Abstract
Complex systems, characterised by high interconnectedness and unpredictability, demand structured approaches to support decision-making, system integration, and operational efficiency. This study aims to develop a comprehensive understanding of these systems to facilitate the implementation of Digital Twins (DTs) in Complex Operations Environments (COEs). A Systematic Literature Review (SLR), following PRISMA methodology, is conducted and complemented by a grounded theory approach to identify, organise, and synthesise system characteristics. The result is a hierarchical conceptualisation comprising eight core categories that capture the essential dimensions of complexity, including: Process and System Interoperability, Human Relationships, Behaviour/Nature, Products and Services, Multi-processes, Performance, System Structure, and Management and Skill Requirements. This framework offers a structured foundation for designing and implementing DT-based decision support in COEs.

Abstract
Organizations can transform their businesses and create more value by adopting Industry 4.0 initiatives. During evaluating these projects, the decision-maker must assess significant uncertainties (risks) resulting from socio-technical, economic, and financial factors. One of the main objectives of this study was to identify the necessary building blocks to develop a framework for project implementation in high-risk scenarios, as in the case of Industry 4.0. A multi-criteria framework divided into three stages was proposed, integrating knowledge from Front-End-Innovation (FEI), Innovation Decision Process (IDP), Traditional Project Evaluation Methods, and Real Options Valuation (ROV). The first step is to identify an investment opportunity. The second step is the definition of a business model. The third step is the simulation of different implementation strategies to give managerial flexibility to decision-makers to decide the best strategy to mitigate risks. A real case study was used to test the framework. According to the results, managers can use this framework to create different project implementation scenarios and determine the best strategy to mitigate risks. However, we must still understand whether uncertainties behave discretely, dynamically, or both, the interactions between elements, and how to calculate them to improve our model.

Abstract
Manufacturing industry is experiencing another revolution towards the digitalization of industrial processes. Different value chain actors must share specific and sensitive data according to business and data requirements. Digital architectures must ensure seamless and comprehensive communications between actors according to agreed-upon vocabularies. The digital representation of machines and other types of equipment, including crucial information about their static and dynamic operational data, is made possible by the ontological modelling of Asset Administration Shells (AAS), which is proposed in this paper as modular and semantically interoperable resources. These Cognitive Digital Twins are herein defined with de facto domain ontologies that model the semantics of the current operation, status and configurations of assets. This paper reports a proof-of-concept technical implementation that demonstrates an innovative digital architecture that connects and communicates active and modular Digital Twin of a machine in a bi-directional, connecting this asset to a digital manufacturing service provider.

Abstract
In the Fourth Industrial Revolution era, commonly known as Industry 4.0, the manufacturing industry is undergoing a profound transformation driven by the convergence of technological advancements. Industry 4.0 technologies are revolutionising how products are manufactured, from design to production to delivery. These technologies, such as collaborative robotics, digital twins, IoT, and data analytics, enable manufacturers to improve efficiency, productivity, and quality. As Industry 4.0 continues to evolve, the demand for skilled engineers who can effectively design, implement, and manage these sophisticated systems is growing rapidly. Future mechanical engineers must be prepared to navigate this complex and data-driven manufacturing landscape. To address this need, the Faculty of Engineering at the University of Porto developed a new course titled Smart Factories, specifically designed to equip master's students with the knowledge and skills necessary to thrive in the factories of the future. This course utilises an innovative, active experimental learning methodology with industry collaborations and a comprehensive curriculum to foster the development of the multidisciplinary skills necessary to excel in this rapidly evolving field. Through this comprehensive and innovative approach, the Smart Factories course aims to prepare future mechanical engineers to become leaders in smart manufacturing, driving innovation and shaping future factories.

Abstract
This paper advocates for Digital Twin (DT) technology as a pivotal solution to address the complexities of Complex Operations Environments (COEs). Recognizing the need for a thorough understanding of COEs and their DTs, a methodology is introduced to bridge existing gaps. Given the lack of a universal definition, the approach leverages the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Latent Dirichlet Allocation (LDA) to extract insights, facilitating the development of a comprehensive definition for COE and DT. The methodology integrates Ontology and Systems Modelling Language (SysML) to provide a semantic and conceptual model of COE and DT. Ontology enriches the semantic understanding, exploring existence and entity relationships, while SysML ensures clear and concise communication through standardized graphical representation. This paper aims to present a methodology to achieve a precise understanding of COEs and their corresponding DTs, providing a robust foundation for addressing operational complexities in dynamic environments.