Special Sessions

The list of accepted special sessions is available below. You can find more information on how to submit a Special Session Paper at the Paper Submission page.

Approved Special Sessions

1. Transferring Mobile Robotics Solutions into Emerging Domains: From complex environments to collaborative fleets (Code: 6m9c1)
2. Robotics Ontologies: Development and Applications (Code: h9ame)
3. Neuro-Symbolic AI: Empowering Trustworthy and Embodied Robotic Autonomous Life Cycle Manufacturing (Code: w8ghk)
4. Human-Robot Collaboration for Futuristic Human-Centric Smart Manufacturing (Code: z8xe4)
5. Intelligent Assembly and Assembly Quality Control (Code: s47m7)
6. Advancements in Modeling, Scheduling, and Control for Robotized Manufacturing Systems (Code: xe2vb)
7. Cognitive Manufacturing Systems (Code: v5374)
8. Emerging Automation Technologies in Core Engineering Systems (Code: 4qca6)
9. Advances in AI-Driven Scheduling and Optimization for Autonomous Manufacturing Systems (Code: g4r1f)
10. 3D Point Cloud Processing and Spatio-Temporal Modeling (Code: cmhjr)
11. Novel Planning and Control Approaches for Semiconductor Manufacturing (Code: x45wg)
12. Collaborative Robot-Empowered Human-Centered and Trustworthy Manufacturing in Industry 5.0 Contexts (Code: psgj8)
13. Leveraging Machine Learning and Big Data Techniques for Reliable and Secure Automation in Healthcare Practices (Code: 5ydsu)
14. System Optimization and Production Control of Intelligent Manufacturing (Code: 4uag3)
15. Analysis and control in the data-driven age, emerging methodologies and applications (Code: vb4c4)
16. AI for Simulation and Optimization in Automation (Code: yj9w4)
17. Frontier of Quality and Reliability Engineering for System Intelligence (Code: x1232)
18. AI-Driven Intelligent Automation for Resilient Complex Systems (Code: 8e67r)
19. Intelligent “Modeling, then Optimization” of Automatic Systems (Code: 4uw8e)
20. Artificial Intelligence-based models and methods for smart logistics, transportation, manufacturing and healthcare (Code: 8m572)
21. Industrial Robot as a Service (IRaaS) (Code: 46g94)
22. Enhancing Healthcare Delivery through Advanced Simulation Techniques (Code: wp5y4)
23. To Automate or To Augment? Towards Human-Centered Systems Advancing Cognitive and Physical Abilities in Industrial Workplaces (Code: 1s972)
24. Advancing Technologies and Sustainability for Production and Service Systems in the AI Age (Code: 4b9ds)
25. 3D Printing and Social Manufacturing (Code: qh296)
26. Automated Optimization for Energy Systems (Code: b7b59)
27. Modeling and Control to Enable Automation, Adaptability, and Reconfiguration in Manufacturing Systems (Code: yc88x)
28. Emerging Data Science in Manufacturing: Autonomous and Software-Defined Factory (Code: e5925)
29. Federated and Distributed Learning for Cyber-Physical Systems (Code: cc9i7)
30. Data Analytics and Optimization for Edge Computing-Enabled Smart Systems (Code: 1vxm5)
31. Process Level Modeling, Control, and Optimization to Address Advanced Manufacturing Needs for Flexibility and Sustainable Automation (Code: 1kn37)
32. Trustworthy virtual experiments and digital twins Design for Metrology 4.0 in the context of Smart Manufacturing (Code: k3xv8)
33. Data-driven and physics-informed modelling for complex cyber-physics industrial systems (Code: kvp15)
34. AI-Powered Collaborative Manufacturing and Scheduling for Industry 5.0 (Code: 256tf)

#1 Transferring Mobile Robotics Solutions into Emerging Domains: From complex environments to collaborative fleets (Code: 6m9c1)

Code: 6m9c1

Mobile robotics has achieved notable success across various fields, delivering market-ready solutions with satisfactory levels of robustness, reliability, and economic viability. However, there are specific areas where such solutions remain absent:

1. For certain user groups, such as small and medium-sized enterprises (SMEs), investing in and operating mobile robotic systems is not currently economically viable due to high acquisition costs and the need for skilled personnel.
2. Complex tasks, such as close collaborative activities or mobile manipulation, are not yet comprehensively addressed by mobile robotics.
3. Most human-centric environments are not suitable for the deployment of current market-available mobile robotic solutions.

The challenges arising from these gaps span technical and economic issues as well as complex research questions. Standardization holds promise for expanding user reach and can make the deployment of mobile robotic systems more feasible and economical, especially for SMEs. Current standardization efforts focus on communication with and among mobile robots. A significant challenge is the coexistence of autonomous mobile robots (AMRs) with fleets of centrally controlled automated guided vehicles (AGVs).

Another critical area is the standardization of communication between robots and their environments. This becomes increasingly important as mobile robots are deployed in less controlled, human-centric environments. With more complex environments, the requirements for mobile robotic systems grow in terms of environmental perception, navigation, and human interaction.Examples of such environments include hospital wards, production facilities and pedestrian areas.

Furthermore, new and complex task domains like collaboration between robots or between a mobile manipulator and humans are gaining attention. Examples of such applications include collaborative material handling by multiple robots, robot-assisted manufacturing, and robotic assistance systems in healthcare.

This session invites contributions that explore these challenges and propose innovative solutions to advance the integration of mobile robotics in specialized applications.

Sub-topics

1. Standardization in mobile robotics communication
2. Fleet coordination for heterogenous fleets
3. Navigation in human-centric environments
4. Environmental perception in complex environments
5. Collaborative Transport
6. Collaborative Manufacturing
7. Motion Planning in Mobile Manipulation
8. Robotic assistance systems

Organizers

• Dennis Lünsch, Research Fellow, Fraunhofer Institute for Material Flow and Logistics IML, E-mail: dennis.luensch@iml.fraunhofer.de
• Fabian Menebröker, Research Fellow, Fraunhofer Institute for Material Flow and Logistics IML, E-mail: fabian.menebroeker@iml.fraunhofer.de

#2 Robotics Ontologies: Development and Applications (Code: h9ame)

Code: h9ame

The special session, “Robotics Ontologies: Development and Applications,” highlights the transformative potential of ontology-based frameworks in shaping intelligent and autonomous robotic systems. By organizing and structuring knowledge, robotics ontologies enable advanced capabilities such as environment perception, adaptive decision-making, and seamless human-robot collaboration. As robotics systems grow more complex within the evolving contexts of Industry 4.0, 5.0, and the emerging Industry 6.0, the role of ontologies becomes increasingly vital for achieving interoperability, scalability, and intelligence.

This session aims to:

1. Investigate the foundational techniques and tools for developing robotics ontologies.
2. Demonstrate real-world applications, focusing on their contributions to agility, sustainability, and enhanced intelligence.
3. Examine emerging challenges in designing collaborative, adaptive, and sustainable robotic systems.

A key focus will be on how ontology development supports the seamless integration of diverse subsystems and applications, enabling real-time data interpretation, more responsive systems, and robust autonomous decision-making.

Given the rapid integration of robotics into domains such as smart manufacturing, autonomous vehicles, healthcare, and environmental sustainability, this session serves as a timely platform for fostering interdisciplinary exchange. Researchers, practitioners, and industry leaders will discuss theoretical advancements and practical applications, aiming to establish pathways toward fully autonomous, human- centric robotics aligned with Industry 5.0 goals.

Sub-topics

1. Ontology Development in Robotics. Exploring innovative tools and techniques for creating and optimizing ontologies across robotic applications.
2. Ontology-Based Planning and Reasoning. Discussing frameworks that enhance task representation, autonomous planning, and real-time task execution.
3. Human-Robot Interaction (HRI) and Collaborative Robotics. Presenting approaches to improve adaptive interactions and ensure safe, effective human-robot collaboration.
4. Ontologies in Industry 4.0, 5.0, and Smart Manufacturing. Examining applications of ontology- driven frameworks in advanced manufacturing and cyber-physical systems.
5. Ontology and Standardization. Proposing common ontological frameworks to enhance cross- domain and cross-system interoperability.
6. Ontology-Driven Machine Learning and AI. Demonstrating how structured knowledge improves machine learning and AI in robotic systems.
7. Sustainability in Robotics. Addressing ontology-based solutions to optimize resource use, support circular economy models, and enable sustainable robotics.
8. Agile Robotic System Design. Developing modular, reconfigurable robotic systems that adapt to dynamic environments.
9. Ontology for Sensor Fusion and Perception. Using ontologies to integrate multimodal sensor data for enhanced context awareness.
10. Merging Ontological Approaches. Discussing techniques for aligning and integrating diverse ontological frameworks to improve semantic coherence.

Organizers

• Prof. Maki K. Habib, Professor of Robotics and Mechatronics. The American University in Cairo, Egypt. Email: maki@aucegypt.edu
• Prof. Paulo Jorge Sequeira Gonçalves, Professor, Instituto Politécnico de Castelo Branco, Portugal. Email: paulo.goncalves@ipcb.pt

#3 Neuro-Symbolic AI: Empowering Trustworthy and Embodied Robotic Autonomous Life Cycle Manufacturing (Code: w8ghk)

Code: w8ghk

The primary objective of this special session is to explore the integration of neuro-symbolic AI techniques in the context of robotic autonomous life cycle manufacturing, including design, processing, assembly, disassembly, end-of-life treatment with a focus on enhancing autonomy, agility, trustworthiness, and embodiment. Specifically, the session aims to:

1. Foster Cross-Disciplinary Collaboration: Bring together researchers from AI, robotics, sustainability, and industrial engineering to discuss the latest advancements and challenges in neuro-symbolic AI for robotic life cycle manufacturing.
2. Highlight Innovations in Neuro-Symbolic AI: Showcase cutting-edge research that leverages the strengths of both neural networks and symbolic reasoning to address the complexities of autonomous life cycle manufacturing.
3. Address Trustworthiness and Embodiment: Discuss methodologies and frameworks that ensure the secure and reliable operation of robotic systems in processing, assembly, disassembly, end-of-life treatment tasks, particularly in unstructured, dynamic and adversarial environments.
4. Identify Future Research Directions: Provide a platform for identifying gaps in current research and outlining potential future directions for neuro-symbolic AI in robotic life cycle manufacturing.

The integration of neuro-symbolic AI with embodied robotics presents a transformative opportunity for the field of autonomous life cycle manufacturing. Traditional AI approaches often struggle with the complex, real-world environments encountered in processing, assembly, disassembly, end-of-life treatment tasks, which require both high-level reasoning and precise physical interaction. Neuro-symbolic AI, which combines thestrengths of neural networks (for learning and adaptability) with symbolic reasoning (for interpretability and robustness), offers a promising solution to these challenges:

1. Complexity of Lifecycle Manufacturing Tasks: For example, disassembly processes often involve intricate sequences of actions, requiring both high-level reasoning (e.g., planning) and low-level perception (e.g., object recognition). Neuro-symbolic AI, which combines neural networks with symbolic reasoning, is well-suited to handle such hybrid tasks.
2. Need for Trustworthiness: In industrial settings, the reliability and trustworthiness of robotic systems are paramount. Neuro-symbolic AI can enhance the transparency and interpretability of decision-making processes, thereby increasing trust in autonomous systems.
3. Embodiment Concerns: Embodiment enables robots to interact with their environment, performing tasks with precision and adaptability. Neuro-symbolic AI, by incorporating robust symbolic reasoning, tactile sensors, vision systems, and advanced technologies, offers solutions for high-precision disassembly tasks.
4. Sustainability and Circular Economy: The increasing emphasis on sustainability and the circular economy necessitates efficient and automated disassembly, end- of-life treatment processes. Neuro-symbolic AI can optimize these processes by enabling robots to make intelligent decisions based on both learned patterns and explicit rules.

