For decades, Manufacturing Execution Systems (MES) have been the steadfast operational core for many in the paper and packaging industry. These systems, often developed 20 to 30 years ago, have been meticulously customized to the unique demands of our production environments, from pulp processing to papermaking to intricate converting and printing lines. They have been the guarantors of efficiency and a key pillar of competitive advantage, deeply intertwined with the shop floor. However, what was once a strength is now, for many, becoming a technological bottleneck.

The Challenge of Aging Champions

The reality is that many of these legacy MES solutions, while functionally rich, are technologically antiquated. The skilled individuals who possess the deep knowledge to maintain and adapt these systems are gradually heading into retirement, creating a critical knowledge gap. Furthermore, these robust but rigid systems are struggling to keep pace with the rapid IT advancements we're witnessing elsewhere, particularly in the ERP space with the rise of Generative AI and intelligent agents. The very highly specific design that made them powerful now makes them inflexible in the face of new opportunities. This technological lag isn't just an operational concern; it also impacts cyber-resilience. Outdated systems can present security vulnerabilities, a factor increasingly scrutinized by banks and investors during company valuations, potentially affecting financial standing and investment appeal.

Shifting Focus: From Process Links to Data Goldmines

Historically, the primary focus of MES-ERP integration was on transactional efficiency – think seamless production order confirmations and goods movement postings. While this remains important, the advent of AI has dramatically shifted the spotlight towards achieving easy, unified, and semantically rich data provisioning. This shift is essential for effectively converting raw data into valuable information. The underlying requirement is to integrate detailed data from diverse sources, such as ERP, MES, and process data systems, into a single, semantically rich data platform. Solutions like SAP Business Data Cloud exemplify this approach. SAP Business Data Cloud aims to prevent data replication, thereby preserving the original business context and semantics of the data, avoiding the costs associated with data extraction, and ultimately saving time, resources, and effort.

Paper and packaging companies have been accumulating vast amounts of production data for years, if not decades. This data is the fuel for powerful AI scenarios that can unlock new levels of competitiveness. Imagine being able to accurately predict which customer order configurations are most likely to cause production issues, or identifying subtle process deviations that impact quality before they escalate. This requires a more comprehensive and accessible data strategy than many current MES architectures can support.

Re-evaluating the MES-ERP Interplay: A Call for Holistic Integration

The temptation might be to simply replace an old MES with a newer, off-the-shelf MES. However, this like-for-like replacement may not be enough to unlock the transformative potential that new technologies promise. Instead, a fundamental rethink of the MES-ERP relationship is required. It's time to break down the traditional silos and envision a true "top-floor-to-shop-floor" solution. This involves critically assessing which functionalities are best suited to reside within the ERP versus the MES. For instance, Quality Management, often a core MES function, is increasingly being migrated to ERP systems. This makes sense, as ERPs typically house crucial data and processes related to customer order requirements, certificate generation, and overall compliance – data that is essential for a holistic quality approach.

The Imperative of Technology, Openness, and Adaptability

The dynamic advancements in GenAI, and the future promise of agent-based systems, offer significant potential for optimizing manufacturing operations. To remain competitive, the paper and packaging industry must be able to leverage these innovations. This necessitates that MES and ERP systems are not just loosely coupled but deeply integrated, seamlessly connecting data and allowing processes to flow intelligently. Bespoke, in-house developed MES platforms will find it increasingly challenging to keep pace with this technological evolution and offer the necessary openness and scalability.

A Platform Approach: SAP's Vision for a Modern MES

Recognizing this need, SAP has adopted an approach for its MES solutions that mirrors the successful strategy employed in its ERP systems. SAP Digital Manufacturing (DM) is conceived as an open and adaptable MES platform. SAP provides the core functionalities, estimated to be around 80-95%, forming a broad foundation that caters to a wide spectrum of manufacturing processes, including mill, discrete, and chemical industries. This approach, which transcends specific industries, results in a broader user base and fuels continuous innovation, development, and technological advancement through cross-sector insights.

The crucial in-deep industry-specific capabilities are then delivered by a rich ecosystem of partners. This collaborative model ensures that the MES solution is always at the forefront of technology (including AI capabilities), aligns with the latest data and process best practices, and meets stringent compliance and cyber-resilience requirements – a critical factor in our industry. Importantly, this approach also offers flexibility in user interface design, process workflows, and data model extensions, allowing for precise tailoring to the unique demands of paper and packaging, such as complex labeling, cutting, printing and tracking processes.

Pioneering the New MES Paradigm: Real-World Success

This architectural vision is already being successfully implemented. SAP partner Delaware offers their DM4MILL solution. This was built on SAP Digital Manufacturing and offers a comprehensive MES tailored specifically for mill products industries. This is a great example of how a core platform can be expertly extended to meet deep vertical needs. Similarly, SAP partner CONCIRCLE is offering add-ons through their conMILL Suite for SAP Digital Manufacturing, also designed specifically for mill products industries. Furthermore, SAP partner T.CON is pursuing a comparable strategy, leveraging SAP’s digital manufacturing platform to deliver specialized MES capabilities for die rolls and format manufacturing. Their work, for instance, in addressing how SAP Digital Manufacturing can handle intricate trim and cutting processes, highlights the practical application and benefits of this modern approach.

