Scientific publications

High-mix low-volume (HMLV) production systems are crucial for enabling mass customization to meet diverse and rapidly changing customer demands. These systems require optimization across various time scales to handle fluctuations in job mix and ensure robustness against unforeseen changes, such as rush orders.

System-Performance Engineering (SysPE) encompasses modeling formalisms, methods, techniques, and industrial practices to design systems for performance, where performance is taken integrally into account during the whole system life cycle. Industrial SysPE state of practice is generally model-based. Due to the rapidly increasing complexity of systems, there is a need to develop and establish model-driven methods and techniques.

In applied research projects, we want to create impact by developing innovative solutions, and land them in the organization. Successfully landing innovations typically requires a direct collaboration with stakeholders in the organization. They need to adopt the results in the system they develop or embed them in their systems development process and way-ofworking.

We are a leading applied research center for systems design and engineering in the high-tech equipment industry. We work in close collaboration with Dutch industry as well as in strong association with the fundamental research of academia, both nationally and internationally. We contribute to society and the economy by driving advances in high-tech systems technology through a strong shared research program, dedicated innovation support, a focused competence development program and various knowledge- and experience-sharing activities.

This work presents the latest advancements in leveraging artificial intelligence (AI) and digital twin (DT) technologies to automate the operation of electron microscopes, with a particular focus on exploring the feasibility of automated electron microscope alignment. The alignment of electron microscopes is a laborious process, demanding significant time and expert knowledge.

This document describes an initial version of the ASIMOV reference architecture, which consists of definitions of terms and multiple architectural views. In addition, this document contains a delimitation of the ASIMOV project tasks in terms of this reference architecture. The description of the reference architecture consists of both a generic architecture and applications to the use cases from the ASIMOV project.

In this document, the specifications, commonalities, and fundamental differences between industrial use cases were gathered, analysed, and reported. Electron Microscopy (EM) and Unmanned Utility Vehicles (UUV) are the two target industrial use cases, which comprise sub use cases, are detailed down and described.

This report describes the state of the art in reinforcement learning and digital twin-based learning, and the first ideas on their application in the ASIMOV use cases. It will be the foundation for further work on researching these techniques in the ASIMOV use cases, which will enable expansion of the knowledge in these fields for general application in the high-tech industry.

WP3 is concerned with the development of a technical approach and a reference architecture for DT-supported AI-based system optimisation. System optimisation can be performed by connecting AI to both the physical system and its DT. By allowing the AI to take control over the DT, a learning cycle based on reward and punishment can be constructed to validate its actions.

The ASIMOV project investigates the combination of Digital Twins (DTs) and Artificial Intelligence (AI) to find the opportunities and challenges for automated optimization and calibration of complex high-tech systems in complex environments. In many cases the actual system is not available for training AI components, therefore a dedicated digital twin or digital model is set up for providing that training data.