The proposed special session is expected to yield several key outcomes:

1. State-of-the-Art Insights: Participants will gain a comprehensive understanding of the current state of neuro-symbolic AI in robotic life cycle manufacturing, including both theoretical advancements and practical applications.
2. Innovative Solutions: The session will highlight novel approaches and solutions that address the challenges of autonomy, agility, trustworthiness, and embodiment in autonomous life cycle manufacturing systems.
3. Collaborative Opportunities: Researchers from diverse fields will have the opportunity to form new collaborations, leading to interdisciplinary research projects that push the boundaries of neuro-symbolic AI in robotics.
4. Roadmap for Future Research: The session will produce a roadmap for future research, identifying critical areas that require further exploration and development to fully realize the potential of neuro-symbolic AI in robotic life cycle manufacturing.
5. Publication Impact: The session will result in high-quality publications in IEEE CASE proceedings, contributing to the academic literature and advancing the field of neuro-symbolic AI in robotics.

Sub-topics

1. Neuro-Symbolic AI Frameworks for Robotic Autonomous Life Cycle Manufacturing
2. Vision Perception and Tactile Sensing in Processing, Assembly, Disassembly, End-of-life Treatment
3. Planning and Sequencing in Autonomous Processing, Assembly, Disassembly, End-of- life Treatment
4. Human-Robot Collaboration in Processing, Assembly, Disassembly, End-of-life Treatment Tasks
5. Trustworthiness and Transparency in Neuro-Symbolic Systems
6. Neuro-symbolic AI in Circular Economy Initiatives
7. Embedding knowledge into Neural Networks for efficient learning
8. Interdisciplinary Approaches to Autonomous Processing, Assembly, Disassembly, End- of-life Treatment
9. Future Directions and Research Gaps in Neuro-Symbolic AI and Embodied Robotics

Organizers

• Prof. Chen Ming, Professor, Shanghai Jiao Tong University, China. E-mail: mingchen@sjtu.edu.cn
• Dr. Wang Zhigang, Research Scientist, Intel Labs China. E-mail: zhi.gang.wang@intel.com
• Dr Chao Liu, Lecturer, College of Engineering and Physical Sciences, Aston University, UK. E-mail: c.liu16@aston.ac.uk
• Prof. Jef R. Peeters, Associate Professor, KU Leuven, Belgium. E-mail: jef.peeters@kuleuven.be
• Prof. Guo Weizhong, Professor, Shanghai Jiao Tong University, China. E-mail: wzguo@sjtu.edu.cn
• Prof. Jiang Zhigang, Professor, Wuhan University of Science and Technology, China. E-mail: jzg100@163.com
• Prof. Zhang Hongshen, Professor, Kunming University of Science and Technology, China. E-mail: hongshen@kust.edu.cn
• Prof. Cai Yinghao, Professor, Institute of Automation, Chinese Academy of Sciences, E-mail: yinghao.cai@ia.ac.cn
• Prof. Zhou Chuangchuang, Professor, Zhongyu Intelligent Laboratory, Henan Academy of Sciences, E-mail: chuangchuang.zhou@hnas.ac.cn

#4 Human-Robot Collaboration for Futuristic Human-Centric Smart Manufacturing (Code: z8xe4)

Code: z8xe4

In line with the human-centric concerns of Industry 5.0, modern factories are striving for an ever-higher degree of flexible and resilient production, as conventional automation approach has reached its bottleneck considering mass personalization with increasing complicatedness and complexity. To achieve it, human- robot collaboration (HRC) becomes a prevailing strategy, which combines high accuracy, strength, and repeatability of industrial robots with high flexibility and adaptability of human operators to realise optimal overall productivity. Cutting-edge technologies, including robot learning and control, cognitive computing, mixed reality/metaverse, generative AI/large language models, industrial IoT, and advanced data analytics create the potential to bridge the gap of knowledge distilling and information sharing between onsite operators, robots and the manufacturing system with mutual cognitions. Therefore, this special session aims to bring together specialists in different fields of manufacturing systems, robotics, artificial intelligence, and other engineering domains to address the foreseeable HRC-empowered human-centric smart manufacturing paradigm characterized with high-level teamwork skills.

Sub-topics

This special session aims to present the state-of-the-art, informatics-based approaches, tools, systems, and cases to enable the readiness and realization of HRC for futuristic human-centric smart manufacturing. To contribute to those areas, this special session includes the following topics, but are not limited to:

1. Human-robot collaborative dis-/assembly
2. Predictable human-robot-task execution loop
3. Cognitive human-robot collaboration systems
4. MR/Metaverse-assisted human-robot interaction
5. Robotic skill learning from human demonstration
6. Self-organizing multiple human-robot collaboration
7. Intuitive safety concerns in human-robot interaction
8. Multi-modal Intelligence for human-robot collaboration
9. Large foundation models for human-robot collaboration
10. Human-robot collaboration cases, systems, and implementations in manufacturing

Organizers

• Dr. Pai Zheng, Associate Professor, The Hong Kong Polytechnic University, E-mail: pai.zheng@polyu.edu.hk, Phone: +[852] – [27665633]
• Dr. Tao Peng, Associate Professor, Zhejiang University, E-mail: tao_peng@zju.edu.cn, Phone: +[86] – [0571-87951145]
• Dr. Xi Gu, Associate Professor, Rutgers University, E-mail: xi.gu@rutgers.edu
• Dr. Yongjing Wang, Associate Professor, University of Birmingham, E-mail: y.wang@bham.ac.uk
• Dr. Hyun-Jung Kim, Associate Professor, Korea Advanced Institute of Science and Technology, E-mail: hyunjungkim@kaist.ac.kr
• Dr. Yunbo Zhang, Assistant Professor, Rochester Institute of Technology, E-mail: ywzeie@rit.edu
• Dr. Jinsong Bao, Professor, Donghua University, E-mail: bao@dhu.edu.cn, Phone: +[86] – [021-67792583]
• Dr. George Huang, Professor, The Hong Kong Polytechnic University, E-mail: gqhuang@hku.hk, Phone: +[852] – [2766 6586]
• Dr. Lihui Wang, Professor, KTH Royal Institute of Technology, E-mail: lihuiw@kth.se, Phone: +[46] – [8-790-83-05]
• Dr. Duc Truong Pham, Professor, University of Birmingham, E-mail: d.t.pham@bham.ac.uk, Phone: +[44] – [(0)29-2087-4696]

#5 Intelligent Assembly and Assembly Quality Control (Code: s47m7)

Code: s47m7

Assembly technology is a critical component in the manufacturing process of complex products such as airplanes, high-speed trains, ships, automobiles, and machine tools, and it has emerged as one of the major bottlenecks affecting the performance and development cycles of complex products. With the extreme development trends in the operating conditions of complex products, influenced by factors such as complex loads, discontinuous structures, and uneven assembly processes, the role of the assembly stage in ensuring the service performance and reliability of complex products is becoming increasingly prominent. With the advancement and increasing industrial applications of robot, artificial intelligence, machine learning and advanced sensing technologies, new theoretical approaches and technical means have been provided to address the challenges faced in the assembly process of complex products.

In this special session, the organizers focus on intelligent assembly and assembly quality control from an industrial automation perspective, and collect the latest researches and achievements with innovative machine learning, AI, and robot technologies for assembly. By sharing the state-of-the-art intelligent assembly in this session, it is believed that the advanced assembly methods or technical means powered by innovative machine learning, AI, and robot technologies can promote the development of assembly process and assembly quality control both at academical and industrial levels.

Sub-topics

1. Human-machine Collaborative Assembly
2. Data-Driven Methods for Assembly Quality Control
3. Optimization of Assembly Process
4. Automated Assembly
5. Digital Assembly
6. Virtual Reality and Augment Reality for Assembly
7. In-line Detection and In-situ Control of Assembly
8. Digital Twin for Assembly
9. Other Automated Assembly Technologies, Machine Learning Methods, and Intelligence Technologies for Assembly and Assembly Quality Control

Organizers

• Changhui Liu, Associate Professor, Tongji Universiy. E-mail: liuchanghui@tongji.edu.cn
• Juan Du, Assistant Professor, The Hong Kong University of Science and Technology (Guangzhou). E-mail: juandu@ust.hk

#6 Advancements in Modeling, Scheduling, and Control for Robotized Manufacturing Systems (Code: xe2vb)

Code: xe2vb

The rapid advancement of automation and robotics in manufacturing has created a growing demand for efficient and intelligent solutions in robotized manufacturing systems, including cluster tools, robotic cells, and hoist and crane scheduling. These systems are essential across various industries, such as semiconductor and display manufacturing, automotive production, and steel processing, where precision, efficiency, and adaptability are crucial. This proposed special session aims to highlight the latest developments in modeling, scheduling, and control for these systems, addressing their unique challenges and presenting innovative strategies to improve overall performance.

Sub-topics

1. Modeling techniques for robotized manufacturing systems
2. Advanced scheduling algorithms for cluster tools
3. Integrated control strategies for multi-robot systems
4. Dynamic and real-time scheduling for flexible manufacturing systems
5. Energy-efficient scheduling and control
6. Multi-criteria optimization in robotic cell scheduling
7. Application of AI in scheduling and control
8. Crane scheduling and optimization
9. Optimization of material handling systems

Organizers

• Hyun-Jung Kim, Associate Professor, Korea Advanced Institute of Science and Technology. E-mail: hyunjungkim@kaist.ac.kr
• Tae-Sun Yu, Associate Professor, Pukyong National University. E-mail: tsyu@pknu.ac.kr
• Yan Qiao, Associate Professor, Macau University of Science and Technology. E-mail: yqiao@must.edu.mo

#7 Cognitive Manufacturing Systems (Code: v5374)

Code: v5374

Manufacturing industries worldwide are undergoing a transition from technology-centric to value-centric development. Intelligent technologies are being incorporated to create and improve products and services with existing and future ICT infrastructure, including automation, computation, informatization, and digitalization. With advancement in the development of data-rich and knowledge-intensive systems, technologies should actively center on the next generation of artificial intelligence. Cognitive manufacturing systems place an emphasis on the adoption of advanced concepts, i.e., self-learning, self-optimization or referred to as Self-X, thereby endowing manufacturing systems with the ability to perceive, comprehend, and learn autonomously. Cognitive manufacturing facilitates a more profound understanding of market demands, optimization of production processes, and the innovation of products and services. This, in turn, propels the direction towards value-driven life cycle sustainability. Cognitive intelligent technologies hold significant potential in various aspects, including human-machine/robot collaboration, human-system interactive design, dynamic optimization, predictive maintenance, and collaborative decision-making However, the profitable and secure implementation of cognitive manufacturing systems remains uncertain. Additionally, fundamental challenges persist, such as the alignment of data and knowledge architecture across multiple processes, entities, and stages. Hence, this special session aims to bring together specialists in smart/intelligent/advanced manufacturing, information and communication, artificial intelligence, cognitive computing, and other science and engineering domains to address these pressing yet long-standing industrial needs.

