Is AI a CEO Task in Manufacturing Companies?

Unlocking the Potential of AI in Global Manufacturing Strategies

I am following the development of AI in the manufacturing industry with a lot of interest and constantly evaluate the status quo, options and opportunities. As part of my network on this topic, I had a very interesting chat with a great Operation Research (OR) specialist and friend, Michael Mederer (https://www.dhirubhai.net/in/michael-m-8b454b/) about the current situation of AI in manufacturing and where he mentioned how difficult it is to get the understanding of the capabilities of OR and AI into companies and bring it to the attention of executives. That led me to write and summarize my current point of view on AI in manufacturing organizations with all the great opportunities and challenges.

Artificial Intelligence (AI) is transforming industries across the globe, and manufacturing is no exception. While many organizations see AI as a tool for automating processes or enhancing production lines, AI’s full potential in manufacturing extends far beyond the shop floor. Today, AI can drive innovation, improve efficiency, and enhance decision-making at every level of a manufacturing organization.

Despite this, AI is still largely seen as an IT or digital transformation tool, often disconnected from core business strategy. But in an industry as competitive and process driven as manufacturing, CEO/COOs must take ownership of AI initiatives to ensure their companies remain competitive in a global market. This article will explore the unique challenges and opportunities for AI in manufacturing, the role of process automation and process mining, and how a global manufacturing footprint impacts AI strategy. We'll also discuss how to overcome barriers like data quality and process culture, and whether AI should be centralized or region-specific in global operations.

AI in Manufacturing: An IT Function or a CEO Task?

Manufacturing companies have long been at the forefront of process optimization, thanks to innovations like lean manufacturing and automation. However, AI presents a new frontier that requires a different approach. Traditionally, AI has been viewed as an IT function, focused on automating repetitive tasks, predictive maintenance, and enhancing supply chain operations. While these applications are valuable, AI’s potential in manufacturing extends far beyond these areas.

AI can serve as a key enabler of strategic decision-making, helping manufacturers optimize global operations, improve quality control, enhance customer personalization, and even drive product innovation. This means AI should not be siloed in the IT department but must be part of the CEO’s strategic agenda.

Why AI Strategy Matters

A well-defined AI strategy is essential for manufacturing companies to harness AI's full potential. According to a 2023 Deloitte survey, only 18% of manufacturers have a comprehensive AI strategy, while many are still exploring how to integrate AI into their operations. Without a clear strategy, AI initiatives often remain fragmented and fail to deliver meaningful business results.

A successful AI strategy in manufacturing should address key areas like demand and supply-planning, inventory planning, production optimization, supply chain management, continuous improvement, problem solving and quality control, while also considering how AI can support product innovation and customer needs. Importantly, the strategy should also consider the company’s global footprint and how AI can drive value across multiple regions.

Process-Driven Systems and AI: A Clash in Manufacturing?

One of the biggest challenges in manufacturing is that the industry is highly process driven. Systems like ERP (Enterprise Resource Planning) and MES (Manufacturing Execution Systems) have been designed to optimize and standardize processes, ensuring consistency and efficiency across production lines. These systems are typically rule-based, highly structured, and often rigid - qualities that can make them incompatible with AI, which thrives in flexible, dynamic, and data-driven environments.

In process-driven environments like manufacturing, integrating AI requires a delicate balance. ERP systems are essential for managing core business processes, but their structured nature can limit AI’s ability to generate insights and optimize performance. AI, by contrast, relies on vast amounts of unstructured data to identify patterns, make predictions, and offer recommendations that improve decision-making and process efficiency. As Michael mentioned – “load the full ERP database into the cloud and send back the findings of the AI results in form of master-data, rule adjustments”

?The Role of ERP and MES Manufacturers

ERP and MES manufacturers are aware of these challenges and are beginning to integrate AI capabilities into their platforms. Companies like SAP and others, for example, are developing AI modules to enhance their ERP systems, while MES vendors are introducing AI tools to improve real-time data analysis on the shop floor.

However, the adoption of AI in these structured environments has been slow, primarily due to the complexity of integrating AI with existing systems. Manufacturers must work closely with ERP and MES providers to ensure that AI can be seamlessly embedded into their operational workflows, allowing them to unlock the full potential of AI-driven insights. Has what was once considered best-in-class now become a burden?

Process Automation and AI in Manufacturing: Opportunities and Challenges

Process automation has long been a priority for manufacturing companies, and AI offers new opportunities to extend automation beyond traditional boundaries. With AI, manufacturers can move from rule-based automation to cognitive automation, enabling machines and systems to not only follow instructions but to understand, learn, and improve over time.

Examples of AI-driven process automation in manufacturing include:

• Demand and supply-planning – commonly know as SIOP (Sales, Inventory and Operation Planning): a holistic model of optimization is giving a “best cost”, or “best profit” opportunity.
• Predictive Maintenance: AI can analyze sensor data to predict when equipment will fail, reducing downtime and optimizing maintenance schedules.
• Supply Chain Optimization: AI can analyze global supply chain data to forecast demand, identify bottlenecks, and suggest more efficient routes, reducing costs and improving customer satisfaction.
• Quality Control: AI-powered computer vision systems can inspect products in real-time, identifying defects and ensuring consistency in quality at a scale that far exceeds human capabilities.

