INTRODUCTION
Information Modelling (IM) has been under the spotlight of both academia and industry for decades. Important aspects of IM include methods and practices for representing concepts, relationships, constraints, rules, and operations in order to specify data semantics for a chosen domain of interest. In response to the IM challenge, a number of modelling paradigms and languages have emerged, ranging from ERM, UML, and ORM to OWL and Knowledge Graphs, together with a wide range of systems supporting the life cycle of information models.
Despite past success, existing approaches and systems for IM struggle to cope with new challenges arising from global industrial digitalisation, enterprise-wide software integration, and the increasing use of AI and large language models in industrial and organisational settings. These developments require advanced information models that support fully computerised, software-driven automation of production processes, information exchange across heterogeneous systems, and trustworthy AI-based decision support.
Such trends are an important part of Industry 4.0 and the Industrial Internet of Things. They require IM approaches that can, for example, capture the functionality of and information flow between different assets in a plant, such as equipment and production processes. Moreover, they require information models based on ISA and IEC standards with desirable properties such as reusability, explainability, scalability, and simulability.
At the same time, the growing adoption of LLMs in industrial applications introduces new challenges related to trustworthiness, bias, accountability, and transparency. Information modelling can play an important role in structuring domain knowledge, representing assumptions and constraints, detecting or mitigating bias, and supporting explainable and auditable AI systems. This includes applications such as human resource management, recruitment, organisational decision-making, political bias detection, compliance checking, and other high-stakes industrial or socio-technical domains where biased or untrustworthy AI outputs may have significant consequences.
These challenges require new theory, methodology, best practices, systems, and shared experiences, developed and discussed by a wide range of stakeholders. In this workshop, we aim to gather researchers and practitioners who address these challenges with the help of semantic technologies. We particularly invite IM experts who are excited and committed to pushing the frontiers of IM further and supporting modern industry in its current technological and AI-driven transformation. The workshop welcomes novel methods, systems, solutions, experiences, and practices for semantic industrial information modelling, including trustworthy and bias-aware information modelling for LLM-based applications.