Sub-topics

This special session aims to present the state-of-the-art theories, methods, tools, systems, and applications to discuss the challenges and future in future cognitive manufacturing. To contribute to those areas, this special session includes the following topics, but are not limited to:

1. Advances in cognitive intelligence theories
2. Fundamentals in cognitive manufacturing systems
3. Cognitive computing for manufacturing
4. Knowledge representation for cognitive intelligence
5. Knowledge graph and graph-based manufacturing decision-making
6. Large language models in manufacturing systems
7. Large time-series model in manufacturing systems
8. Cognition in innovative product/system design
9. Cognitive digital twin for manufacturing systems
10. Cognitive human-robot collaboration systems
11. Cognitive prognostics and health management
12. Self-organizing manufacturing networks/systems
13. Self-anomaly-detection in manufacturing processes
14. Self-coordinating supply chains/networks
15. Self-diagnosis of intelligent machines/products
16. Cases, applications, and implementations in cognitive manufacturing systems

Organizers

• Dr. Tao Peng, Associate Professor, Zhejiang University. E-mail: tao_peng@zju.edu.cn, Phone: +86 – 0571 87079137
• Dr. Pai Zheng, Associate Professor, The Hong Kong Polytechnic University. E-mail: pai.zheng@polyu.edu.hk, Phone: +852 – 2766 5633
• Dr. Xinyu Li, Research Professor, Donghua University, E-mail: lixinyu@dhu.edu.cn, Phone: +86 – 15000258061
• Dr. Feng Ju, Associate Professor, Arizona State University. E-mail: fengju@asu.edu, Phone: +1 – 6085562998
• Dr. Juan Du, Assistant Professor, The Hong Kong University of Science and Technology (Guangzhou), E-mail: juandu@ust.hk, Phone: +86 – 020 88335989
• Dr. Lihui Wang, Professor, KTH Royal Institute of Technology. E-mail: lihuiw@kth.se, Phone: +46 – 8 790 8305
• Dr. George Huang, Professor, The Hong Kong Polytechnic University. E-mail: gq.huang@polyu.edu.hk, Phone: +852 – 2766 6586
• Dr. Weiming Shen, Professor, Huazhong University of Science and Technology. E-mail: wshen@ieee, Phone: +86 – 027-87543770

#8 Emerging Automation Technologies in Core Engineering Systems (Code: 4qca6)

Code: 4qca6

This special session focuses on the integration of automation technologies in core engineering domains such as civil, mechanical, manufacturing, aerospace, and automobile engineering. It aims to explore innovations that enhance efficiency, safety, sustainability, and productivity across these disciplines.

Automation has become a cornerstone of modern engineering systems, driving advancements in design, analysis, manufacturing, and operations. The session addresses challenges and opportunities associated with implementing automation technologies in traditional engineering fields, fostering interdisciplinary research and applications.

Key objectives include:

• Highlighting the latest automation techniques and tools in core engineering sectors.
• Showcasing case studies of successful implementation in real-world projects.
• Exploring sustainable and cost-effective automation solutions.
• Encouraging collaboration among researchers and industry professionals from diverse engineering backgrounds.

Expected outcomes of the session include insights into emerging trends, practical strategies for adopting automation technologies, and a platform for researchers to share their findings.

Sub-topics

1. Automation in Construction and Infrastructure Management
2. Smart Manufacturing Systems and Industry 4.0
3. Intelligent Maintenance Systems for Aerospace and Automotive Applications
4. Robotics in Heavy Machinery and Industrial Equipment
5. Digital Twin Applications in Core Engineering Fields
6. AI and Machine Learning in Structural Health Monitoring
7. Energy Optimization in Automated Engineering Systems
8. Safety and Reliability in Automated Systems for Core Industries

Organizers

• Dr. Ritesh Bhat, Professor & Head of Department, Mechatronics Engineering, Rajalakshmi Engineering College, Thandalam, Kancheepuram, Tamil Nadu, India. E-mail: riteshbhat.rb@rajalakshmi.edu.in
• Dr. Shilpa Suresh, Department of Mechatronics Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India. E-mail: shilpa.suresh@manipal.edu
• Dr. Shilpa Gite, Symbiosis Centre for Applied AI, Symbiosis International, Pune, Maharashtra, India. Email: shilpa.gite@sitpune.edu.in

#9 Advances in AI-Driven Scheduling and Optimization for Autonomous Manufacturing Systems (Code: g4r1f)

Code: g4r1f

This special session explores advancements in AI, optimization, and autonomous decision-making for scheduling in modern manufacturing systems. It addresses challenges in dynamic and uncertain environments, focusing on AI-driven scheduling, machine learning, real-time decision-making, and hybrid optimization techniques. The session invites researchers and practitioners to present innovative approaches, algorithms, and frameworks for tackling complex scheduling challenges in dynamic and uncertain production settings, including working around human teammates. Topics of interest include, but are not limited to, AI-driven scheduling, machine learning applications, human-centric production scheduling and optimisation, autonomous systems for real-time decision-making, optimization of production workflows, and hybrid methodologies combining heuristic and metaheuristic techniques.

Sub-topics

1. AI-Driven Scheduling Algorithms
2. Machine Learning in Production Planning
3. Real-Time Decision-Making Systems
4. Advanced Optimization Techniques
5. Digital Twins for Scheduling
6. Ergonomics Optimized Scheduling
7. Human-Machine Collaborative Scheduling
8. Human-Centric Production Optimization
9. Reinforcement Learning for Scheduling
10. Human-in-the-Loop Scheduling
11. Human-Robot Task Allocation and Scheduling
12. (Continuous) Dynamic Production Scheduling
13. Knowledge Representation and Learning in Scheduling
14. Mixed-AI Techniques (heuristic, metaheuristic, symbolic AI, RL, GPT etc.) for Scheduling
15. Advanced Project Scheduling
16. Case Studies and Applications

Organizers

• Hyun-Jung Kim, Associate Professor, Korea Advanced Institute of Science and Technology, E-mail: hyunjungkim@kaist.ac.kr
• Yuqian Lu, Senior Lecturer, The University of Auckland E-mail: yuqian.lu@auckland.ac.nz
• Weiming Shen, Professor, Huazhong University of Science and Technology, Email: wshen@ieee.org
• Xinyu Li, Professor, Huazhong University of Science and Technology, Email: lixinyu@hust.edu.cn

#10 3D Point Cloud Processing and Spatio-Temporal Modeling (Code: cmhjr)

Code: cmhjr

The processing and modeling of 3D point clouds and spatio-temporal data are critical in various applications, such as advanced manufacturing, medical imaging, robotics, and environmental monitoring. As 3D sensing technologies, such as LiDAR, structured light scanners, and depth cameras, continue to evolve, the availability of large-scale, high-resolution 3D spatio-temporal datasets has significantly increased. However, effectively extracting meaningful information from these datasets and modeling their dynamic, spatio-temporal characteristics remain significant challenges due to their high dimensionality, irregularity, and temporal variability. Rapid advancements in AI, machine learning, and geometric processing algorithms offer novel theoretical approaches and technical solutions for tackling these challenges.

In this special session, the organizers aim to highlight the latest research and innovations in 3D point cloud processing and spatio-temporal modeling. This session seeks to bridge the gap between academic research and industrial applications by showcasing solutions that enhance the efficiency, accuracy, and scalability of spatio-temporal data processing and modeling techniques.

Sub-topics

1. Geometric Deep Learning for 3D Point Clouds
2. Spatio-Temporal and High-Dimensional Data Modeling
3. Multi-Sensor Fusion for 3D and Temporal Data Integration
4. Advanced (Multivariate and 3D) Time Series Analysis and Forecasting
5. Geometric Shape Descriptor Models and Applications
6. Scalable Processing of Large-Scale Point Clouds
7. Digital Twin Development Using Spatio-Temporal Data
8. Real-World Applications of Spatio-Temporal Data Analysis
9. Enhanced Visualization Techniques for 3D and Spatio-Temporal Data

Organizers

• Dr. Yinan Wang, Assistant Professor, Rensselaer Polytechnic Institute, E-mail: wangy88@rpi.edu, Phone: +1-518-276-6837
• Dr. Michael Biehler, Assistant Professor, University of Wisconsin – Madison, Email: mbiehler@wisc.edu, Phone: +1-608-262-2686

#11 Novel Planning and Control Approaches for Semiconductor Manufacturing (Code: x45wg)

Code: x45wg

Semiconductor manufacturing is one of the most complex manufacturing processes. Due to the increased use of new technologies, the global economy is significantly impacted by the semiconductor industry. Competition in the semiconductor sector is fierce. As a result, the semiconductor industry needs to continually reinvent itself and be resourceful at all decision levels. Efficient design, analysis, and operation of semiconductor wafer manufacturing facilities and corresponding supply chains are essential.

• Various factors can disrupt certain chains of the production networks. For example, the global supply chain has been disrupted by the Covid-19 pandemic which spreads around the world, resulting, for instance, in chip shortage. All this has an impact on the production (product quality, supply, etc.). Subsequently, effective decisions must be taken to conduct production operations at different levels of the supply chain links while addressing disruptions.
• To ensure reliable results and reduced operational costs, highly automated manufacturing systems are used to carry out operations and make millions of decisions per day. This requires planning and scheduling methods in manufacturing execution systems (MES) and in logistics/supply chain management tools to support their automated operation. As new trends such as sustainable production, cloud computing, and Industry 4.0 emerge, they also need to be addressed.
• The development and application of planning and scheduling methods for these high-cost systems and supply-chains are critical elements in improving their operations. The purpose of the proposed session is to highlight cutting-edge research on semiconductor manufacturing planning, scheduling, and supply- chain management.

Sub-topics

1. Network planning in semiconductor supply chains
2. Demand planning for the semiconductor supply chains
3. Capacity planning and master planning for semiconductor supply chains
4. Operational planning approaches in the semiconductor domain
5. Reference models for semiconductor supply chain planning and control
6. Models for the interactions between production and development
7. Novel scheduling approaches for semiconductor manufacturing
8. Flexibility and robustness in planning and scheduling
9. Digital twins for semiconductor manufacturing
10. Machine learning and other artificial intelligence approaches and tools such as ontologies for planning and scheduling.

Organizers

• Lars Mönch, Prof., University of Hagen, Germany. E-mail: Lars.Moench@fernuni-hagen.de
• Claude Yugma, Prof., Mines de Saint-Etienne, France. E-mail: Yugma@emse.fr

#12 Collaborative Robot-Empowered Human-Centered and Trustworthy Manufacturing in Industry 5.0 Contexts (Code: psgj8)

Code: psgj8

Collaborative robots play a significant role in the Industry 5.0 revolution. They can assist in human-centered manufacturing tasks at both informational and physical levels. Meanwhile, the role of humans is not weakened as humans have unmatched problem-solving skills and are able to provide creative and unique solutions compared to robots. Therefore, the collaboration between humans and robots for human-centered manufacturing will create new opportunities and benefits to optimize production, boost efficiency, improve task quality, enhance safety, and increase worker flexibility for industry sectors. Industry 5.0 upgrades the interconnection and nature of human-robot partnerships with emphases on humans’ well-being, facilitating talents and diversity, and expanding societal value. In this context, human-robot collaboration will highly enable human expertise and creativity, transparentizing manufacturing processes, advancing manufacturing trustworthiness and reliability, and democratizing manufacturing industries with high productivity and low cost. In both academia and industries, new interdependent and cross- disciplinary research issues, opportunities, and challenges arise and need to be addressed to make human-robot teams more productive and ergonomic for human-centered and trustworthy manufacturing in Industry 5.0 Contexts. This special session aims to bring researchers, scientists, engineers, and managers engaged in frontier research and technologies of robotics, automation, human factors, manufacturing, artificial intelligence, and cyber-physical systems to investigate and solve different open questions in the field of human-centered and trustworthy manufacturing. Prospective authors are invited to share their state-of-the-art research findings to address the gaps in this area.

