Expert Insights Series: AI and Digital Twins in shaping the future of Supply Chains.
AI and Simulation: A Powerful Combination for Smart Supply Chains
As AI continues to revolutionize industries, its integration with simulation tools is opening new doors for efficiency, risk management, and decision-making in supply chains. We spoke with Kees van der Klauw, former Chairman of the Netherlands AI Coalition and former Research Executive at Philips, about the synergy between AI and simulation, and what the future holds for smart supply chains.

Can you share your background and how you became involved in AI and innovation?
My career started at Philips, where I was involved in digital transformations across various domains—semiconductors, LCD displays, TV innovations, and LED lighting systems. Over a decade ago, in Philips Research, we already explored AI applications such as automated design of electronic circuits and customer preference analysis. Later, I played a key role in establishing AIOTI (now a leading IoT and Edge Computing initiative) and led the strategy and development of the Netherlands AI Coalition. This coalition now consists of over 500 parties and nearly 2,000 experts working on AI applications across multiple industries.
What is simulation? What is AI? How do they complement each other?
Simulation is used to predict outcomes in scenarios that are too complex, costly, or risky to test in real life. With advancements in computing power, we can now create highly detailed simulations, but they still require strong domain knowledge. AI, on the other hand, is built on statistical models trained on massive datasets. While simulation models are generally based on known physical principles, AI finds correlations within data, sometimes revealing hidden insights. The great opportunity we now have is not to replace the one with the other but to augment simulation models (which usually have a very limited number of parameters) with AI algorithms that add statistical intelligence on effects that until now were too complex or simply unnoticed to include in simulation tools.
Where does AI running on a digital twin differ from AI running on raw input data?
Many AI models are trained on large datasets to detect patterns and correlations, but this does not necessarily mean they understand causal relationships. Training AI solely on raw data requires extensive resources, while digital twins integrate domain knowledge, providing faster, more accurate, and explainable insights. By combining AI with simulation models, we leverage expert knowledge for efficient and precise system behavior predictions. This hybrid approach enables accurate manufacturing simulations while accounting for unpredictable factors like human behavior or equipment failures, ultimately leading to smarter decision-making.
How is simulation shaping supply chain management, and how will AI enhance it?
Simulation has already revolutionized the supply chain industry, particularly in material handling and manufacturing, where internal goods flow management relies heavily on simulation models. Today, these models extend across multiple production sites, enabling integrated and efficient operations. Smart Industry initiatives further enhance this by facilitating programmable manufacturing lines where production stages communicate seamlessly. In complex assembly lines with multiple suppliers, ship-to-line logistics has become a standard practice.
Optimizing logistics—covering warehouse space utilization, cycle times, time-critical deliveries, loading, transport costs, and more—is achievable through simulation. However, a key challenge remains: various stages, sites, machines, and transport systems are often managed by different entities and suppliers. Since these elements are not always part of a unified simulation model, fine-tuning is essential to ensure accuracy. To achieve this, models must be parametrically adjustable, tuned by domain experts, and supported by strong data-sharing collaboration across the value chain. This need for seamless data exchange becomes even more critical with AI integration.
With advancements in AI, we will soon be able to create highly accurate digital twins of complex logistics flows—both within individual companies and across entire supply chains. These digital twins will not only drive efficiency and support risk analysis but also act as real-time decision-making companions during supply chain disruptions. Beyond optimizing logistics, AI-powered simulation will contribute to sustainability by tracking CO₂ footprints, improving reliability and flexibility, and ensuring compliance with regulations.
How can the combination of AI and simulation improve decision-making?
Simulation provides outcomes based on set parameters, but human decision-makers must still interpret the results, considering aspects like regulations, financial risks, and operational constraints. AI can enhance this process by augmenting simulation engines with AI models (not replacing them), creating comprehensive digital twins that support real-time, data-driven decision-making.
“
AI is not just an efficiency tool – it is a competitive necessity.
What industries are leading the way in AI and simulation innovation?
