Engin Topan

The University of Twente is a public technical university located in Enschede, Netherlands. The university has been placed in the top 170 universities in the world by multiple central ranking tables. In addition, the UT was ranked the best technical university in The Netherlands by Keuzegids Universiteiten, the most significant national university ranking. The UT collaborates with Delft University of Technology, Eindhoven University of Technology and the Wageningen University and Research Centre under the umbrella of 4TU and is also a partner in the European Consortium of Innovative Universities (ECIU). Together with colleagues from my university, as part of University of Twente, we are part of two initiatives, each with a strong business case: (Relevant to Cluster 4) Working with high-tech SME manufacturers in the Netherlands and their challenges, we develop advanced AI-driven planning and quotation systems. We particularly focus on the challenges of SME manufacturers operating in high-mix, low-volume (HMLV) environments. Existing planning systems lack the flexibility, adaptability, and predictive capabilities required to manage resource allocation effectively, respond to customer demands, and optimize production processes. Moreover, these challenges are not only relevant for manufacturers managing their own production lines but also for equipment manufacturers and machine builders offering their systems as a service (Manufacturing-as-a-Service, MaaS). These companies seek to create added value by enabling customers to optimize the utilization of installed machines and manufacturing systems. To address the challenges of HMLV manufacturing, we focus on: • Hybrid AI Approach: Combining model-based, rule-based, and AI-driven methods to ensure both computational efficiency and flexibility in planning. • Feature-Based Planning: Capturing product-specific attributes early in the process to improve resource allocation and scheduling. • Simulation-Enhanced Quotation: Predicting lead times, throughput, and bottlenecks to provide more accurate and reliable quotations. • Interactive Human-AI Collaboration: Ensuring that AI-driven planning aligns with human decision-making through explainable AI and reinforcement learning from human feedback (RLHF) to improve usability and flexibility. By integrating hybrid AI, simulation-based decision support, and interactive learning mechanisms,we provide a scalable and explainable solution for SME manufacturers and machine builders offering MaaS, improving quotation accuracy, production efficiency, and scheduling adaptability. Related projects COLMAN TIMELY SCOPE TRANSFORM Relevant calls: HORIZON-CL4-2025-04-DIGITAL-EMERGING-07: Enhanced Learning Strategies for General Purpose AI: Advancing GenAI4EU (RIA) (AI/Data/Robotics Partnership) HORIZON-CL4-INDUSTRY-2025-01-TWIN-TRANSITION-02: Physical and cognitive augmentation in advanced manufacturing (Made in Europe Partnership) (RIA) (Relevant to Cluster 5) Working a regional consortium on multi modal shift for construction logistics. We explore use of inland waterways, Twente Kanaal, to reduce the growing pressure on environmental impact of emissions from logistics and transportation. Our goal is to develop a collaborative multimodal transport model for construction industry utilizing the Twente Canal for transporting construction materials and waste. We aim to simulate use of the canal and benefits, and develop innovative business models to share these benefits among stakeholders to encourage further collaboration beyond our consortium, and finally we aim at establishment of a dedicated construction hub. Twente Kanaal is a also strategical inland waterway on the North Sea-Baltic corridor, connecting to major European sea ports and hinterlands, our research also contributes to the current\[1\] and future EU initiatives. \[1\] https://www.inlandwaterwaytransport.eu/ Relevant calls: HORIZON-CL5-2026-01-D6-10: Integrating inland waterway transport in smart shipping and multimodal logistics chains