Sub-topics

1. Characterization and modeling of human factors in human-robot collaborative manufacturing
2. Modeling of manufacturing reliability in human-robot collaborative contexts
3. Task scheduling optimization in human-centered collaborative manufacturing
4. Learning from demonstrations for intelligent and trustworthy manufacturing
5. AI-based approaches for human-robot trustworthy manufacturing
6. Multi-human-robot collaboration for human-centered manufacturing tasks
7. Human intention reasoning/prediction in human-centered manufacturing
8. Ethical issues in human-robot collaboration
9. Smart sensory systems for human-robot collaborative manufacturing
10. Design and development of novel collaborative robots
11. Advanced robot motion/action planning algorithms for human-robot collaboration
12. Virtual reality/augmented reality/extended reality for human-robot collaborative manufacturing
13. Collaborative manufacturing task quality modeling and evaluation

Organizers

• [Weitian Wang], [Associate Professor, IEEE Senior Member] [School of Computing, Montclair State University, USA] E-mail: [wangw@montclair.edu] Phone: [+1-9736555269]
• [Mengchu Zhou], [Dist. Professor, IEEE Fellow] [Dept. of Electrical and Computer Engineering, New Jersey Institute of Technology, USA] E-mail: [zhou@njit.edu] Phone: [+1-9735966282]
• [Xiwang Guo], [Associate Professor, IEEE Senior Member] [Dept. of Information and Control Engineering, Liaoning Petrochemical University, China] E-mail: [x.w.guo@163.com] Phone: [+1-6266888772]
• [Yan Qiao], [Associate Professor, IEEE Senior Member] [Institute of Systems Engineering, Macau University of Science and Technology, China] E-mail: [yqiao@must.edu.mo] Phone: [+853-65643512]

#13 Leveraging Machine Learning and Big Data Techniques for Reliable and Secure Automation in Healthcare Practices (Code: 5ydsu)

Code: 5ydsu

Objectives: This special session seeks to examine how machine learning and big data analytics can transform a broad range of healthcare practices, from diagnostics and treatment planning to public health and system management. By presenting recent breakthroughs in machine learning and big data techniques for image analysis, and real‐time data processing, the session aims to highlight how data‐driven insights enhance clinical decision‐making, optimize resource allocation, and support personalized patient care. This session will also explore methodological and infrastructural considerations, such as data integration, regulatory compliance, and adoption strategies across diverse healthcare environments, that enable the sustainable incorporation of advanced analytics into day‐to‐day operations.

Justifications: Recent years have seen a surge in healthcare data collection, encompassing electronic and mobile health records, imaging repositories, wearables, and genomic data. Traditional rule‐based solutions often struggle with the complexity and scale of this information, whereas machine learning and big data tools can detect intricate patterns and generate actionable predictions. Meanwhile these techniques raise pressing questions regarding privacy, data governance, algorithmic transparency, and the potential for bias. A dedicated forum is therefore essential for exchanging insights on how to deploy these tools ethically, effectively, and in ways that truly benefit patients. By gathering professionals from clinical, academic, and technical backgrounds, the session will create a space to compare experiences, share innovative research, and devise strategies to overcome implementation hurdles.

Projected Outcomes: Attendees will gain a deeper understanding of how cutting‐edge machine learning and big data approaches are reshaping multiple facets of healthcare, including clinical operations, disease surveillance, and personalized medicine. The session will offer perspectives on interdisciplinary collaborations that can break down silos between healthcare providers, data scientists, and policymakers.

Participants will leave with a stronger grasp of the opportunities and limitations that accompany the adoption of data‐centric tools, as well as a clearer sense of next steps for advancing research, fostering robust industry partnerships, and ensuring equitable access to these transformative technologies.

Sub-topics

1. Data Fusion for Healthcare Automation
2. Robust and Resilient Machine Learning Models
3. Automated Diagnosis and Decision Support Systems
4. AI‐Driven Predictive Analytics
5. Privacy‐Preserving Machine Learning in Healthcare
6. Reliability in AI‐Driven Medical Systems
7. Efficient Exchange of Medical Data
8. Secure Data Sharing and Interoperability in Healthcare

Organizers

• Yuxin Wen, Assistant Professor, Chapman University. E‐mail: yuwen@chapman.edu
• Jiajing Huang, Assistant Professor, Kennesaw State University E‐mail: jhuang24@kennesaw.edu
• Feng Liu, Assistant Professor, Stevens Institute of Technology, E‐mail: fliu22@stevens.edu
• Jia Chen, Assistant Professor of Teaching, University of California, Riverside, E‐mail: jia.chen@ucr.edu
• Chao Wang, Postdoctoral Research Fellow, University of Maryland, E‐mail: cwang@umces.edu
• Xin Shen, PhD Student, University of California, Riverside, E‐mail: xshen049@ucr.edu

#14 System Optimization and Production Control of Intelligent Manufacturing (Code: 4uag3)

Code: 4uag3

The era of intelligent manufacturing is characterized by the features of connectivity, optimization, transparency, predictive, and agility. The rapid advancement of artificial intelligence (AI) technologies, particularly large language models (LLMs), has ushered in a new era of innovation and transformation across various industries. The integration of these techniques into manufacturing systems is revolutionizing the way these systems are built, managed and controlled, offering unprecedented opportunities for enhancing efficiency, quick response, self-organization, servitization, agile scheduling, predictive maintenance, quality control, and decision-making processes. However, they also introduce significant challenges, including the effective data acquisition and utilization, system complexity, and the need for advanced human-machine collaboration. This special session aims to explore the advancements in system optimization and production control of intelligent manufacturing by researchers and practitioners from different disciplines. This session invites contributions that explore these challenges and propose innovative solutions to advance the intelligence and efficiency of manufacturing systems.

Sub-topics

1. Predictive analytics to forecast product demands
2. Self-organization production systems planning and design
3. Integration of IoT and Cyber-Physical Systems for intelligent manufacturing
4. Advanced optimization algorithms for manufacturing processes
5. Integration of AI approaches in production planning and control
6. Real-time monitoring and predictive maintenance strategies
7. Digital twin technology for manufacturing system modelling, simulation, and optimization
8. Human-machine collaboration in manufacturing process
9. Supply chain optimization in intelligent manufacturing environments
10. New methods or tools for Lean management in intelligent manufacturing
11. Manufacturing management information systems and technology
12. AI-enhanced quality control, system optimization, and production control
13. Sustainable and energy efficiency for intelligent manufacturing
14. Optimization and production control of servitized manufacturing system

Organizers

• [Zhibin Jiang], [Professor], [Antai College of Economics and Management, Shanghai Jiao Tong University]. E-mail: [zbjiang@sjtu.edu.cn]
• [Liping Zhou], [Associate Professor], [Institute of Intelligent Manufacturing and Service Management, Shanghai Jiao Tong University]. E-mail: [zhoulp@sjtu.edu.cn]

#15 Analysis and control in the data-driven age, emerging methodologies and applications (Code: vb4c4)

Code: vb4c4

The acquisition, processing, and analysis of “big data” have become pivotal topics in science and engineering, driving renewed interest in data-driven techniques for automatic control. Traditional model-based control strategies often struggle with the inherent complexity of modern process systems. These challenges arise from the difficulty of establishing models that are both accurate and efficient for control purposes. First- principles models, while theoretically robust, are often highly intricate, making their derivation and reduction a daunting task. Dynamic models, typically obtained through system identification based on intentional process perturbations, demand significant time and human intervention to ensure their quality. Examples of such complex systems include oil refineries, chemical plants, pharmaceutical and fermentation processes, and semiconductor manufacturing systems. The macroscopic and phenomenological dynamics of these systems are governed by the transport of mass, energy, and momentum under constraints imposed by thermodynamic laws.

This context underscores the necessity to address two critical methodological aspects for the advanced control theory in the data-driven age:

1. Formal Reachability Analysis: A formal characterization of the reachability properties of plant models is an integral part of system design and validation, particularly for safety-critical applications. This enables the assurance of system safety by exhaustively exploring all potential outcomes. This aspect is essential for industries where operational safety is paramount, including chemical plants, autonomous vehicles, and aerospace systems.
2. Development of Efficient Data-Driven Control Strategies: The growing capability to collect high- volume and the high-variety datasets from process operations have catalyzed the adoption of machine learning techniques for process data analytics. Building on reachability analysis, data-driven control strategies can leverage these datasets to address non-convex control problems efficiently. By integrating ideas from control systems theory and artificial intelligence, these strategies aim to unlock novel control architectures that can handle the complexity of dynamic systems.

The research objectives consist in: exploring and developing advanced methodologies for data-driven analysis and control of dynamic systems; establishing computationally efficient tools for addressing non-convex control problems in complex environments such as industrial plants and smart cities; integrating control theory and machine learning approaches to design innovative and robust control architectures; enhancing safety and operational efficiency in critical systems through advanced reachability analysis.

In addition, the increasing complexity of modern process systems necessitates a shift from traditional model- based approaches to data-driven methodologies. These methods not only offer scalability and adaptability

but also capitalize on the growing availability of large datasets from industrial operations. Moreover, the integration of artificial intelligence techniques introduces a new dimension to control theory, enabling more accurate, efficient, and resilient solutions for managing complex, high-dimensional systems. By addressing non-convex control problems, these advancements can unlock significant operational improvements in industries such as manufacturing, energy, and urban infrastructure.

Possible Outcomes:

1. Innovative Control Architectures: Development of new control frameworks that merge insights from control systems theory and artificial intelligence to handle complex nonlinear dynamics.
2. Efficient Computational Tools: Creation of algorithms and numerical methods to perform reachability analysis and solve non-convex control problems efficiently.
3. Improved System Safety: Enhanced ability to analyze and guarantee the safety of critical systems through formal reachability studies.
4. Enhanced Operational Efficiency: Optimization of process performance in industrial plants and smart cities by leveraging data-driven insights.
5. Cross-Disciplinary Collaboration: Foster a deeper synergy between the fields of control systems and artificial intelligence, leading to a broader impact on both theoretical advancements and practical applications.

The special session aims then to provide a platform for discussing recent advancements in data-driven analysis and control of dynamical systems. It seeks to address the pressing challenges posed by complex industrial and urban environments, demonstrating how the fusion of control theory and machine learning can drive innovative solutions. By focusing on computational efficiency and practical applicability, this session aims to set the foundation for the next generation of control strategies, paving the way for safer, smarter, and more efficient systems.

Sub-topics

1. Reachability analysis
2. Constrained control
3. Reinforcement learning
4. Statistical tools
5. Autonomous systems
6. Intelligent transportation systems
7. Smart buildings
8. Safety and security of cyber-physical systems

Organizers

• Domenico Famularo, Associate Professor, DIMES, University of Calabria, Italy. E-mail: domenico.famularo@unical.it
• Giancarlo Fortino, Professor, DIMES, University of Calabria, Italy. E-mail: g.fortino@unical.it
• Vicenç Puig, Professor, Automatic Control Department and Institute of Robotics, Universitad Politecnica de Catalunya, Spain. E-mail: vicenc.puig@upc.edu

• MengChu Zhou, Professor, ECE- New Jersey Institute of Technology, NJ, USA. E-mail: mengchu.zhou@njit.edu

#16 AI for Simulation and Optimization in Automation(Code: yj9w4)

Code: yj9w4

Simulation and optimization is part of a main procedure to design, analyze, and improve the performance in many automation systems, such as smart grids, smart buildings, smart cities, intelligent transportation systems, just to name a few. The recent advances in analytical tools and data acquisition technologies have fundamentally reshaped the business in this field. It has become common practice to build digital twins aside the physical automation system, together with decision making boosted by artificial intelligence (AI). On the one hand, it is the unique structural property in some systems that allows for model reduction and efficient simulation. On the other hand, it is the upcoming artificial general intelligence (AGI) that might provide efficient decision making based on historical data as well as near-human-level reasoning capabilities. While embracing the exiting advances in particular fields of AI, such as AI generated content (AIGC), we see both promising future as well as challenges in applying AI for simulation and optimization in automation.

The focus of this special session is to call for submissions reporting the state of art in this field. In particular, we welcome positioning papers reviewing the challenges and future research directions, as well as case studies to show the applications in specific simulation and optimization problems in automation.

This special session is endorsed by the IEEE RAS TC on Machine Learning for Automation.

