@techreport{fairwork_d3.3, title = {Final DAI-DSS Research Collection - D3.3}, author = {Lucas Paletta, Herwig Zeiner, Gustavo Vieira, Sylwia Olbrych, Alexander Nasuta, Johanna Werz, Johannes Zysk, Noushin Qeybi, Stefan Böschen, Marlene Mayr, Christian Muck}, editor = {Stefan Böschen}, url = {https://fairwork-project.eu/deliverables/D3.3%20DAI-DSS%20Research%20Collection%20v1.0-preliminary.pdf}, year = {2025}, date = {2025-02-28}, urldate = {2025-02-28}, abstract = {This deliverable shows the “Final DAI-DSS Research Collection” as part of the Horizon Europe project FAIRWork. The deliverable aims to describe guidelines, methods and tools for democratising the production process in the light of their flexibilization utilizing artificial intelligence (AI), Optimisation, Human Factors Analytics, and multi-agent systems (MAS) as mediators in the form of prototypes, physical experiments in laboratories, implemented questionnaires, modelling tools or semantic model of criteria catalogues. Importantly, this collection is published in the FARIWork Innovation Shop and represents the key support features for the Democratic AI-based Decision Support System Support System (DAI-DSS). The FAIRWork Innovation shop is accessible online as Deliverable 3.3. This document is considered the accompanying document of the deployed online version. https://innovationshop.fairwork-project.eu/ This deliverable also presents the scientific basis for the FAIRWork project across seven research tracks: The “Democratization of Decision-Making in Socio-Technical Settings” examines the dynamics of democratization in industry through MAS by exploring the contextual conditions for implementing a DSS within socio-technical frameworks. The “Decision-Making Using Multi Agent Systems” explores the potential of MAS for decentralized, adaptive decision-making in industry. By balancing technical, human-centric, and ethical aspects, MAS enhances efficiency, inclusivity, and scalability in complex systems. The “Digital Human Factors Analytics” outlines the use of wearable sensors to capture critical information on human physiological, cognitive-emotional, and resilience states, including the intelligent sensor box (ISB). It also details a novel framework using Personas as Human Digital Twins for Decision Making in Industry 5.0. The “Optimization in Decision Support Systems” is crucial for defining clear goals in AI-driven manufacturing, addressing challenges such as process optimisation, automation, and resource allocation. Various techniques, including AI, new algorithms and heuristics, support decision-making and efficiency in manufacturing companies. The “AI-Enriched Decision Support Systems” explores how AI methodologies, particularly machine learning (ML), can optimise decision-making in manufacturing, with an emphasis on dynamic tasks. It also addresses the gap between industry and developers by proposing a structured categorisation of DSS, enabling developers to select appropriate AI methods for industrial applications. The “Model-based Knowledge Engineering for Decision Support” presents a structured approach to AI adoption in enterprises by proposing a three-layered framework: Identification, Specification, and Configuration. It highlights the role of conceptual and technical models from the identification of the problem setting to the configuration of AI to ensure the alignment with business needs. Furthermore, it also reflects the integration of different AI techniques, such as retrieval-augmented generation (RAG) and large language models (LLMs), within use-case-specific prototypes, demonstrating how model-based methodologies can support AI configuration. It also investigates how such design models can be reused to support the explanation of decision scenarios on a high abstraction level using OMiLAB’s Scene2Model tool. The “Reliable and Trustworthy AI” demonstrates the importance of transparency for trusting AI systems and that transparency needs to be adapted to the target group. AI systems have to be understandable, which is why a system-dependent approach that sets the user in the center is recommended. A developed transparency matrix with additional individual consulting workshops for the developers has shown to be successful in implementing transparency and accuracy communication into AI services.}, howpublished = {https://fairwork-project.eu/deliverables/D3.3%20DAI-DSS%20Research%20Collection%20v1.0-preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d4.3, title = {Final DAI-DSS Prototype, Documentation and Test Report - D4.3}, author = {Herwig Zeiner, Lucas Paletta, Julia Tschuden, Michael Schneeberger, Gustavo Vieira, Rui Fernandes, Johanna Lauwigi, Alexander Nasuta, Damiano Falcioni, Marlene Mayr, Christian Muck, Rishyank Chevuri}, editor = {Marlene Mayr}, url = {https://fairwork-project.eu/deliverables/D4.3%20Final%20DAI-DSS%20Prototype%20v1.0-preliminary.pdf}, year = {2025}, date = {2025-02-28}, abstract = {This deliverable, “D4.3 – Final DAI-DSS Prototype, Documentation and Test Report”, provides a comprehensive overview of the final implementation of the DAI-DSS. Building on the foundations established in “D4.1 – DAI-DSS Architecture and Initial Documentation and Test Report” and “D4.2 – Initial DAI-DSS Prototype”, this document details the integration, orchestration, and deployment of the components for AI-based decision-making within industrial