SS01 - Industrial Cybersecurity Methods and Technologies

Special Session Organized by

Paulo C. Bartolomeu, University of Aveiro, Portugal and Stefano Marrone, University of Campania "Luigi Vanvitelli", Italy

Download Call for Papers

Click here to download the session cfp.

Focus

Industrial cybersecurity has become a key research topic in recent years due to the massive connectivity brought by the Internet of Things and the rise of cyberattacks against industrial assets. While fostering contemporary applications and use cases, ubiquitous Internet access has also exposed legacy operational technologies to new and challenging security threats that must be addressed. This Special Session focuses on novel security, safety, and privacy-enhancing technologies for current and future industrial applications.

Topics under this track include (but not limited to):

  • Security, safety, or privacy-enhancing technologies in industrial systems
  • Modeling of cybersecurity threats
  • Applications of distributed ledger technologies/blockchains in Industry 4.0
  • Self-sovereign identity for M2M and decentralized device-to-device communication
  • Hardware advances for securing Industrial devices and networks
  • Quantitative evaluation and/or interconnections among non-functional aspects (e.g., reliability vs. safety, security vs. performance)
  • Run-time methods and technologies for Complex Event Detection systems
  • Software engineering methods and techniques for high dependable control systems
  • Case studies / lessons learned of security, safety, or privacy assessments of industrial systems

Complete Details

Click here for complete details.

SS02 - Software Engineering for Cyber-Physical Production Systems (SECPPS)

Special Session Organized by

Istvan Koren, RWTH Aachen University, Germany and Kristof Meixner, TU Wien, Austria and Felix Rinker, TU Wien, Austria and Andreas Wortmann, University of Stuttgart, Germany

Download Call for Papers

Click here to download the session cfp.

Focus

With the emergence of Cyber-Physical Production Systems (CPPSs), systems engineers are currently facing a dramatic increase in the complexity of developing and operating systems. In particular, software plays a crucial role in the effective and efficient operation of CPPSs. Despite the tremendous progress in software engineering approaches and technologies, they do not seem to reach industry. More comprehensive and systematic views on all aspects of systems and their development process are required. The Special Session on Software Engineering for Cyber-Physical Production Systems aims to discuss challenges in adopting state-of-the-art software engineering approaches and technologies to CPPSs, and highlight new methods for the design of software for production systems.

Topics under this track include (but not limited to):

  • Software engineering improvements for and transfer of best practices to CPPSs (e.g., agile methods)
  • Operation, evolution, and management of CPPS software (e.g., DevOps)
  • Software modeling and languages for CPPSs (e.g., model-driven engineering)
  • Interdisciplinary collaboration in the engineering and operation of CPPS software
  • Software engineering education for CPPS engineers
  • Security, resilience and sustainability of CPPS software by design
  • Usability of software development environments for CPPS engineering

Complete Details

Click here for complete details.

SS03 - Capability- and Skill-based Engineering of Manufacturing Systems

Special Session Organized by

Roman Froschauer, University of Applied Sciences Upper Austria, Austria and Aljosha Kocher, Helmut Schmidt University, Germany and Kristof Meixner, CDL-SQI, TU Wien, Austria and Siwara Schmitt, Fraunhofer IESE, Germany

Download Call for Papers

Click here to download the session cfp.

Focus

As customer requirements change more frequently, pursuing flexible and adaptive automation approaches becomes necessary. Such approaches demand an explicit description of a production system's functionality and the products to be manufactured. Recent research has introduced approaches based on capabilities and skills using holistic data models (i.e., ontologies, DSLs, variability models ...). While capabilities are understood as an abstract description of (manufacturing) processes a system can perform, skills are often described as their executable counterparts (i.e., modelling an invocation interface such as OPC UA). To find solutions for customer requirements automatically, required tasks and domain-specific constraints must be matched with capabilities provided by automation components. This can be achieved by various techniques such as AI planning or knowledge graph exploration and reasoning. Process plans can then be orchestrated by combining skills related to capabilities found in the previous step. Finally, simulation and optimization of such process plans can be performed before executing them.