Sub-topics

1. AI for simulation
2. AI for optimization
3. Reinforcement learning
4. Energy management
5. Autonomous driving
6. Manufacturing
7. Logistics
8. Smart grid

Organizers

• (Samuel) Qing-Shan Jia, Professor, Tsinghua University, China. E-mail: jiaqs@tsinghua.edu.cn
• Bing Yan, Assistant Professor, Rochester Institute of Technology, USA. E-mail: bxyeee@rit.edu
• Shuo Feng, Associate Professor, Tsinghua University, China. E-mail: fshuo@tsinghua.edu.cn
• Bengt Lennartson, Professor, Chalmers University of Technology, Sweden. E-mail: bengt.lennartson@chalmers.se
• Maria Pia Fanti, Professor, Polytechnic University of Bari, Italy. E-mail: mariapia.fanti@poliba.it

#17 Frontier of Quality and Reliability Engineering for System Intelligence (Code: x1232)

Code: x1232

In the landscape of modern automation, the paradigm of Secured and Trustworthy Automation has become the cornerstone for the successful deployment of various systems. Automation, spanning from collaborative robotics on smart factory to autonomous vehicles on the roads, smart manufacturing setups, and intelligent healthcare delivery mechanisms, is increasingly relying on cutting-edge quality and reliability engineering.

As the systems become more complex and interconnected, the quality and reliability engineering for system intelligence have become of utmost importance. Imperfect system design, biases in data collection, anomaly in manufacturing, variation propagation in multi-stage assembly, and potential vulnerabilities in machine learning algorithms can lead to catastrophic failures. For example, in aircraft manufacturing, a minor glitch in sensor quality could result in serious accidents. In smart city’s transportation automation system, if the reliability of control algorithms is compromised, it could cause safety hazards to human life. These issues highlight the pressing need to delve into the frontier of quality and reliability engineering for System Intelligence. By doing so, we can fortify the integrity of automation systems, making them more secure, trustworthy, and resilient in the face of complex real-world scenarios.

Sub-topics

1. Advanced Quality Assurance Techniques for AI-Driven Intelligent Systems
2. Reliability Modeling and Analysis in Complex Systems
3. Data-Driven Quality Improvement for System Intelligence in Smart Manufacturing
4. Ensuring Reliability in Edge Computing for Intelligent IoT Devices
5. Active Learning and Design of Experiment for Quality Improvement
6. Quality and Reliability Challenges in Machine Learning – Based Predictive Maintenance Systems
7. Predictive Analytics for Quality and Reliabiltiy
8. Reinforcement Learning for System Automation

Organizers

• Dr. Xiaowei Yue, Department of Industrial Engineering, Tsinghua University. Email: yuex@tsinghua.edu.cn
• Dr. Jianguo Wu, Department of Industrial Engineering and Management, Peking University. Email: j.wu@pku.edu.cn
• Dr. Yongxiang Li, Department of Industrial Engineering, Shanghai Jiao Tong University. Email: yongxiangli@sjtu.edu.cn
• Dr. Mohammed Nabhan, Department of Industrial and Systems Engineering, King Fahd University of Petroleum and Minerals. Email: nabhan@kfupm.edu.sa

#18 AI-Driven Intelligent Automation for Resilient Complex Systems (Code: 8e67r)

Code: 8e67r

The rapid integration of artificial intelligence (AI) and Internet of Things (IoT) technologies is revolutionizing automation in complex industrial and service systems, spanning smart manufacturing, healthcare, power grids, and consumer-driven supply chains. IoT platforms enable seamless connectivity among diverse devices, generating vast volumes of heterogeneous data that capture intricate system behaviors in real time. While these data-rich environments offer unprecedented opportunities to enhance system performance, they also pose significant challenges in harnessing data for real-time adaptation, predictive control, and autonomous decision-making. AI has emerged as a cornerstone in addressing these challenges, providing powerful tools for self-learning automation, dynamic process optimization, and anomaly detection, thereby enabling complex systems to operate with greater resilience and efficiency.

This special session focuses on AI-driven intelligent automation as a key enabler for building resilient complex systems that can withstand disruptions and adapt to evolving conditions. Modern industries operate in highly dynamic and interconnected environments, where challenges such as volatile market demands, supply chain disruptions, fluctuating energy conditions, and multi-modal data heterogeneity require adaptive, scalable, and robust automation solutions. Traditional rule-based or static models often struggle to handle the complexity, uncertainty, and interdependencies inherent in these systems. In contrast, AI-driven approaches introduce a paradigm shift—enabling automation systems to continuously learn from data, dynamically adjust to real-world changes, and proactively mitigate risks, thereby enhancing resilience, reliability, and efficiency. This session aims to explore cutting-edge AI techniques that drive automation solutions capable of self- optimization, predictive decision-making, and intelligent adaptation. Contributions will highlight novel methodologies and real-world applications that bridge the gap between theoretical AI advancements and practical system automation in dynamic, high-stakes environments.

Sub-topics

1. Interpretable AI and Machine Learning Technologies
2. Data-Driven System Modeling
3. Knowledge-infused Data Analytical Methods for Quality and Reliability
4. Process Anomaly Detection and Diagnosis
5. Federated Learning for Distributed Automation
6. AI-enabled Operation and Maintenance Optimization
7. Hybrid Offline and Online Learning for Adaptive Scheduling and Control
8. Other Industrial Applications of AI technologies

Organizers

• [Yu An], [Research Fellow], [Department of Industrial Systems Engineering and Management, National University of Singapore]. E-mail: [yu.an@nus.edu.sg]
• [Xi Zhang], [Associate Professor], [Department of Industrial Engineering and Management, Peking University]. E-mail: [xi.zhang@pku.edu.cn]

#19 Intelligent “Modeling, then Optimization” of Automatic Systems (Code: 4uw8e)

Code: 4uw8e

With the advancement of information technology, automatic systems have been implemented across various domains, offering transformative opportunities to revolutionize people’s daily life. However, many real-world analytics challenges in automatic systems entail two pivotal aspects: modeling and optimization. Traditional Artificial Intelligence (AI) tools for these tasks overlook the optimization problem structure when designing the modeling phase, solely focusing on minimizing modeling errors without considering how the modeling results will be utilized in subsequent optimization processes. Furthermore, in the era of big data, automatic systems face complexities that hinder the straightforward application of the “modeling, then optimization” paradigm. These complexities arise from data collected from multiple sensors with diverse modalities, as well as the influence of physical mechanisms and environmental factors on the systems.

This special session aims to delve into the recent advancements in intelligent “modeling, then optimization” of automatic systems. Potential topics include, but are not limited to, interactive learning for modeling and optimization, multi-sensor fusion, physics-informed AI, multi-modal intelligence, reinforcement learning, large models, and theoretical AI analysis, all aimed at enhancing system automation.

Sub-topics

1. Modeling and optimization-integrated tasks for system automation
2. Multi-sensor fusion of process modeling, control, diagnosis, and decision making
3. Physics-informed AI modeling of automatic systems
4. Multi-modal intelligence of automatic systems
5. Optimization and control in improving system automation
6. Application of reinforcement learning in automation
7. Large models for automatic systems
8. Theoretical analysis of AI models for system automation

Organizers

• Di Wang, Associate Professor, Shanghai Jiao Tong University, E-mail: d.wang@sjtu.edu.cn
• Yuanxiang Wang, Associate Professor, Tongji University, E-mail: yuanxiangwang@tongji.edu.cn
• Jingsi Huang, Research Associate, Peking University, E-mail: jingsi.huang@pku.edu.cn
• Minhee Kim, Assistant Professor, University of Florida, E-mail: mkim@ise.ufl.edu

#20 Artificial Intelligence-based models and methods for smart logistics, transportation, manufacturing and healthcare (Code: 8m572)

Code: 8m572

This special session deals with the problem of enhancing the control and management of smart logistics, manufacturing and healthcare systems by using Artificial Intelligence (AI). AI is the current trend in different research fields and its applicability must be further investigated in several engineering sectors to ensure safe and accurate decision processes.

The goal of this session is to present AI-based approaches and models that can improve the efficiency and safety of logistics, intelligent transportations and manufacturing operations and support the human decisions and diagnosis in healthcare.

Smart logistics and Intelligent Transportation Systems involve the optimization of the flow of goods, people and services. It can be enhanced by AI technologies to improve efficiency, reduce costs, and enhance customer satisfaction. Route optimization, supply chain optimization, traffic forecasting, predictive maintenance are some of the issues that can be addressed more efficiently via AI.

AI for smart manufacturing focuses on improving productivity, quality, and flexibility in the production process. The integration of AI is part of the broader concept of Industry 4.0/5.0, where automation, data exchange, and smart systems play a key role. AI can be used for supply chain management, process optimization, predictive quality control, and Digital Twins design.

In healthcare, AI can be used to improve diagnosis, personalize treatments, manage healthcare operations and predict patient health outcomes. In this context, AI-based methods can allow for enhancing medical imaging analysis, clinical decision support, personalized medicine, robotic surgery, virtual healthcare assistants.

Sub-topics

1. Artificial Intelligence Optimization
2. Electric and autonomous fleet management Demand forecasting in transportation
3. Process optimization
4. Predictive maintenance
5. Automation in manufacturing, logistics and Intelligent transportation
6. Digital Twins in manufacturing, logistics and intelligent transportation
7. Medical diagnosis
8. Medical imaging analysis Clinical decision support

Organizers

• Maria Pia Fanti, Polytechnic University of Bari, Italy (email: mariapia.fanti@poliba.it)
• Cristian Mahulea, Universidad de Zaragoza, Spain (email: cmahulea@unizar.es)
• Agostino Marcello Mangini, Polytechnic University of Bari, Italy (email: agostinomarcello.mangini@poliba.it)
• Michele Roccotelli, Polytechnic University of Bari, Italy (email: michele.roccotelli@poliba.it)
• Birgit Vogel-Heuser, Technical University of Munich, Germany (vogel-heuser@tum.de)
• Mengchu Zhou, New Jersey Institute of Technology, USA (email: zhou@njit.edu)

#21 Industrial Robot as a Service (IRaaS) (Code: 46g94)

Code: 46g94

Analogous to service-oriented software architectures, the concept of Robot as a Service (RaaS) represents a paradigm shift in robotics, where the focus changes from explicitly defining how a robot performs a task to merely specifying what it should accomplish. The execution is handled autonomously through AI-driven planning and control, enabling broader accessibility beyond expert users. In an industrial context RaaS holds the potential to provide cost-efficient, flexible, and intuitive automation solutions for companies across various industries and scales. However, current implementations face significant challenges, particularly in terms of user interaction complexity, adaptation to specific use cases, and seamless hardware-software integration.

The continuous advancements in Large Language Models (LLMs) open new possibilities for the configuration and programming of robotic applications, while simulated training environments enable robots to acquire perceptive capabilities before deployment in real-world settings. This special session explores the applicability of these emerging technologies in industrial environments by addressing four key research challenges:

Availability of Digital Models

The effective deployment of robotic systems in industrial settings relies on high-fidelity digital representations of the operational environment to facilitate accurate planning, simulation, and control. However, there remains a significant discrepancy between available digital models and the requirements of AI-driven robotic systems. A major challenge is the integration of heterogeneous data sources, such as CAD files and 3D scans, into a coherent environmental model. This requires advanced data fusion algorithms capable of reconstructing digital twins from fragmented, incomplete and dynamically changing inputs. However, capturing every environmental detail at high resolution is neither feasible nor efficient, as overly complex models increase computational overhead and hinder real-time processing. Instead, modeling strategies must prioritize critical features while maintaining a lightweight and adaptive structure that enables efficient data processing.

Sim-to-Real Gap

The training of AI models for robotic control increasingly relies on simulation-generated datasets. However, the effectiveness of such models in real-world applications is limited by the Sim-to-Real Gap—the discrepancy between simulated environments and real-world conditions. Differences in physical properties, sensor noise, and environmental variations often lead to failures when transitioning from simulation to real world deployment. Addressing this challenge requires domain adaptation techniques, domain randomization, and transfer learning to enhance the generalization capabilities of AI models. Moreover, realistic physics simulation, incorporating material properties, lighting conditions, and sensor imperfections, is essential to improve the fidelity of training environments and ensure robust AI performance across diverse industrial scenarios.