applications. The prototype is designed to enhance operational efficiency and support decision-making for various scenarios. In this last version, the DAI-DSS Prototype extends its applicability across multiple industrial use cases, including workforce allocation, production planning, machine maintenance, and validation of documents. The demonstration materials can be accessed via the following link: https://innovationshop.fairwork-project.eu/ By leveraging a modular and scalable architecture, the system facilitates the interaction between AI services, UIs, and structured data repositories. The implementation of DAI-DSS consists of several integrated building blocks: The DAI-DSS User Interface collects multiple UI components for the different scenarios and AI services to enable stakeholders to visualize data, interact with decision-making tools, and monitor industrial workflows. The DAI-DSS Orchestrator component serves as the central coordination engine, managing workflows, microservices, and AI-driven recommendations to ensure the system operation. It includes different approaches that range from centralized to decentralized prototypes. The DAI-DSS Configurator consists of a tool designed to enhance decision support systems through configuration and integration frameworks. It consists of the Configuration Framework, which assists in creating decision models and strategies, and the Configuration Integration Framework, which generates system configurations. It allows for microservices and workflow configuration, featuring an interface with a wizard for UI components combination. The DAI-DSS Knowledge Base is highlighted as a central data repository, storing user properties, sensor data, and processed data. It plays a key role in the system's data flow, integrating with the Configurator and using REST API for data retrieval. The DAI-DSS AI Enrichment incorporates various decision-making techniques and AI services, including neural networks, decision trees, constraint programming, Multi-Agent Systems (MAS), Large Language Model (LLM), and retrieval-augmented generation (RAG), to provide tailored recommendations and automation support. The final DAI-DSS Prototype delivers several advancements over previous iterations by 1) supporting AI-driven decisions that aim to enhance decision-making and information access with different AI techniques while including reflections on AI and data reliability, 2) ensuring scalability and adaptability of the system by its component-based architecture that allows for integration with various applications and expansion into new domains, 3) advancing data utilization and processing with efficient storage, retrieval, and processing of industrial data, in the Knowledge Base and Vector Databases and 4) proposing a flexible approach to enable the extension with new prototypes. The DAI-DSS marks a step forward in AI-powered decision support for industrial environments. Its modular and scalable architecture provides a foundation for future AI enhancements, data integration, and broader enterprise adoption. In particular, the results and prototypes aim to be used as starting point for use cases in the area of robots in manufacturing settings for example supporting decisons in maintenance or optimal robot-task and line allocation. Furthermore, findings and implementations documented in this deliverable contribute to advancing intelligent, and effective decision-support solutions in industrial ecosystems, and aim to contribute to future European AI research and reference architectures. }, howpublished = {https://fairwork-project.eu/deliverables/D4.3%20Final%20DAI-DSS%20Prototype%20v1.0-preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d3.2, title = {First DAI-DSS Research Collection - D3.2}, author = {Lucas Paletta, Herwig Zeiner, Michael Schneeberger, Julia Tschuden, Martin Pszeida, Andreas A. Mosbacher, Gustavo Vieira, Sylwia Olbrych, Alexander Nasuta, Johanna Werz, Noushin Gheibi, Stefan Böschen, Magdalena Dienstl, Marlene Mayer, Christian Muck}, editor = {Lucas Paletta}, url = {https://fairwork-project.eu/deliverables/D3.2_First-DAI-DSS-ResearchCollection_V1.0_preliminary.pdf}, year = {2024}, date = {2024-04-30}, urldate = {2024-04-30}, abstract = {This report focuses on deliverable “First DAI-DSS Research Collection”, which is part of the Horizon Europe project FAIRWork. The deliverable aims to describe a first iteration on guidelines, methods and tools for democratising the production process in the light of their flexibilisation while using Artificial Intelligence, Optimisation, Human Factors Analytics, and Multi Agent Systems (MAS) as mediators in form of prototypes, physical experiments in laboratories, implemented questionnaires, modelling tools or semantic model of criteria catalogues. It presents a collection of concepts, methods, studies, and services of a research framework within the Democratised AI Decision Support System (DAI-DSS). The DAI-DSS research collection is based on the fundamental principles related to the research intended within the frame of this project that was described in Deliverable “D3.1 DAI-DSS Research Specification” and incorporates cross-connections with Deliverable “D4.1.1 DAI-DSS Architecture and initial Documentation and Test Report“ on the basis of functional components of the DAI-DSS system architecture. The first part of the report provides an overview of the individual research