Embracing Change for a Competitive Future

The MES systems that have served the paper and packaging industry so well are reaching an inflection point. Continuing with technologically outdated, siloed systems is a path towards diminishing returns and increased operational risk. The future lies in embracing integrated, open, and intelligent platforms that can harness the power of data and AI. By rethinking the MES-ERP relationship and looking towards collaborative, platform-based solutions, a new era of manufacturing excellence and sustained competitiveness in the paper and packaging sector can be paved.

To find out more about SAP Digital Manufacturing click here.  For more about how SAP has partnered with the Mill Products industry for more than 50 years click here.  

I have recently been considering the potential danger of artificial intelligence (AI). I have been watching interviews of people like Marvin Minskey (co-founder of MIT AI laboratory), Geoffrey Hinton (the godfather of AI), and others. Of course, anyone that has watched any Black Mirror episode would be afraid. Considering that AI is a significant attribute of my profession and credentials, I should consider the concerns sincerely.

In a prior article I mentioned the uncertainty in defining AI. The definition I will use is any computer application that passes the Turing test. Alan Turing was a brilliant mathematician that, among his many accomplishments, includes the origin of this AI definition. Any computer application which produces results indistinguishable from human intelligence or better is artificial intelligence. That is the Turing test. One of those applications is automation. If you can’t tell whether a really smart operator or a computer is controlling your process, it qualifies as AI (in my opinion).  

If we agree to that definition, what is the danger? They consist of:

• Unemployment: if computers/machines can replace humans, are cheaper, and more productive, then humans won’t have a way to earn income.
• Subservience: if computers become smarter than us, they can make us slaves in the same way we use cows to give us milk and burgers.

These are frightening concerns, and people like Hinton are warning us that we are getting close to a tipping point. They say now is the time to get AI under our control before it gets away from us.

Another prominent actor in AI is Ray Kurzweil, who was the subject of one of my prior articles. Kurzweil calls the tipping point the “singularity”. In the opinion of Kurzweil, he sees tremendous benefit from AI that will make humans more powerful. He doesn’t seem to share the concerns of Hinton.

I am not a futurist. I have read authors like Alvin Toffler who seem to have an impressive way of reading trend lines and predicting where they will go in the future. I am much less confident in my prognostication. However, I do have some opinions.

To begin, I am going to constrain the domain to industrial automation. What is not universally understood outside of manufacturing industries is that the AI techniques that apply to chatbots, finance, and social media are not always the best when applied in industry. Methods such as Neural Networks can have a place, but there are inherent problems when Neural Networks try to overfit noisy data that has not been pre-processed by humans. Results can be disastrous when applied to closed loop control. We have first principles that tell us steady state models rarely have more than 1 inflection point, and dynamic models don’t need more than 2^nd order dynamics with lead and deadtime accounted for. We know what the models should look like. Tried and true methods with first principle foundations qualify as AI, but have been time tested and are not scary.

If we step outside of our industrial domain, we need to ask if there is any attribute of humanity that computers cannot achieve. In “Algorithms to Live By” (Christian and Griffiths), it is made clear that the AI techniques that fail are those that try to be too perfect. The techniques that are successful more closely approach our human imperfections. That suggests that there is an attribute of humanity that is asymptotic. That attribute is creativity.

The way I explain this to students at Miami University is as follows:

• Premise 1: There will always be problems to solve
• Corollary 1: There will always be a demand for solutions to problems
• Premise 2: Some problems require creativity (https://www.businessinsider.com/accidental-inventions-that-changed-the-world-2014-5)
• Premise 3: Humans are the ideal creative machine
• Thesis (Corollary 1 + Premise 2 + Premise 3) = Machines will not entirely replace human labor
• If you want to make the world better, have a baby

Of course I could be wrong, but since that has never happened that seems improbable.

I think we should be cautious about any technology. Just like nuclear weapons and recovery boilers, there is potential for AI to be dangerous. If you have competent and enlightened people to work with when you are applying AI, your will be safe and profitable.

I am so old that back when I started, the "M" in MES stood for manual. And AI was HAL in "2001-a Space Odyssey." Then there is the Catholic Church who now has a Pope younger than me. I digress.

I was just talking a couple of days ago to a senior leader in our industry. He has one foot in agribusiness and the other in pulp and paper, as I do. We were agreeing that for 150 years agriculture technology of all aspects--mechanical, technical and biological--has led the way in the modern world.

We are catching up in industry, particularly pulp and paper. Much of our advancement in the last three or four decades has been in computerization. Much of that in MES and ERP. In the 1980's, for instance, a trained paper machine planner could beat any computer at "trimming out" a machine. No more. In the 1980's, a blip in your computer system did not stop the paper machine--it does now because you can't print roll labels.

We accept these hazards for the benefits they provide 99.9% of the time (your backup power supply needs to be high on your maintenance list as well as your cyber security).

Now comes AI. The promises look great; the dangers are of the order of magnitude of HAL mentioned above. Reread Pat Dixon's column.

IoT technology is becoming a cornerstone of digital transformation across global industries, but cost and complexity are still holding back more advanced use, according to GSMA Intelligence’s latest report, “IoT & eSIM for Digital Industries.”

AI is everywhere. And it's growing rapidly in IoT as implementers are learning that cloud-based AI dependency is slow (due to latency), expensive (owing to power-hungry servers and transmission costs), not always reliable (due to link downtime), and poses a privacy risk. Performing more AI processing locally on IoT devices helps address these issues. 

A recent whitepaper by ABI Research delves into the far-reaching impacts of Trump’s trade policies on the tech industry.

Learn how to balance scalability, flexibility and security for smarter, data-driven production using on-premises, cloud and/or edge data infrastructures.