Calibration of AI Models

Beyond achieving high accuracy, AI models must also quantify the confidence in their predictions to ensure reliability in industrial applications. Currently, there is a lack of robust methodologies for accurately assessing AI confidence levels in robotic deployments. Conventional approaches such as Platt or Temperature Scaling often fail to adapt to dynamic environments and sensor variations. A critical research need lies in the development of adaptive confidence calibration techniques, enabling AI models to continuously refine their uncertainty estimations through active learning mechanisms. Furthermore, establishing standardized evaluation metrics is crucial to systematically determine whether models with low confidence require refinement through real-world data collection.

Availability of Real-World Training Data

The effectiveness of AI-driven robotics depends on high-quality training data, yet data acquisition remains challenging due to scarcity, high costs, and lack of diversity across operating conditions. The absence of standardized datasets further limits AI model generalization. While combining real and synthetic data improves robustness, new methods are needed to efficiently capture real-world training data with minimal cost and effort. Research should focus on automated data labeling, self-supervised learning, and adaptive sampling to streamline training data collection.

Sub-topics

1. LLM-based Programming and Configuration of Industrial Robot Applications
2. Automated Environment Mapping and 3D Reconstruction for Robotic Workspaces
3. Bridging the Sim-to-Real Gap in Industrial Robot Applications
4. Synthetic and Self-Supervised Learning for Industrial Robotics
5. Confidence Calibration and Uncertainty Estimation in AI-driven Robotics
6. Advances in Visual Servoing and Zero-shot Learning for Robot Control
7. Flexible and Reconfigurable Robotics for Circular Manufacturing
8. Business Models for IRaaS

Organizers

• Rüdiger Daub, Professor, Technical University of Munich (TUM), Fraunhofer Institute for Casting, Composite and Processing Technology (IGCV). E-mail: ruediger.daub@iwb.tum.de
• Lukas Tanz, Head of Department Assembly Technology and Robotics, Technical University of Munich (TUM),. E-mail: Lukas.tanz@iwb.tum.de

#22 Enhancing Healthcare Delivery through Advanced Simulation Techniques (Code: wp5y4)

Code: wp5y4

In the rapidly evolving field of healthcare, the integration of advanced simulation techniques and artificial intelligence (AI) presents a transformative opportunity to enhance healthcare service delivery and patient outcomes. This special session explores the cutting-edge theoretical foundations and applications of simulation and AI in healthcare environments, emphasizing its role in optimizing processes, reducing costs, and improving care quality. The session will focus on automation in healthcare, where simulation and AI are applied to streamline operations, manage patient flow, and enhance decision-making in various healthcare settings. The session will also highlight innovative methodologies and the latest technological advancements in simulation, providing a platform for researchers and practitioners to share insights. By fostering a deeper understanding of simulation’s potential, this session aims to inspire ongoing research and practical implementations that address critical challenges in healthcare delivery.

Sub-topics

1. Capacity Planning and Resource Allocation
2. Simulation in Emergency Room Management
3. Simulation in Critical Care Delivery
4. Surgical Procedure Optimization
5. Simulation in Operating Rooms
6. Medical Decision Making
7. Data Analytics and Health Informatics
8. Public Health and Health Policy

Organizers

• Feifan Wang, Assistant Professor, Tsinghua University. Email: wangfeifan@tsinghua.edu.cn
• Xiang Zhong, Associate Professor, University of Florida. Email: xiang.zhong@ise.ufl.edu

#23 To Automate or To Augment? Towards Human-Centered Systems Advancing Cognitive and Physical Abilities in Industrial Workplaces (Code: 1s972)

Code: 1s972

The emerging shift towards human-centered automation holds the promise of transforming future work by augmenting humans—both cognitively and physically—rather than replacing them with algorithms and machines. This special session will explore cutting-edge research at the intersection of AI, robotics, human modeling, and spatial computing, aiming to bring this vision to life. Discussions will focus on innovative intelligent systems designed to empower human workers with diverse cognitive and physical abilities to perform tasks that are (a) beyond their current capabilities and (b) infeasible to fully automate. While submissions in all areas related to the scope of this session are welcome, the focus will specifically be on: (a) data-driven perception of cognitive and physical states using multimodal sensor data streams (e.g., RGB-D, gaze, hand/body pose, physiological signals), (b) context-aware interventions through generative AI models informed by real-time cognitive and physical states, and (c) adaptive interactions with virtual (e.g., augmented reality (AR) interfaces) and physical (e.g., collaborative robots) agents aimed at enhancing human abilities, both cognitive and physical. Further motivating this special session is the urgent need to foster an inclusive workforce that addresses skill shortages and the human-technology mismatch in future workplaces.

Sub-topics

1. Human activity and task understanding (e.g., action recognition/anticipation, error detection).
2. Cognitive and emotional state estimation (e.g., workload, stress, frustration).
3. Attention tracking and intent recognition/anticipation.
4. Recent advances in AI that enable context-aware analysis, intervention and multimodal interaction (e.g., for training, feedback).
5. AR guides to facilitate skill acquisition, task performance, human-robot/machine interaction.
6. Seamless human-robot and human-agent collaboration.
7. Wearable technologies for cognitive and physical augmentation.
8. Societal impacts and implications for individuals with cognitive and physical disabilities.

Organizers

• Mohsen Moghaddam, Associate Professor, Georgia Institute of Technology, mohsen.moghaddam@gatech.edu.
• Shakiba Davari, Postdoctoral Researcher, Georgia Institute of Technology, sdn6@gatech.edu.
• Sean Andrist, Principal Researcher, Microsoft Research, sean.andrist@microsoft.com.
• Dan Bohus, Senior Principal Researcher, Microsoft Research, dbohus@microsoft.com.
• Stacy Marsella, Professor, Northeastern University, s.marsella@northeastern.edu.

#24 Advancing Technologies and Sustainability for Production and Service Systems in the AI Age (Code: 4b9ds)

Code: 4b9ds

In the wave of the global manufacturing industry’s accelerated transformation and upgrading, intelligent manufacturing, advanced production systems, and smart logistics, as key fields, are profoundly changing the face of traditional industries and jointly building the cornerstone of a modern industrial system. They are intertwined and develop synergistically, becoming the core drivers for promoting enterprise innovation, enhancing competitiveness, and achieving sustainable development. On one hand, intelligent manufacturing is a new manufacturing model that integrates advanced information technology, automation technology, artificial intelligence, etc., aiming to achieve the intelligence, automation, and flexibility of the production process. By widely deploying sensors in production equipment, real-time production data is collected, and through in-depth mining and analysis using big data analytics, machine learning, and artificial intelligence algorithms, enterprises can achieve precise control and optimization of the production process. On the other hand, smart logistics is a new logistics model that uses technologies such as the Internet of Things, big data, and artificial intelligence to conduct intelligent management and optimization of the logistics process. It covers multiple links such as warehousing, transportation, and distribution, aiming to improve logistics efficiency, reduce logistics costs, and enhance customer service levels. The intelligent warehouse management system can monitor the inventory status in real-time and automatically make inventory warnings and replenishment decisions according to sales data and production plans, effectively improving inventory turnover. In the transportation link, the intelligent transportation management system uses technologies such as satellite positioning and the IoTs to achieve real-time tracking and dispatching of transportation vehicles. By optimizing transportation routes and reasonably arranging vehicle loading, enterprises can reduce transportation costs and improve transportation efficiency. In the distribution link, the intelligent distribution system automatically plans the optimal distribution route according to information such as customers’ locations and order requirements through artificial intelligence algorithms and big data analysis, achieving the rapid and accurate delivery of goods. With the use of new distribution methods such as intelligent express lockers and drone delivery, enterprises can further improve distribution efficiency and meet customers’ personalized needs.

We are pleased to announce a special session call for papers on the cutting-edge topics that are reshaping our world: Intelligent Manufacturing, Smart Logistics, AI, Robotics, and Sustainable Development. This special session aims to bring together researchers, practitioners, and innovators to share the latest research findings, innovative applications, and new perspectives in these rapidly evolving fields. Intelligent manufacturing, production systems, and smart logistics, as the core components of modern manufacturing, are interrelated and mutually promoting. The development of intelligent manufacturing promotes the intelligent upgrading of production systems and improves production efficiency and product quality; advanced production systems provide a solid foundation and support for intelligent manufacturing; and smart logistics closely connects production and the market, achieving the efficient allocation of production resources and the rapid circulation of products. With the continuous progress and innovation of technology, these three fields will continue to deeply integrate, bringing broader prospects for the development of the global manufacturing industry.

Sub-topics

1. Energy efficient and environment friendly manufacturing and service systems
2. Collaborative robots in sustainable manufacturing systems
3. Smart logistics management in sustainable manufacturing and service systems
4. Human-machine interaction for sustainable production optimization
5. Meta-verse in sustainable manufacturing and services
6. Resilience in manufacturing
7. Real-time control of sustainable production and service process
8. Data-driven modelling, monitoring and control of sustainable production and service process
9. Service-oriented smart manufacturing and Robot as a Service (RaaS)
10. AI based design and optimization in sustainable production and service system
11. Digital-twin technology and service-oriented manufacturing technology
12. Green production and service operations management
13. Lean-sustainable industry and services
14. Sustainable supply chain management
15. Predictive maintenance for sustainable operations
16. Eco-design in product development
17. Smart factories and the IoT
18. Data-driven decision making in production and operations
19. Robotics and automation for eco-friendly operations
20. Advanced process control for sustainable operations

Organizers

• [Chao‐Bo Yan], [Professor] [Xi’an Jiaotong University] E‐mail: [chaoboyan@mail.xjtu.edu.cn] Phone: +[86] – [17791257080]
• [Zhi Pei], [Professor] [Zhejiang University of Technology] E‐mail: [peizhi@zjut.edu.cn] Phone: +[86] – [18858123635]
• [Feng Ju], [Associate Professor] [Arizona State University] E‐mail: [fengju@asu.edu] Phone: +[1] – [6085562998]

• [Congbo Li], [Professor] [Chongqing University] E‐mail: [congboli@cqu.edu.cn] Phone: +[86] – [17830505106]
• [Junfeng Wang], [Professor] [Huazhong University of Science & Technology] E‐mail: [wangjf@hust.edu.cn] Phone: +[86] – [13545148279]
• [Chunlong Yu], [Assistant Professor] [Tongji University] E‐mail: [chunlong_yu@tongji.edu.cn] Phone: +[86] – [18221788551]

#25 3D Printing and Social Manufacturing (Code: qh296)

Code: qh296

As technological advancements accelerate, 3D printing has emerged as a transformative force in global manufacturing, offering innovative solutions to overcome the limitations of traditional production methods. This disruptive technology enables personalized, highly efficient, and flexible manufacturing processes by providing precise construction, reducing material waste, and supporting limitless customization. Beyond optimizing production, 3D printing fosters industry-wide innovation, creating new possibilities across sectors.

Simultaneously, the rise of social manufacturing, a collaborative and platform-based production model, is reshaping the manufacturing landscape. Leveraging the power of the Internet, cloud computing, open platforms, and crowdsourcing, social manufacturing connects global producers, consumers, and resources, creating a dynamic ecosystem that enables seamless collaboration. This model challenges the traditional manufacturing monopoly, empowering individuals and organizations worldwide to contribute to production and innovation, thereby maximizing social and economic benefits.

The primary value of social manufacturing lies in its inclusivity and efficiency. By shifting the focus from mass production to on-demand and customized services, this model enables highly flexible supply chains and promotes sustainable, low-waste manufacturing practices. 3D printing plays a pivotal role in supporting this shift, offering adaptable production methods that minimize resource consumption, reduce carbon emissions, and contribute to the broader goals of sustainable development.