tracks related to the various research strategies that incrementally shape and extend the research collection within the frame of this project and in the context of the scientific as well as industrial communities. It covers the most significant research domains such as democratisation of decision-making as well as digital shadows for resilience risk stratification or human experts. Additionally, it explores technical approaches like Artificial Intelligence (AI) and MAS crucial for improving Decision Support Systems (DSS). This section also presents the state-of-the-art in key aspects of today's technology, particularly reliability and trustworthiness in AI. The output of this research overview leads to the outline of research activities in multiple domains that will be addressed within the FAIRWork project. The second part of the report focuses on the research collection in terms of the concrete research methods and services employed to investigate the technical aspects of decision-making processes, human aspects in the process, and digital Human Factors measurements. It presents research approaches for the successful implementation of AI and MAS-based technologies into DSS. Methods such as data-driven modelling, prototyping, and testing are proposed within the AI and MAS domains. Additionally, the report outlines the use of wearable sensors to capture critical information about the physiological, cognitive-emotional, and resilience state of humans, including implementation details of the Intelligent Sensor Box (ISB). Furthermore, the novel framework using Personas as Human Digital Twins for Decision Making in the context of Industry 5.0 is described in detail. The third part covers initial observations on explainability and fairness in FAIRWork from an algorithmic point of view and summarises some reviews and surveys in the field. The report also provides results of the strategy for scientific dissemination in the context of the research methodology of the FAIRWork project. The objective is to continuously disseminate project achievements, raise awareness about the project, and gather feedback to improve the created research artefacts. }, howpublished = {https://fairwork-project.eu/deliverables/D3.2_First-DAI-DSS-ResearchCollection_V1.0_preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d4.1.1, title = {DAI-DSS Architecture and Initial Documentation and Test Report - D4.1.1}, author = {Rishyank Chevuri, Herwig Zeiner, Lucas Paletta, Michael Schneeberger, Robert Woitsch, Magdalena Dienstl, Gutavo Vieira, Sylwia Olbrych, Lauwigi Johanna, Alexander Nasuta, Christian Muck, Roland Sitar}, editor = {Rishyank Chevuri}, url = {https://fairwork-project.eu/deliverables/D4.1.1_DAI-DSS_ArchitectureAndInitialDocumentationAndTestReport_V1.0_preliminary.pdf}, year = {2024}, date = {2024-04-30}, urldate = {2024-04-30}, abstract = {This deliverable “D4.1.1 – FAIRWork Architecture and initial Documentation and Test Report” is the second iteration of the deliverable D4.1 which has been submitted in the month M6 of the project. This iteration of the deliverable builds upon it and provides a comprehensive overview of the current status of the implementation of DAI-DSS components and services required for the decision support. This iteration outlines key performance indicators for FLEX, covering "Automated Test Building," "Worker Allocation," and "Machine Maintenance After Breakdown," as well as for CRF, focusing on "Workload Balance," "Delay of Material," and "Quality Issues." These key performance indicators (KPIs) have been revised to align with the FAIRWork goals and objectives. These metrics will be used as benchmarks for evaluating the solutions proposed by the DAI-DSS for the described challenges. The list of AI services proposed to support the use case challenges are revised in this iteration and further classified into three categories based on their development status: "Initial Prototype", "Development in Progress" and "Planned for Implementation" indicating their status of implementation. In addition to this the prerequisites and usage descriptions for the services have been added. These additional descriptions outline the specific circumstances under which the service is designed to operate, which aids in the selection process, ensuring that end users choose the most appropriate service for their specific scenarios. This iteration further specifies the software and hardware requirements essential for the operation of DAI-DSS components and provides the status of the functional capability’s implementations, including User Interface, Configurator, Orchestrator, Knowledge Base, AI Enrichment, and Sensor Boxes. Furthermore, the information related to the security considerations critical to the DAI-DSS architecture, highlighting mechanisms for data protection, authentication, and authorization has been described. This includes employing robust security practices such as SSL certificates, two-factor authentication, role-based access control, and secure data transmission protocols, which ensures the system is safeguarded against unauthorized access and data breaches, ensuring the protection of sensitive information. Additionally, an outlook on the necessary security measures and implementations aimed at enhancing the DAI-DSS's security has been described. Finally, the iteration provides an overview of the testing methods employed to evaluate the