Despite these benefits, the integration of AI into existing process-driven environments remains a challenge. Many manufacturing companies struggle with outdated infrastructure, data silos, and a lack of skilled talent, all of which can hinder the success of AI initiatives.

Process Mining and AI: Unlocking New Insights

Process mining is an emerging tool that holds significant promise for manufacturing companies. Process mining uses data from ERP, MES, and other systems to analyze how processes are executed, rather than how they are documented. It identifies bottlenecks, inefficiencies, and deviations, offering valuable insights into areas for improvement.

When combined with AI, process mining can deliver even greater value. AI can use the insights generated by process mining to predict future inefficiencies, recommend process changes, and continuously optimize operations in real-time. For manufacturers, this means moving from reactive process management to a proactive, data-driven approach that continuously improves performance. Still and here we have the same technological constraint are the existing tools that are based on prior generation algorithms.

Global Footprint: Should AI Be Centralized or Regional?

For manufacturing companies with a global footprint, one of the key strategic questions is whether AI should be managed centrally or regionally. Should there be a "one-size-fits-all" AI approach across the entire organization, or should AI initiatives be tailored to the needs of each region?

There are pros and cons to both approaches. A centralized AI strategy ensures consistency and standardization across the organization, making it easier to manage AI initiatives, share data, and scale successful use cases. However, regional variations in manufacturing processes, customer demands, political, and regulatory (data protection law) requirements may limit the effectiveness of a centralized approach.

On the other hand, a regional AI strategy allows manufacturing companies to tailor AI solutions to the specific needs of each market. For example, production lines in Europe may require different AI-driven optimizations than those in Asia, due to differences in customer preferences, labor costs, and regulations.

In my opinion the best approach will depend on the company’s specific needs, but in most cases, a hybrid strategy works best. AI initiatives can be centrally coordinated, with regional adaptations to address local market conditions and operational differences.

Why AI Isn’t Yet Showing Up in the P&L

Despite the growing investment in AI, many manufacturing companies have yet to see the financial benefits reflected in their profit and loss (P&L) statements. Several factors are contributing to this delay:

• Data Availability and Quality: AI relies on vast amounts of high-quality data, but many manufacturers struggle with data silos, incomplete datasets, and inconsistent data standards.
• Process Culture: Many manufacturing organizations have a deeply ingrained process culture that values standardization and control, which can conflict with AI’s dynamic, data-driven nature.
• Integration Challenges: The complexity of integrating AI into legacy ERP, MES, and other systems is a significant barrier. This often requires costly upgrades and new infrastructure investments.
• Lack of AI Talent: Many manufacturers lack the in-house expertise needed to develop and implement AI solutions effectively.

AI and Lean Manufacturing: A Powerful Combination

Lean manufacturing and AI have a natural synergy. Both aim to reduce waste, improve efficiency, and optimize processes, but they approach these goals in different ways. Lean manufacturing relies on human-driven continuous improvement, while AI uses data to automate and optimize processes.

When combined, AI can enhance Lean manufacturing by identifying inefficiencies more quickly and accurately. For example, AI can predict demand fluctuations, enabling more efficient production planning, or identify subtle quality issues that human inspectors might miss, improving overall quality control.

Process Culture: Enabler or Hinderer of AI?

The process culture within a manufacturing organization can either support or hinder AI adoption. Companies with a rigid process culture may struggle to embrace the flexibility and experimentation required for successful AI implementation. To overcome this, manufacturers must foster a culture that values data-driven decision-making, continuous learning, and agility.

Future Competitive Advantages for Manufacturers Using AI

For manufacturers, the future competitive advantages of AI are significant. Companies that successfully integrate AI into their operations can expect:

• Improved Operational Efficiency: AI can automate routine tasks, optimize production schedules, and reduce downtime, driving greater efficiency.
• Enhanced Quality Control: AI-powered systems can inspect products with greater accuracy and speed, reducing defects and improving customer satisfaction.
• Faster Decision-Making: AI provides real-time insights that enable executives to make faster, more informed decisions, improving agility and responsiveness.
• Product Innovation: AI can analyze market trends and customer preferences, helping manufacturers design new products that better meet consumer demands.

Preparing for AI: Data Availability and Quality

To fully leverage AI, manufacturers must focus on improving data availability and quality. This includes breaking down data silos, implementing robust data governance practices, and investing in modern data infrastructure that can support AI workloads.

By ensuring that high-quality data is available across the organization, manufacturers can unlock the full potential of AI, improving efficiency, enhancing quality, and driving innovation.

Every problem, ever incidence, every claim, every accident, just everything – including meeting minutes are providing the source of data that will give AI the power to deliver value. In my understanding it is not a single cut-off point that counts, but the continuous process of data generation across the whole organization.

Conclusion

For manufacturing companies, AI is far more than an IT tool—it is a strategic imperative that should be driven by CEO/COOs. From process automation to process mining, and from global operations to quality control, AI has the potential to revolutionize every aspect of manufacturing. However, success requires overcoming challenges like data availability, rigid process cultures, and integration with existing systems. By developing a robust AI strategy that aligns with their global footprint and business objectives, manufacturers can unlock new competitive advantages and ensure long-term success in a rapidly evolving market.

Let me know what is your experience and understanding of AI in manufacturing today!

Yours sincerely

Hans