieving levels of control that surpass what’s possible with traditional methods.

What sets AI apart is its ability to understand and manage the complex interrelationships between these variables. It is simply not possible for human-driven analyses to incorporate and dissect data relationships that are this complex. But if your paper mills are to overcome the plateaus of traditional analysis, understanding and acting based on these multivariate relationships is crucial for continuous improvement. 

For instance, if your team implements a parameter change designed to improve tensile strength, they may inadvertently negatively impact bulk or porosity. AI can navigate these multivariate trade-offs, finding the optimal balance that meets all required specifications.

Moreover, AI can help manufacturers compound the value of their improvements. As the system optimizes one variable, it simultaneously analyzes how these changes affect other key variables, ensuring that improvements in one area don’t come at the cost of degradation in another. This holistic approach generates a snowball effect of positive impacts across the entire operation.

Use case: ROI achieved in one week with SPC reporting

Oji Paper, a global specialty paper producer, wanted to optimize its paper coating process and reduce costs. Facing strong market competition, Oji Paper’s teams identified chemical coating as their main source of cost variability, with 65% of batches exceeding the target coating amount. Even with this knowledge, though, they lacked the nuanced information to hone in on the right variables that would reduce 

The team brought in Braincube to map the entire production process and integrate data from multiple sources. Once data was integrated and contextualized, process experts ran AI analyses on 2,450 process variables within the chemical coating process. AI revealed a combination of just four key parameters that could maintain coating targets while reducing chemical consumption.

Braincube provides process engineers with powerful Big Data tools to analyze the connections between all the process variables, to immediately highlight the variables having an impact, and to create strong multi-variable rules.

Lincoln Tsuruta, Oji Paper Process Engineer

By implementing these AI-derived optimal settings and providing real-time recommendations to operators via Braincube Live interfaces, Oji Paper was able to reduce chemical consumption by 10% within days of implementation. The team also improved overall stability of their coated paper products. 

Overall, the Oji Paper team found AI was remarkably helpful in identifying the most impactful variables and determining optimal setpoints in their complex manufacturing processes. By focusing on these key parameters, Oji Paper was able to rapidly reduce costs, minimize waste, and improve product quality—ultimately enhancing their competitiveness in the market.

Conclusion

The integration of AI into real-time process control represents a significant leap forward for the paper manufacturing industry. By leveraging advanced centerlining, real-time alerts, and improving the controllability of specific variables, paper manufacturers can achieve unprecedented levels of efficiency, quality, and consistency.

It’s important to remember that the journey doesn’t end here. As manufacturers become more adept at using AI for real-time process control, they pave the way for even more advanced use cases. The next stage involves moving from reactive and real-time control to predictive and prescriptive analytics, in addition to autonomous operations. In this stage, AI controls current processes while also optimizing future operations and guiding strategic decision-making.

The key to success in your AI journey is a methodical, step-by-step approach. By starting with foundational use cases like those discussed in this article, paper manufacturers can build the organizational capabilities and AI expertise needed to tackle more complex challenges in the future. As the paper manufacturing industry continues to evolve, those who embrace AI-powered process control will be well-positioned to lead the way in efficiency, quality, and innovation.