Sustainability is now a cornerstone of global manufacturing. Through 3D printing, production processes become more precise, material usage is optimized, and waste is minimized. This technology significantly improves energy efficiency, aligns with green manufacturing initiatives, and fosters a more sustainable future. The integration of 3D printing with social manufacturing platforms enhances collaboration between businesses and consumers, enabling eco-friendly, cost-effective, and efficient production solutions.

Moreover, this new model of social manufacturing needs more flexible automation, especially in the era of AI. In the past and even nowadays, making customization fully automated is difficult or very costly. But with the development of the robot technology and large language models, there is larger and larger possibility for low-cost fully automated customization, making social manufacturing more and more popular. In the future, people can create things in the metaverse with a Non-Fungible Token (NFT), and customers can choose to print the one they like and pay for the IP and manufacturing service. This may be the future manufacturing.

This special session aims to provide a comprehensive platform for academics, industry experts, innovators, and manufacturing professionals to exchange ideas, share real-world applications, and address key challenges in the integration of 3D printing and social manufacturing. Participants will explore cutting-edge solutions for improving production efficiency, reducing costs, and driving sustainable manufacturing practices. The session will also highlight successful industry case studies and future opportunities for innovation.

By fostering cross-disciplinary collaboration, this session will contribute to the advancement of a smarter, more sustainable global manufacturing ecosystem, enabling the industry to become more resilient and flexible. Security and trustworthy automation are indispensable as the metaverse plays a more important role in the new manufacturing model. The research on the 3D Printing and Social Manufacturing can provide a perfect match for the diverse background of audiences in IEEE CASE 2025.

Sub-topics

1. Parallel Intelligence or Digital Twin for 3D Printing and Social Manufacturing.
2. 3D Printing in Personalized Manufacturing: Applications & Challenges.
3. Integration of 3D Printing and AI.
4. Intelligent Decision Support in Social Manufacturing.
5. Data-Driven Decision Systems in 3D Printing & Social Manufacturing.
6. Real-Time Monitoring and Quality Control in 3D Printing.
7. Distributed Production and Collaboration in 3D Printing.
8. Sustainability Challenges in 3D Printing for Social Manufacturing.
9. Robot Technology in 3D printing.
10. LLM Based Automation for 3D printing.
11. Metaverse Based Customization Manufacturing.

Organizers

• Fei-Yue Wang, Professor, Institute of Automation, Chinese Academy of Sciences. E-mail: feiyue.wang@ia.ac.cn
• Pingyu Jiang, Professor, Xi’an Jiaotong University. E-mail: pjiang@mail.xjtu.edu.cn
• Qiang Huang, Professor, University of Southern California. E-mail: qiang.huang@usc.edu
• Chunyuan Pian, Professor, Xinxiang University. E-mail: pianchunyuan69@xxu.edu.cn
• Di Wang, Professor, South China University of Technology. E-mail:mewdlaser@scut.edu.cn
• Zhen Shen, Professor, Institute of Automation, Chinese Academy of Sciences. E-mail: zhen.shen@ia.ac.cn

#26 Automated Optimization for Energy Systems (Code: b7b59)

Code: b7b59

This invited session aims at bringing together research experts in the broad scientific area and industrial application domain of automated optimization in energy systems, ranging from automated thermal control of energy efficient buildings to optimal energy management in smart energy communities and in the power distribution grid. Automation and Optimization in Energy Systems has received extremely high research attention in the last years in the context of applied science and engineering. One of the organizers, Prof. Mariagrazia Dotoli, Senior Editor of IEEE TASE, General Chair of the CASE 2024, will be a keynote speaker at CASE 2025.

Sub-topics

1. Automation for Energy and Sustainability
2. Optimization and Optimal Control
3. Building Automation
4. Demand Side Management
5. Distributed Generation and Storage
6. Modelling, Simulation and Validation of Cyber-physical Energy Systems
7. Plug-in Electric Vehicles
8. Power and Energy Systems automation
9. Smart Home and City
10. Smart Grids

Organizers

• Prof. Sergio Grammatico (TU Delft, The Netherlands) [s.grammatico@tudelft.nl]
• Prof. Mariagrazia Dotoli (Politecnico di Bari, Italy) [mariagrazia.dotoli@poliba.it]
• Dr. Raffaele Carli (Politecnico di Bari, Italy) [raffaele.carli@poliba.it]
• Dr. Paolo Scarabaggio (Politecnico di Bari, Italy) [paolo.scarabaggio@poliba.it]
• Mr. Nicola Mignoni (Politecnico di Bari, Italy) [nicola.mignoni@poliba.it]

#27 Modeling and Control to Enable Automation, Adaptability, and Reconfiguration in Manufacturing Systems (Code: yc88x)

Code: yc88x

In the past several years, manufacturers have seen an increase in the demand for customized production, faster delivery times, and higher product quality. This has been a challenge for manufacturers as traditional manufacturing systems have focused on mass production with limited flexibility. Furthermore, these challenges have been exacerbated by a dwindling manufacturing workforce and the ever‐changing technological landscape.

This special session focuses on leveraging dynamic modelling, optimization, and control theory to enable automation, adaptability, and reconfiguration in manufacturing systems. Specifically, this session explores how to improve system‐level behaviour of various manufacturing systems using the latest manufacturing system technology, standards, and tools. Potential topics of interest include system‐level models and software architectures, dynamic models for manufacturing systems, Digital Twins for manufacturing, optimization and control methods for system‐level control, reconfigurability and flexibility of manufacturing systems, and other general approaches toward modeling and control of manufacturing systems. The session will gather an international group of researchers to collaboratively explore opportunities, challenges, and future directions in the area of modeling and control of manufacturing systems.

Sub-topics

1. Modeling
2. Control
3. Optimization
4. Planning
5. Digital Twins
6. Reconfiguration
7. Adaptability
8. Learning
9. Automation
10. Advanced Manufacturing

Organizers

• Kira Barton, Professor in Robotics, University of Michigan, USA. E‐mail: bartonkl@umich.edu
• Ilya Kovalenko, Assistant Professor in Mechanical and Industrial Engineering, Penn State University, USA. E‐mail: iqk5135@psu.edu
• Efe Balta, Head of Control and Automation Research Group, Inspire AG, Switzerland. Email: efe.balta@inspire.ch
• Sara Wang, Research Fellow in Aerospace Manufacturing, University of Nottingham, UK. Email: Sara.Wang@nottingham.ac.uk

#28 Emerging Data Science in Manufacturing: Autonomous and Software-Defined Factory (Code: e5925)

Code: e5925

Manufacturing is characterized by capital/labor-intensive, the short product life cycle, rapid technology migration, long production lead-time, and complex production networks. These characteristics bring more challenges and difficulties to the manufacturing management. This session focuses on how the data science or machine learning techniques support problem-solving and enhance the core competence in manufacturing industry. The special session focuses on data science and engineering in the broad area of manufacturing. Theoretical research or empirical study are all welcome. The topics in this session include autonomous manufacturing, software-defined factories, defect classification, maintenance scheduling, predictive maintenance, and process parameter optimization, etc. It will explore the architecture, implementation, and operational advancements enabled by these technologies.

This session would like to provide a platform that offers opportunities to discuss, debate, and exchange ideas, in particular, in a world-side view of manufacturing system. We invite all the researchers, scholars, and graduates who would like to develop the mathematical/empirical models and benefit the automation and data science field.

Sub-topics

1. Agent-based collaborative automation systems
2. AI/Smart manufacturing and factory automation
3. Automated fault detection, diagnostics, and prognostics
4. Autonomous factory and software-defined factory (SDF)
5. Big data, data mining, and machine learning
6. Capacity planning and scheduling
7. Circular Economy
8. Cyber physical production systems and industry 4.0
9. Data-driven Modelling Methodologies
10. Digital Product Passports for Recycle and Re-use
11. Digital Twins and Simulation Models
12. Modeling, simulation, and optimization of automation systems
13. Process Mining for Manufacturing
14. Smart logistics and supply chains
15. Sustainability and green manufacturing

Organizers

• [Chia-Yen Lee], [Professor] [Chia-Yen Lee, National Taiwan University, Taiwan] E-mail: [chiayenlee at ntu dot edu dot tw] Phone: +[886] – [ 233661206]
• [Chia-Yu Hsu], [Professor] [Chia-Yu Hsu, National Taiwan University of Science and Technology, Taiwan] E-mail: [cyhsu at mail dot ntust dot edu dot tw] Phone: +[886] – [227376337]
• [Shu-Kai S. Fan], [Professor] [Shu-Kai Fan, National Taipei University of Technology] E-mail: [morrisfan at ntut dot edu dot tw] Phone: +[886] – [ 227712171#2382]
• [Feng Ju], [Associate Professor] [Arizona State University] E-mail: [fengju@asu.edu]
• [Jakey Blue], [Associate Professor] [Jakey Blue, National Taiwan University, Taiwan] E-mail: [jakeyblue at ntu dot edu dot tw] Phone: +[886] – [ 233661531]
• [Young Jae Jang], [Associate Professor] [Young Jae Jang, KAIST, South Korea] E-mail: [yjang@kaist.edu] Phone: +[82] – [(0) 42 350 31 30]
• [Anders Skoogh], [Professor] [Anders Skoogh, Chalmers University of Technology, Sweden] E-mail: [anders.skoogh@chalmers.se] Phone: +[46] – [31 772 48 06]
• [Giovanni Lugaresi], [Assistant Professor] [Giovanni Lugaresi, KU Leuven, Belgium] E-mail: [giovanni.lugaresi@kuleuven.be] Phone: +[32] – [16 32 68 22]

#29 Federated and Distributed Learning for Cyber-Physical Systems (Code: cc9i7)

Code: cc9i7

Complex cyber-physical systems (CPS) can be found in a variety of domains, including manufacturing, agriculture, and healthcare. Within each of these domains there is an ever-increasing amount of distributed data. Although such data provide great opportunity for improving prediction and decision-making, the distributed nature of the collected data creates a number of challenges. Specifically, data privacy concerns have intensified in recent years and drove the demand to store and analyze data at the edge of networks.

Federated learning (FL) and distributed learning (DL) frameworks have been introduced as a solution to these concerns. These approaches enable model training to occur in a distributed fashion (locally across the different nodes), while facilitating collaborative learning without direct data sharing, ensuring privacy preservation and compliance with regulatory constraints.

This special session on “Federated and Distributed Learning for Cyber-Physical Systems” aims to explore the latest developments, challenges, and applications of FL and DL in CPS. Key objectives are summarized as follows:

• Showcasing innovative methodologies in federated and distributed learning for real-world CPS applications.
• Handling challenges associated with efficiency, model convergence, communication overhead, and security in decentralized learning environments.
• Covering approaches for handling high-dimensional, multimodal, and dynamic data in FL/DL frameworks.
• Discussing the role of uncertainty quantification, causal inference, and Bayesian techniques in federated settings.
• Examining deployment strategies for FL and DL systems, including scalability, latency optimization, and resource allocation.

This session will enable discussions on cutting-edge research and practical implementations across both academia and industry, providing insights for how FL and DL can improve decision-making in distributed CPS while having robust, privacy-preserving, and interpretable solutions.