Advancements are happening across many industries, but the most impactful innovations transform tedious yet expertise-driven tasks. Key sectors benefiting from AI include logistics, manufacturing, healthcare, cybersecurity, and energy. AI is optimizing everything from transportation flows to medical diagnostics, driving efficiency and accuracy. There is also an impressive contribution by AI in advertising and marketing and administrative processes.
However, because data usage is subject to privacy and security regulations, AI adoption is progressing fastest in less sensitive areas—focusing on machines rather than personal data. In supply chain management, the potential is enormous, offering opportunities to enhance efficiency, resilience, and decision-making on an unprecedented scale.
What excites you most about AI developments in simulation software?
While there is a lot of hype around generative AI and large language models, I believe the most meaningful advancements will come from dedicated machine learning models tailored to specific fields such as supply chain management, healthcare diagnostics and drug development, education, security, energy, and transportation. Machine Learning will drive major improvements in efficiency, quality, and cost reduction by automating tedious human tasks. Additionally, I foresee AI-powered simulation tools becoming more efficient, running on small-footprint systems rather than energy-intensive data centers.
This could mean local servers within companies or even AI-driven IoT devices embedded in equipment or transport vehicles. Such a distributed approach offers significant advantages, including enhanced cybersecurity, resilience, and in energy management, making AI adoption more sustainable and practical across industries.
How do you see AI contributing to sustainability in supply chain operations?
Sustainability is complex, often requiring trade-offs between different environmental and economic factors. AI models can process large-scale, statistical data to develop more holistic sustainability strategies. By integrating AI with simulation, businesses can automate environmental impact assessments, optimize energy usage, and improve waste management.
“
The great opportunity we now have is not to replace the one with the
other but to augment simulation models (which usually have a very
limited number of parameters) with AI algorithms that add statistical
intelligence on effects that until now were too complex or simply
unnoticed to include in simulation tools.
What are the biggest challenges companies face when integrating AI into their systems?
The primary challenge is the availability and quality of data to train AI systems. Many companies struggle to collect and extract meaningful insights from dispersed systems. Key data—such as machine uptime, cycle times, and waiting times for transport robots—often remains siloed and underutilized.
Another challenge is acquiring the right expertise. Companies typically need to bring in data scientists or partner with startups, as existing personnel may lack the specialized skills for AI projects. At the same time, experienced employees are vital for identifying high-value use cases and offering domain knowledge.
Finally, strong management commitment is crucial. Leaders must educate themselves on AI’s broader implications, including dependencies on external platforms and control over key business processes, rather than simply following trends.
What key skills should companies develop to maximize AI in simulation?
Companies should first master simulation for their core processes, ensuring that AI enhances rather than replaces their models. Additionally, data management expertise is crucial, as AI depends on high-quality data. Businesses must also educate employees on AI’s role helping them in their daily work, fostering a culture of data-driven decision-making.
“
With advancements in AI, we will soon be able to create highly accurate digital twins of complex logistics flows—both within individual companies and across entire supply chains.
What are the biggest pitfalls executives should watch for?
Executives must differentiate between primary and secondary processes when applying AI. For example, using AI for marketing content generation is a low-risk secondary process, while using AI in core operations—such as supply chain optimization—requires deep expertise and business control. Another pitfall is relying too heavily on external AI platforms, which can create long-term dependencies instead of offering real competitive advantages for one’s business.
What if companies do not adopt AI technology?
AI is not just an efficiency tool—it is a competitive necessity. Companies that fail to adopt AI risk losing market relevance as AI-powered competitors optimize costs, mitigate risks, and unlock new business models. However, adopting AI should be strategic, ensuring it enhances core competencies rather than creating dependencies.
Final Thoughts
AI and simulation are not competing technologies—they are complementary tools that, when combined, create more accurate, scalable, and intelligent digital twins. As businesses navigate an increasingly complex and unpredictable world, AI-enhanced simulation will be a game-changer for supply chain optimization, sustainability, and decision-making.
With AI advancing rapidly, companies must embrace innovation, invest in expertise, and develop a data-driven strategy to stay ahead in the ever-evolving supply chain landscape.