integrity and functionality of the DAI-DSS architecture. Describing the methodologies for assessing the interactions between both internal and external components, the functional evaluation of each distinct component, and the integration testing of the components. This ensures that the DAI-DSS operates seamlessly, maintaining high standards of quality and reliability in its performance. The contents of this deliverable will contribute to the “D4.3 Final DAI-DSS Prototype, Documentation and Test Report”, scheduled for M30. }, howpublished = {https://fairwork-project.eu/deliverables/D4.1.1_DAI-DSS_ArchitectureAndInitialDocumentationAndTestReport_V1.0_preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d5.2, title = {DAI-DSS Infrastucture and Setup Report at Use Case Site - D5.2}, author = {Aleassandro Cisi, Roland Sitar}, editor = {Roland Sitar}, url = {https://fairwork-project.eu/deliverables/D5.2_DAI-DSS_InfrastructureAndSetupReportAtUseCaseSite_v1.0_preliminary.pdf}, year = {2024}, date = {2024-04-30}, abstract = {The purpose of the document, "D5.2 – DAI-DSS INFRASTRUCTURE AND SETUP REPORT AT USE CASE SITE" is to provide an update on the progression and current state of implementation on use case sites of the Democratic AI-based Decision Support System (DAI-DSS) Architecture as part of the FAIRWork project. Section 2 of the deliverable describes the overall infrastructure setup process at the use case partner side. It also describes the setup of the isolated computing system for the DAI-DSS tool to avoid risks to the corporate network. Section 3 then details the technical preparation tasks such as data collection, human expert training, user selection and implantation of the current DAI-DSS prototype as a use case site and integration with the legacy systems. Section 4 describes the overall testing procedure. This includes a brief description of the user evaluation and the corresponding KPIs for testing the use cases in general. In summary, this deliverable provides a brief update on the deployment of the first iteration of the FAIRWork DAI DSS. The first iteration has clearly shown that the DAI-DSS is a useful tool for future decision making in Flex and CRF.}, howpublished = {https://fairwork-project.eu/deliverables/D5.2_DAI-DSS_InfrastructureAndSetupReportAtUseCaseSite_v1.0_preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d4.2, title = {Initial DAI-DSS Prototype - D4.2}, author = {Herwig Zeiner, Lucas Paletta, Michael Schneeberger, Gustavo Vieira, Higor Rosse, Rui Fernandes, Sylwia Olbrych, Alexander Nasuta, Magdalena Dienstl, Christian Muck, Rishyank Chevuri }, editor = {Gustavo Vieira}, url = {https://fairwork-project.eu/deliverables/D4.2_Initial%20DAI-DSS%20Prototype%20v1.0a-preliminary.pdf}, year = {2023}, date = {2023-12-29}, abstract = {This deliverable, “D4.2 – Initial DAI-DSS Prototype”, is the first reference implementation focusing on the synthesis of fundamental components to establish a shared comprehension of the operational principles behind DAI-DSS. It aims to describe how the different DAI-DSS components fit together, forming a complete and cohesive structure in which the functionalities of the blocks are well-defined, the complementarity of the structure is explored, and the technological component is described in terms of fulfilling its purposes and integrating with the neighbouring structures that are part of the architecture. The first part of this documentation deals with the description of the building blocks that form part of the architecture described in deliverable “D4.1 – DAI-DSS Architecture and Initial Documentation and Test Report“. Initially, a quick overview of what the building blocks are and a brief explanation of their general structure are given. We then move on to the integration of the DAI-DSS User Interface (UI) into the system. Taking Industrial Partner CRF's Workload Balance Use Case Scenario as a reference, an overview of the UI in this setting is provided. It describes how the UI offers the possibility of visualising the important components directly linked to decision-making in this example, as well as the added value offered by this building block. The UI is essential for efficient decision-making. It provides the appropriate interface between the decision-maker and the system in order to enable a clear visualisation of the decision-making process. Some images provide an introductory view of the UI in this initial prototype. Next is the DAI-DSS Orchestrator. The Orchestrator plays a very important and central role in the architecture of this system. Its purpose is to provide coordination in the exchange of information and decisions between the various parts of the system. Like the conductor in an orchestra, the DAI-DSS carries out the decision-making process, and the involved individual services retrieve the relevant data from the knowledge database. This currently takes place in two parallel ways. At a more advanced stage is the Workflow-based Orchestrator. In this orchestrator, a workflow engine offers the possibility of executing workflows, i.e., the execution of tasks that follow a well-defined path leading to the triggering or realisation of an AI service that offers a recommendation for decision-making on issues identified in the business process for which the system is built. For demonstration purposes, a workflow was built in which worker data is retrieved, and the order of a