Topics under this track include (but not limited to):

  • Modeling of capabilities, skills and services: Data Modeling, Modeling Languages, Knowledge Graphs, Rule Engines, Knowledge-based Systems, Asset Administration Shell
  • Algorithms to find matching capabilities: Planning, Artificial intelligence, Capability-task-matching, Knowledge Graph Exploration
  • Skill-based production: Generation/Modeling of process plans, Orchestration, Execution, Optimization
  • Simulation of a proposed plan: Optimization, simulation techniques for skills
  • Engineering methods: Automated code generation, model-based programming, automated generation of models
  • Organization of marketplaces and supply chains via services

Complete Details

Click here for complete details.

SS04 - Design and Enforcement of Safety, Security, Privacy and Law in Society 5.0

Special Session Organized by

Muhammad Taimoor Khan, University of Greenwich, UK and Dimitrios Serpanos, ISI Athena, ECE, University of Patras, Greece and Howard Shrobe, MIT CSAIL, USA and Kunio Uchiyama, AI Chip Design Centre, Japan

Download Call for Papers

Click here to download the session cfp.

Focus

Computing constitutes a fundamental component of the emerging Society 5.0, which combines cyber and physical spaces (i.e., processes) and requires control and monitoring techniques for its operation and management. In Society 5.0, people, things, and systems are connected in cyberspace and operate exploiting automated methods, including machine learning (ML) and artificial intelligence (AI). Such operation and management bring new value to industry and society in ways not previously possible. Typical cyber physical systems (CPS) are based on (I)IoT (Industrial - Internet of Things) and (I)CPS (Industrial - Cyber Physical Systems) and have applications in all critical infrastructure domains with strict real-time requirements, such as healthcare, electric grid, transportation, to name a few. Intentional or accidental errors/failures/attacks to these systems have highly severe consequences. Therefore, novel design methodologies are required to ensure that design of real-time cyber physical systems and applications in the emerging Society 5.0 are free of vulnerabilities, threats and attacks. Since the physical part of CPS involves several processes, typically, it is challenging to ensure that the design is free from all known vulnerabilities. It is necessary to develop run-time monitoring and analysis techniques that can help to detect run-time incidents by observing the processes and their data. Furthermore, adequate modelling of CPS physical processes and corresponding cyber and physical attacks is fundamental to systematically model, analyse and verify real-time security of CPS. Importantly, since AI and machine learning have demonstrated their success in many application areas including cyber security, this special session focuses on investigating AI, machine learning and formal methods-based techniques to develop safe, secure, privacy and law-aware real-time cyber physical systems, digital twins and smart cities at all levels, from hardware components to applications.

Topics under this track include (but not limited to):

  • Design-time and run-time safety, security, privacy and law in modern systems, e.g., Society 5.0, Digital Twins, ICPS, and IIoT
  • Data-driven (AI and Machine Learning-based)/model-based
  • Safety, security, privacy and law in cyber-physical systems (CPS), networks and communication
  • Prevention, detection and mitigation techniques for real-time CPS (RT-CPS) applications against cyber and physical threats
  • Hardware design for safe, secure, privacy and law-aware RT-CPS
  • Vulnerability analysis of RT-CPS applications
  • Attack modeling and performance analysis of RT-CPS
  • Formal methods (FM)-based safety and security of critical systems at design-time and run-time
  • Safety, security and privacy of citizens in Society 5.0 including pandemics and disasters
  • Methodologies and tools for analysis, compliance and enforcement of law and regulations for safety, security and/or privacy
  • Methodologies and tools for compliance testing and standardization
  • CAD tools for AI-based cyber-physical systems (CPS)
  • CAD tools for safe, secure, privacy, and law-aware RT-CPS
  • Case studies for AI and machine learning-based RT-CPS
  • Case studies for digital law compliance and regulations in RT-CPS
  • Benchmarks for security, safety, privacy and or/law in RT-CPS
  • Challenges in modelling, analysis, safety, security, privacy and law of RT-CPS

Complete Details

Click here for complete details.