Sub-topics

1. Federated Causal Learning
2. Federated Continuous Optimization
3. Bayesian Federated Learning
4. Deployment of Federated Systems (e.g., challenges in latency, efficiency, and sampling)
5. Federated Learning in High-Dimensional and Multimodal Data
6. Uncertainty Quantification for Federated Learning (e.g., conformal prediction, etc.)
7. Federated Process Monitoring and Prognostics
8. Federated Learning for Distributed Manufacturing, Agricultural, and Healthcare Systems

Organizers

• Mostafa Reisi Gahrooei, Assistant Professor, University of Florida, E-mail: mreisigahrooei@ufl.edu
• Xubo Yue, Assistant Professor, Northeastern University. E-mail: x.yue@northeastern.edu
• Nathan Gaw, Assistant Professor, Air Force Institute of Technology, E-mail: nathan.gaw@au.af.edu

#30 Data Analytics and Optimization for Edge Computing-Enabled Smart Systems (Code: 1vxm5)

Code: 1vxm5

Edge computing has emerged as a transformative paradigm in IoT and smart systems, enabling real-time processing and decision-making at the network edge. As IoT applications continue to grow in scale and complexity, edge computing has become a critical paradigm to reduce latency, enhance privacy, and optimize resource utilization. This session will focus on data analytics for edge computing-enabled smart systems. Specifically, the session discusses AI/ML techniques, statistical methods, and optimization frameworks that enable efficient decision-making, real-time processing, and adaptive learning at the edge. This session seek contributions that push the boundaries of computational efficiency, resource optimization, and decision-making in edge environments, particularly in applications such as industrial automation, advanced manufacturing, smart cities, healthcare, and intelligent transportation systems.

Sub-topics

1. Holistic Data Management and Real-Time Processing for IoT in Edge Computing Environments
2. Bayesian Optimization and Uncertainty Quantification for Decision-Making in Edge-Enabled IoT Systems
3. Machine Learning Techniques for Hyperspectral Data Analysis in Edge-Enabled IoT Applications
4. Microrobotics and Multi-Scale Robotics for Edge-Enabled IoT Applications
5. Reinforcement Learning for Dynamic Sampling in Edge Computing
6. Statistical Process Control and AI for Anomaly Detection in IoT-Enabled Edge Systems
7. Federated Learning for Decentralized Edge Analytics in IoT Systems
8. Energy-Efficient Edge Computing for IoT and Smart Systems
9. Digital Twins and Edge Computing for Smart Systems Optimization
10. Meta-learning and fast adaptation using edge computing
11. Scalable approaches in data analytics for edge computing-enabled systems

Organizers

• Elif Konyar, Postdoctoral Fellow, Georgia Institute of Technology. E-mail: ekonyar3@gatech.edu
• Mohammad Bisheh, Postdoctoral Fellow, Georgia Institute of Technology. E-mail: mbisheh3@gatech.edu
• Seokhyun Chung, Assistant Professor, University of Virginia, E-mail: schung@virginia.edu

#31 Process Level Modeling, Control, and Optimization to Address Advanced Manufacturing Needs for Flexibility and Sustainable Automation (Code: 1kn37)

Code: 1kn37

Modern manufacturing systems have stringent requirements and constraints due to limited resources and increasing quality requirements. Moreover, resource efficiency plays an increasingly important role in promoting sustainability. Process level optimization and control are fundamental tools to address the needs of modern advanced manufacturing systems and to promote sustainable automation.

This special session focuses on leveraging dynamic modeling, optimization, and control theory to enable automation, adaptability, and reconfiguration in manufacturing processes. Specifically, the session explores data‐driven optimization tools for fast adaptation, model predictive control for robotics systems and various additive manufacturing systems, path planning and hierarchical control for robotics systems, and novel applications at the intersection of advanced manufacturing processes and robotics. The tools in the session focus on using real‐time sensory data for dynamic decision‐making, reconfiguration, and adaptation, making it a timely collection of contributions in the field of manufacturing process automation. The session will gather an international group of researchers to collaboratively explore opportunities, challenges, and future directions in the area of modeling and control of manufacturing processes and related robotics applications.

Sub-topics

1. Data‐driven Optimization Tools
2. Physics‐informed Modeling
3. Model Predictive Control
4. Hierarchical Control
5. Path Planning
6. Robotics
7. Additive Manufacturing
8. Learning and adaptation

Organizers

• Kira Barton, Professor in Robotics, University of Michigan, USA. E‐mail: bartonkl@umich.edu
• Ilya Kovalenko, Assistant Professor in Mechanical and Industrial Engineering, Penn State University, USA. E‐mail: iqk5135@psu.edu
• Efe Balta, Head of Control and Automation Research Group, Inspire AG, Switzerland. Email: efe.balta@inspire.ch
• Sara Wang, Research Fellow in Aerospace Manufacturing, University of Nottingham, UK. Email: Sara.Wang@nottingham.ac.uk
• Douglas Bristow, Professor in Mechanical and Aerospace Engineering, Missouri S&T, USA. Email: dbristow@mst.edu

#32 Trustworthy virtual experiments and digital twins Design for Metrology 4.0 in the context of Smart Manufacturing (Code: k3xv8)

Code: k3xv8

Virtual Experiments (VEs) and Digital Twins (DTs) are critical technologies for advancing digitalisation and sustainability in Industry 4.0 and 5.0. VEs use numerical simulations to model physical systems, while DTs extend this concept by dynamically updating models with real-time measurement data through Physical-to- Virtual (P2V) and Virtual-to-Physical (V2P) connections. These technologies play a crucial role in precision manufacturing, metrology, predictive maintenance, and autonomous quality control. To ensure their reliability in industrial and metrology applications, robust uncertainty evaluation and validation methodologies are required. Differences between calibrated standards (or measurement data obtained with calibrated instruments) and their virtual counterparts need to be quantified to enhance trust, accuracy, and reproducibility. Therefore, this special session aims to discuss recent advancements in uncertainty quantification, validation, and real-world applications of VEs and DTs. It will bring together researchers, engineers, and industry leaders to explore scientific and engineering solutions for building trustworthy digital twins and virtual experiments in metrology, manufacturing, precision engineering, and quality assurance.

Sub-topics

1. Uncertainty quantification for VEs/DTs
2. Validation of VEs/DTs
3. Applications of VEs/DTs in precision metrology
4. Applications of VEs/DTs in smart manufacturing
5. AI-driven VEs and DTs modelling
6. AI-driven uncertainty evaluation
7. Trustworthy and traceable VEs/DTs
8. Surrogate modeling in VEs/DTs
9. Metrology digital twins

Organizers

• Dr. Charyar MEHDI-SOUZANI, Associate Professor, Université Paris-Saclay, ENS-Paris-Saclay, Université Sorbone Paris-Nord, LURPA France. E-mail: charyar.souzani@ens-paris-saclay.fr
• Dr. Gengxiang CHEN, research fellow, Université Sorbone Paris-Nord, France. gengxiang.chen@univ- paris13.fr
• Dr. Nabil ANWER, Professor, Université Paris-Saclay, ENS-Paris-Saclay, LURPA, France. E-mail: nabil.anwer@ens-paris-saclay.fr
• Dr. Hichem NOUIRA, Researcher, LNE, France. E-mail: Hichem.nouira@lne.fr
• Dr. Sonja SCHMELTER, Reasearcher, PTB, Germany. E-mail: sonja.schmelter@ptb.de

#33 Data-driven and physics-informed modelling for complex cyber-physics industrial systems (Code: kvp15)

Code: kvp15

In pursuit of customisation and smartness in industrial systems, data-driven modelling has emerged as a fundamental tool for predictive modelling and process optimisation in manufacturing engineering and broader industrial contexts. Data-driven techniques allow the integration of multi-modal data, machine learning, and advanced analytics to extract insights from complex datasets, enabling applications such as predictive maintenance, defect detection, process optimisation, and quality control. However, the effectiveness of data-driven modelling in industry faces challenges, including labelled data scarcity, the unavailability of data of consistent quality and the lack of interpretable and reliable predictions. Recent advancements in neural operators and physics-informed machine learning further extend the capability of data-driven modelling and bring revolutionary potential to manufacturing engineering. The integration of physics priors and domain knowledge into data-driven models has been a potential direction for compensating for labelled data inefficiency and improving prediction reliability. Therefore, this special session focuses on the methodologies and applications of combining domain knowledge, physics priors to improve the reliability, interpretability, and generalisability of data-driven machine learning models, particularly in steh contexte of smart manufacturing and the developpement of new complex industrial systems.

Sub-topics

1. Data-driven modelling in industry
2. Physics-informed data-driven modelling
3. Data-driven smart manufacturing
4. Neural operators
5. Physics-informed neural networks
6. Transfer learning
7. Scientific machine learning
8. AI for science
9. Hybrid machine learning

Organizers

• Dr. Charyar MEHDI-SOUZANI, Associate Professor, Université Paris-Saclay, ENS-Paris-Saclay, Université Sorbone Paris-Nord, LURPA France. E-mail: charyar.souzani@ens-paris-saclay.fr
• Dr. Gengxiang CHEN, research fellow, Université Sorbone Paris-Nord, France. gengxiang.chen@univ- paris13.fr

#34 AI-Powered Collaborative Manufacturing and Scheduling for Industry 5.0 (Code: 256tf)

Code: 256tf

The advent of Industry 5.0 emphasizes the fusion of cutting-edge technologies with a renewed focus on human-centric values, sustainability, and resilience. Unlike its predecessor, Industry 5.0 not only leverages intelligent systems like Artificial Intelligence (AI) and robotics but also aspires to achieve harmonious collaboration between humans and machines while fostering sustainable production practices and resilient industrial ecosystems. This paradigm shift unlocks opportunities for innovation in productivity, adaptability, and environmental responsibility.

AI technologies have become pivotal in driving Industry 5.0, revolutionizing manufacturing processes with enhanced agility and efficiency. By enabling intelligent decision-making, real-time adaptability, and predictive insights, these technologies bring transformative potential to smart manufacturing. Collaborative manufacturing and intelligent scheduling have emerged as crucial research areas to meet these demands. In scenarios such as predictive maintenance, material transportation, and workforce allocation, integrating AI- driven solutions has shown promise in addressing complex, dynamic decision-making challenges with greater resilience and flexibility.

This special session aims to delve deep into the core challenges and opportunities introduced by collaborative manufacturing under the Industry 5.0 framework. By emphasizing AI-assisted approaches, it seeks to address critical issues related to sustainable production, real-time decision-making amidst uncertainties, and efficient interaction and coordination among humans, robots, and machines. The discussions and outcomes of this session are expected to contribute significantly to advancing manufacturing efficiency, reliability, and engagement while supporting the creation of sustainable and world-class production systems.

Sub-topics

1. AI-Enabled Predictive Maintenance in Collaborative Manufacturing
2. Collaborative Optimization for Material Handling and Logistics
3. Real-Time and Adaptive Scheduling Theories and Methodologies for Industry 5.0
4. Data- and Knowledge-Driven Collaborative Manufacturing and Scheduling
5. Human-Robot Collaboration in Manufacturing
6. Theories and Applications of Multi-Agent Systems for Intelligent Manufacturing
7. Applications of Reinforcement Learning in Manufacturing Scheduling
8. Digital Twin-Driven Optimization for Manufacturing Systems
9. Energy-Aware and Sustainable Manufacturing
10. Human-in-the-Loop Manufacturing
11. Time Series Forecasting for Industry 5.0
12. Emerging Paradigms in Manufacturing and Scheduling Enabled by Large Language Models

Organizers

• Junkai Wang, Assistant Professor, Tongji University. E-mail: jkwang@tongji.edu.cn
• Qing Chang, Professor, University of Virginia, E-mail: qc9nq@virginia.edu
• Andrea Matta, Professor, Politecnico di Milano, E-mail: andrea.matta@polimi.it
• Xi Vincent Wang, Associate Professor, KTH Royal Institute of Technology, E-mail: wangxi@kth.se
• Xiaoou Li, Professor, CINVESTAV-IPN, E-mail: xiaoou.li@cinvestav.mx
• Ying (Gina) Tang, Professor, Rowan University, E-mail: tang@rowan.edu
• Chaobo Yan, Professor, Xi’an Jiaotong University, E-mail: chaoboyan@mail.xjtu.edu.cn
• Haibin Zhu, Professor, IEEE Fellow, Nipissing University, E-mail: haibinz@nipissingu.ca

Contact

For any inquiries please contact the Special Session Chairs.