part is requested in the Workload Balance Use Case Scenario. The step-by-step execution of the orchestration is documented in order to elucidate how the orchestration of the building blocks can take place in a real, concrete case from the manufacturing industry in this example. It presents the DAI-DSS Configurator, a tool designed to enhance decision support systems through efficient configuration and integration frameworks. It explains the configurator's components: the Configuration Framework, which assists in creating decision models and strategies, and the Configuration Integration Framework, which generates system configurations from these models. The report details the prototype's configuration steps, illustrated for clarity. It also allows for microservices and workflow configuration, featuring a user-friendly interface with a wizard for UI components combination. The DAI-DSS Knowledge Base is highlighted as a central data repository, storing user properties, sensor data, and processed data. It plays a key role in the system's data flow, integrating with the Configurator and using REST API for data retrieval. The report covers the integration of AI services using REST-API endpoints, emphasising ease of access and industry-standard practices like Swagger documentation. It explores various algorithms, including Neural Networks and Decision Trees, and the integration of human factors data for assessing worker resilience. Deployment details of the AI services are discussed, focusing on hosting, management, and future enhancements like data catalogues and decision pattern services for resource allocation and predictive maintenance. Finally, the report addresses the extension of the DAI-DSS with new services for various industrial use cases. It discusses customising services through conceptual modelling and the advantages of rule-based approaches over machine learning methods. The services aim to optimise resource allocation, improve productivity, and address human factors, thereby enhancing the system's decision-making capabilities in complex industrial scenarios. }, howpublished = {https://fairwork-project.eu/deliverables/D4.2_Initial%20DAI-DSS%20Prototype%20v1.0a-preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d5.1, title = {DAI-DSS FAIRWork Knowledge Base at Use Case Site - D5.1}, author = {Aleassandro Cisi, Roland Sitar, Marlene Mayr, Magdelena Dienstl, Rishyank Chevuri}, editor = {Rishyank Chevuri}, url = {https://fairwork-project.eu/deliverables/d5-1/D5.1_DAI-DSS%20FAIRWork%20knowledge%20base_v1.0-preliminary.pdf}, year = {2023}, date = {2023-06-30}, urldate = {2023-06-30}, abstract = {This document is Deliverable D5.1, which focuses on the FAIRWork Knowledge Base at the Use Case Site. The deliverable encompasses the results of Task T5.1 “Modelling FAIRWork for Production Processes” and Task T5.2 “Creating FAIRWork Knowledge Base”. The document needs to be seen in the context of the FAIRWork objective to provide a decision support system that (a) integrates digital twins to optimize the overall production process according to multiple parameters, and (b) democratizes decision making granting human worker and machines a say during decision making. The project combines (a) model-based approaches to transparently design, simulate and improve decision making, (b) a co-creation laboratory using models and physical experiments as communication media to all actors and, (c) reliability indication of data and AI algorithms. This deliverable serves as a public demonstration of the project's progress. It provides a concise overview of the work completed in Task T5.1, which involved creating decision process models for two Use Case providers, FLEX and CRF. These models cover various use case categories such as "Automated Test Building," "Worker Allocation," "Machine Maintenance After Breakdown," "Workload Balance," "Delay of Material," and "Quality Issues." These models incorporate details about the production environment, decision-making processes, and involved actors. In this deliverable, Chapter 2 corresponds to the Task 5.1 and provides a comprehensive overview of the methodology employed for decision process modeling in the FAIRWork project, as well as the tools utilized to implement these models. Additionally, there is information provided on how to access these models for demonstration purposes. Section 2.1 delves into the methodology of decision process modeling within the specific context of the FAIRWork project. It outlines the steps and approaches taken to develop effective decision models that can enhance decision-making processes. This section provides insights into the iterative nature of the modeling process, emphasizing the need for clear understanding of a problem, identification of concrete decisions, and relevant decision parameters and aspects. Section 2.2 focuses on the tools utilized to implement the decision process models in the FAIRWork project. It discusses the technological infrastructure leveraged to create these models effectively. The section highlights the utilization of modeling techniques, such as BPMN 2.0, to ensure consistency and compatibility across different decision trees and scenarios. Furthermore, Section 2.3 and 2.4 provide an overview of the decision process models developed specifically for the end users of the FAIRWork project, FLEX and CRF. These sections also present the relevant data