SS05 - Learning on Scarce Data for Data-based Control and Operational Strategy

Special Session Organized by

Jose Lopez Vicario, Universitat Autonoma de Barcelona, Spain and Antoni Morell, Universitat Autonoma de Barcelona, Spain

Download Call for Papers

Click here to download the session cfp.

Focus

The study and application of Machine Learning (ML) algorithms in the industrial domain have become a popular topic of research during the last years. Several authors have proposed data-based approaches to automate and/or optimize industrial activities such as predictive maintenance, quality assurance, stock/demand prediction, energy consumption, etc. Going a step further, ML has also been explored at industrial control level to conduct data-driven operational strategy and implement data-based controllers. A common denominator of these approaches is the need of a huge amount of data to ensure proper training and avoid overfitting. In the context of control and plant operation strategies, deployment on real environments could fall into the small data regime, i.e., the amount of available data is scarce. Scarce Data is not only limited to the volume of the data, but it could also refer to environments with highly unbalanced data or slow dynamics which prevent extracting the underlying pattern from training data. Besides, data-based controllers deal with temporal sequence and sensor data, which also implies an additional challenge to the application of Machine Learning mechanisms. In other ML domains, several strategies have been proposed to overcome the Scarce Data limitations such as the Transfer Learning of models trained at other (big data) scenarios, the generation of artificial data or the adoption of Reinforcement Learning. More recently, Bayesian approaches have also been recovered, where proper probabilistic modeling and the inclusion of prior knowledge is introduced at learning. The aim of this special issue is to address the problem of ML in Industrial Control and operational strategy when dealing with Scarce Data and propose mechanisms to overcome it.

Topics under this track include (but not limited to):

  • Data-efficient algorithms and Machine Learning in the context of Scarce Data
  • Machine Learning and Signal Processing for temporal sequences and sensor data
  • Generative models for industrial data augmentation
  • Reinforcement Learning solutions to tune system operation and data-based control design according to data availability
  • Transfer Learning solutions that use knowledge acquired in big data scenarios to deal with the small data regime
  • Probabilistic Models and Bayesian Learning: how to measure uncertainty and develop solutions to take it into account

Complete Details

Click here for complete details.

SS06 - Machine learning in condition monitoring and prognostics of engineering systems

Special Session Organized by

Miguel Delgado Prieto, Universitat Politecnica de Catalunya, Spain and Juan Jose Saucedo Dorantes, Universidad Autonoma de Queretaro, Mexico

Download Call for Papers

Click here to download the session cfp.

Focus

The massive digitalization of industrial assets carried out in the last decade, allows now the deployment of industrial cyber-physical systems and IIoT architectures including advanced analytics capabilities to improve decision-making processes. In this regard, machine learning (ML) is being included as part of industrial maintenance procedures to improve anomaly detection, fault diagnosis and remaining useful lifetime prediction over assets. However, two main challenges have to be faced for an effective deployment of such ML-based procedures: first, the scalability, that is, hyperparameters selection procedures and interpretable training methods avoiding overfitted algorithms, and, second the reliability, that is, continuous learning approaches avoiding deviations between the supervision model and the operational drifts during the assets useful life.

Topics under this track include (but not limited to):

  • Data-based industrial processes characterization
  • Automated novelty detection in industrial systems
  • Decision support systems and new information fusion structures
  • Condition monitoring and pattern recognition
  • Transfer learning and incremental learning methodologies
  • Forecasting and remaining useful life estimation
  • Data governance and regulations on machine learning in the industrial maintenance

Complete Details

Click here for complete details.