inputs to facilitate decision-making. Concurrently, task T5.2 focused on collecting and storing the data and decision models required for each Use Case scenario to establish a common Knowledge Base. The Knowledge Base is built on internationally available open standards, with the ISO 10303 (STEP) standard for data exchange serving as the foundation for the developed repositories. The Knowledge Base provides REST APIs that enable data access and exchange. Furthermore, the deliverable includes the demonstration of the Knowledge Base through figures presented in chapter 3, showcasing the preliminary project setup, and data accessing methods using REST API and also providing information on the demonstrator videos that are made to show the data accessing and sharing capabilities of the Knowledge Base. The contents of this deliverable will contribute to "D5.2 DAI-DSS Infrastructure and Setup Report at Use Case Site" and "D5.3 Demonstration Report of FAIRWork with DAI-DSS", scheduled for M20 and M36, respectively. These subsequent deliverables will further adapt the Knowledge Base to suit the specific needs of the use cases.}, howpublished = {https://fairwork-project.eu/deliverables/d5-1/D5.1_DAI-DSS%20FAIRWork%20knowledge%20base_v1.0-preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d3.1, title = {DAI-DSS Research Specification - D3.1}, author = {Herwig Zeiner, Lucas Paletta, Michael Schneeberger, Gustavo Vieira, Sylwia Olbrych, Alexander Nasuta, Johanna Werz, Noushin Gheibi, Stefan Böschen, Magdalena Dienstl, Marlene Mayer, Christian Muck}, editor = {Sylwia Olbrych}, url = {https://fairwork-project.eu/deliverables/D3.1_DAI-DSS%20Research%20Specification-preliminary.pdf}, year = {2023}, date = {2023-06-20}, abstract = {This report focuses on the deliverable “D3.1 – DAI-DSS Research Specification”, part of the Horizon Europe project FAIRWork. The deliverable aims to describe the specific research factors in selected use cases of industrial partners FLEX and CRF. It presents a research strategies and factors catalogue that serves as a framework for conducting research within the Democratized AI-based Decision Support System (DAI-DSS). DAI-DSS research specifications are closely related to deliverables “D2.1 Specification of FAIRWork Use Case and DAI-DSS Prototype Report” and “D4.1 DAI-DSS Architecture and Initial Documentation and Test Report”. The first part of the report provides an overview of the relevant literature related to the research intended within the frame of this project. It covers the most significant research domains, such as the democratization of decision-making and digital shadows and twins for human experts. Additionally, it explores technical approaches like Artificial Intelligence (AI) and Multi-Agent System (MAS) crucial for improving Decision Support Systems (DSS). This section also presents the state-of-the-art crucial aspects of today's technology, particularly reliability and trustworthiness in AI. The output of this literature review leads to research questions in multiple domains addressed within the FAIRWork project. The second part of the report focuses on the research methodologies and strategies employed to investigate the technical aspects of decision-making processes, human aspects, and digital human factors measurements. It presents research approaches for successfully implementing AI and MAS-based technologies into DSS. Methods such as data-driven modelling, prototyping, and testing are proposed within the AI and MAS domains. Additionally, the report outlines the use of sensors to capture critical information about humans' mental, affective, and motivational states, including implementation details of the Intelligent Sensor Box (ISB). Furthermore, a novel framework using Personas as Human Digital Twins for Decision Making in the context of Industry 5.0 is described. The final part of the report presents the key research factors identified in the industrial use cases and potential AI services to address them. These research factors are categorized into two main perspectives: the human perspective and the technical perspective. The human perspective factors are derived from the research plan and are observed in given use cases. On the technical side, the requirements for modelling and testing new concepts using AI and MAS technologies primarily focus on data availability (process-relevant and expert knowledge) and the DSS architecture necessary to enable information flow and decision models related to the use cases. The report also provides a strategy for communication and dissemination in the context of the research methodology of the FAIRWork project. The objective is to continuously disseminate project achievements, raise awareness about the project, and gather feedback to improve the created research artefacts.}, howpublished = {https://fairwork-project.eu/deliverables/D3.1_DAI-DSS%20Research%20Specification-preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d4.1, title = {DAI-DSS Architecture and Initial Documentation and Test Report - D4.1}, author = {Remi Lanza, Rishyank, Herwig Zeiner, Lucas Paletta, Michael Schneeberger, Robert Woitsch, Magdalena Dienstl, Jose Barbosa, Gutavo Vieira, Higor Rosse, Sylwia Olbrych, Alexander Nasuta, Christian Muck}, editor = {Rishyank Chevuri}, url = {https://fairwork-project.eu/deliverables/D4.1_DAI-DSS%20Architecture%20v1.0a-preliminary.pdf }, year = {2023}, date = {2023-02-28}, urldate = {2023-02-28}, abstract = {This deliverable “D4.1 – FAIRWork Architecture and initial Documentation and Test Report” for is the first description of the architecture, which will be updated in M20 and M30 in form of updated deliverables. First the high-level architecture is mapped to the use case scenarios that are for FLEX “Automated Test Building”, “Worker Allocation” and “Machine Maintenance After Breakdown” as well as for CRF “Workload Balance”, “Delay of Material” and “Quality Issues”. Based on the process analysis described in more detail in “D2.1 - Specification of FAIRWork and Initial DAI-DSS Architecture” we identified generic challenges to the aforementioned use case scenarios. This ensures that the architecture and the implemented solutions will not exclusively fit to those use cases but also serve similar use cases that are out of the project. The challenges are: “finding similar projects”, “find relevant experts”, “simulate production process”, “allocate worker”, “map workers with profiles”, “find similar problems”, “reschedule production line”, “allocate order to production line”, “assess the impact”. Such challenges are targeted by Microservices that may use Artificial Intelligence (AI) when appropriate. Hence, we propose a list of services that either individually or in a cooperative manner target the aforementioned list of challenges. The architecture distinguishes therefore between (a) the core components that allow the selection, configuration, deployment, and operation of selected (AI) services and (b) a list of (AI) services the user can select and compose a solution that targets the requirements of the particular use case. The (a) cores components are (i) the user interface that is a framework enabling user interface widgets to be deployed in environment like web-pages of MS TEAMS, (ii) the orchestrator that is a framework enabling the controlling of individual services or the orchestration of services, (iii) the knowledge base that integrates data from legacy applications, sensor data streams that require no protection and sensor data stream that require protection in form of an Intelligent Sensor Box (e.g. for human-related information); (iv) the externa data asset marketplace complements the data coming from inside the use case with available data form outside, (v) the configurator that enables the selection and configuration of services to a use-case specific solution and finally (vi) the so-called AI-enrichment which is a service catalogue providing different AI-based services that fulfill the end users’ needs. The (b) list of (AI) services is a collection of available services, commercial products and research prototypes that are partly used from outside the consortium where feasible and partly created during the project. Hence, we currently propose an initial list of services that target the aforementioned challenges and propose different AI realizations to demonstrate the flexibility of our core components and to enable a selection of appropriate AI solutions. This list is therefore seen as indicative, and it is expected that in the duration of the project it will evolve. The final set of services will also be provided as projects results in the so-called innovation shop. An initial plan for testing and reporting procedures is presented, which is currently seen as a plan and will be detailed once the deliverable will be updated in M20. The updated version of the deliverable includes the details of implemented components and services and updated security implementations. as well as implemented architecture testing and reporting methodologies.}, howpublished = {https://fairwork-project.eu/deliverables/D4.1_DAI-DSS%20Architecture%20v1.0a-preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d8.1, title = {FAIRWork Exploitation Tools (Website, Social Media, Flyer) - D8.1}, author = {Wilfrid Utz, Christian Muck, Patrik Burzynski, Michael Car}, editor = {Christian Muck}, url = {https://fairwork-project.eu/deliverables/D8.1-FAIRWork%20Exploitation%20Tools_1.0_preliminary.pdf}, year = {2023}, date = {2023-02-28}, abstract = {This report is the accompanying document for deliverable 8.1 (D8.1), in which communication and exploitation tools were created and instantiated for the FAIRWork project. They will be used to communicate information during the project runtime and support exploitation of created artefacts within and after the project runtime. Additionally, the document discusses the FAIRWork innovation shop, which is not yet available, but will be established in D8.2. The following list provides an overview of the tools, which are discussed in this document and includes important links to where they can be found: - Website: https://fairwork-project.eu/ - Flyer: https://zenodo.org/record/7677298/files/FAIRWork_Flyer_1.0.pdf?download=1 - Brochure: https://zenodo.org/record/7673832/files/FAIRWork_Brochure_1.0.pdf?download=1 - Social Media: + LinkedIn: https://www.linkedin.com/company/fairwork-project/ + Twitter: https://twitter.com/fairwork_eu + YouTube: https://www.youtube.com/@fairwork_eu - Zenodo: https://zenodo.org/communities/fairwork/ - Webinars: + Publicly available webinar recordings on YouTube: https://www.youtube.com/playlist?list=PLDKnDRTHllZrGrXZsiePXmvylV1gHvh9K + Event subpage of the webpage (including the webinars): https://fairwork-project.eu/events/ - Innovation Shop: https://fairwork-project.eu/innovation-shop/ (future link)}, howpublished = {https://fairwork-project.eu/deliverables/D8.1-FAIRWork%20Exploitation%20Tools_1.0_preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} } @techreport{fairwork_d2.1, title = {Specification of FAIRWork Use Case and DAI-DSS Prototype Report - D2.1}, author = {Aleassandro Cisi, Roland Sitar, Wilfrid Utz, Christian Muck, Patrik Burzynski, Robert Woitsch, Magdalena Dienstl, Marlene Mayr, Remi Lanza, Rishyank Chevuri, Sylwia Olbrych, Johanna Werz, Alexander Nasuta, Stefan Böschen, Noushin Gheibi, Higor Rosse, Lucas Paletta}, editor = {Herwig Zeiner}, url = {https://fairwork-project.eu/deliverables/D2.1_Specification%20of%20FAIRWork-v1.0a-preliminary.pdf}, year = {2022}, date = {2022-02-28}, urldate = {2022-02-28}, abstract = {Deliverable D2.1 specifies the design requirements for the first relevant stage in the Horizon Europe project FAIRWork aiming at fair decision making within complex systems in the production domain. From the planned use cases of the industrial partners - FLEX “Automated Test Building”, “Worker Allocation”, “Machine Maintenance After Breakdown” and CRF “Workload Balance”, “Delay of Material” and “Quality Issues” - the report constitutes and specifies the basic design decisions for research and development directions in the project. A major contribution is the provision of the initial architecture of an innovative service framework for decision support systems. The first part of the report is dedicated to the design thinking approach as an efficient choice for the analysis of user requirements. The planned use case scenarios were primarily examined from a user-driven perspective. In several workshops a participatory process was applied in order to deduce the most substantial design decisions from the results of the highly interactive sessions. A model-based approach of the design process was chosen in order to determine the user requirements that were consequently defined and described in detail. In this way, the knowledge about a use case was externalised in the form of conceptual representations, using domain-specific modelling languages that are suitable and provide the required construct for representation and processing. In a further step, the high-level scenarios were designed in a collaborative, interactive, and agile environment involving experts from different backgrounds. Processes are the outcome of this structured approach requiring support for: “finding similar projects”, “find relevant experts”, “simulate production process”, “allocate worker”, “map workers with profiles”, “find similar problems”, “reschedule production line”, “allocate order to production line”, “assess the impact”. In the second part, we give an overview about the key challenges of FAIRWork. In the current production industry there is a need to make the current automated and hierarchical structured production processes more flexible. At the same time digitalization with AI support is seen as a key enabler for more energy efficient and resource efficient services, products or business models, by also enabling process optimization in the overall production process. Therefore, we describe in more detail the main challenges technical challenges such as configuration, resource allocation, and selection aspects. These three challenges are highly relevant for making the process more flexible, adaptive, and resource efficient by using the relevant AI-based decision strategies in our complex distributed decision-making. At the end, the trustworthy AI aspect is a further key challenge to get AI accepted by the involved humans and also utilize its potential. In the third part, the overall methodology of making complex decision-making is outlined. Within this chapter, we describe the overall procedure for implementing complex decision-making processes. Therefore, this chapter gives an overview about relevant concepts for the research direction and implementation of such complex decision making by using AI services. In FAIRWork, AI is used in all our scenarios to automate processes or to make their processes more resource-efficient. Since humans are an important part of the overall decision process, trust in AI and human factors plays an essential role, therefore these aspects are explained. Finally, the technical concepts for a concrete implementation such as digital knowledge base, digital twin, digital shadow, will be discussed as well. Finally, it follows the explanation about the orchestration of decision-making processes by using Microservices. In the fourth part, the initial architecture of the project is presented based on the overall project objectives and requirements. Key components of the FAIRWork service framework are motivated, described, and their relevant features are presented. A detailed description of these components is given in Deliverable D4.1 including the technical implementation of the basic core services or application specific services. Finally, the initial design of the FAIRWork’s architecture is compared with most relevant technical architectures that are commonly used in the industry environment domain, such as, Gaia-X, FIWARE, International Data Space, and RAMI. }, howpublished = {https://fairwork-project.eu/deliverables/D2.1_Specification%20of%20FAIRWork-v1.0a-preliminary.pdf}, keywords = {deliverable}, pubstate = {published}, tppubtype = {techreport} }