Abstract
Industries are becoming more and more digitized to better implement intelligent and predictive maintenance support systems, aligned with Industry 4.0, which requires the progressive digitization of data collection and processes. Maintenance interventions, in an evolving technological context, are increasingly more complex and difficult for technicians to perform. In these environments, the use of Augmented Reality (AR) to help assist and guide in the maintenance operations, can accomplish a considerable gain in productivity. AR allows to superimpose information objects in real scenes, such as text, images, audiovisuals, and 2D/3D model animations, making available contextual information about the process, based on location and perspective. This paper describes the design and implementation of a prototype augmented reality application to support maintenance tasks inside a metal stamping production unit, that produces components for the automotive sector. It aims to train and guide personnel during the maintenance operations, and offering an extra channel to reach expert help.

Abstract
In the nowadays highly unstable manufacturing market, companies are faced, on a daily basis, with important strategic decisions, such as "does the company has the necessary capacity to accept a high volume order?" or "what measures need to be implemented if the product demand increases x% a year?". Decision-makers, i.e. company's managers, rely on their experience and insights supported by classical tools to take such decisions. Classical mathematical solvers or agent-based systems are typical architectural solutions to implement strategic planning tools to support decision-makers on this important task. Within the ARUM (Adaptive Production Management) project, a hybrid strategic planning tool was specified and developed, combining the optimization features of classical solvers with the flexibility and agility of agent systems. This paper briefly presents such architecture and focuses on the generation of the "what-if game" mechanism to support the generation of more intelligent and dynamic planning solutions. © 2015 IEEE.

Abstract
This paper presents a platform for adaptive production management developed in the ARUM1 (Adaptive pRodUct Management, http://arumproject.eu/) project. The design of ARUM platform started with applying a traditional enterprise Service-Oriented Architecture (SOA) paradigm to solving an integration problem for the production ramp-up of highly customized products such as aircrafts, ships, etc. The production of such articles is exceptionally challenging for planning and control, especially in small lot sizes. Often requests for changes at any stage of the production, immature products and processes bring serious additional risks for the producers and customers. To counter such issues requires new strategies, the core elements of most of them include early detection of unexpected situations followed by rapid mitigation actions. Furthermore, human beings cannot cope any longer with processing a massive volume of data that comes with a high velocity from various sources that is a requirement for any modern production shop floor. The traditional IT solutions also fall short when trying to satisfy all those requirements and this motivates the need for ARUM platform to help in effective decision making. © Springer International Publishing Switzerland 2015.

Abstract
The 4th industrial revolution is marked by the use of Cyber-Physical Systems (CPSs) to achieve higher levels of flexibility and adaptation in production systems that need to cope with a demanding and ever-changing market, driven by mass customization and high quality products. In this context, data analysis is a key technology enabler in the development of intelligent machines and products. However, in addition to Cloud-based data analysis services, the realization of such CPS requires technologies and approaches capable to effectively support distributed and embedded data analysis capabilities. The advances in Edge Computing have promoted the data processing near or at the devices that produce data, which combined with Multi-Agent Systems, allow to develop solutions based on distributed and interacting autonomous entities in open and dynamic environments. In this sense, this paper presents a modular agent-based architecture to design and embed cyber-physical components with data analysis capabilities. The proposed approach defines a set of data processing modules that can be combined to build cyber-physical agents to be deployed at different computational layers. The proposed approach was applied in a smart inspection station for electric motors, where agents embedding data analysis algorithms were distributed among Edge, Fog and Cloud layers. The experimental results illustrated the benefits of distributing the data analysis by different computational layers. © 2019, Springer Nature Switzerland AG.

Abstract
The Cyber-Physical System (CPS) paradigm promotes the decentralization and distribution of the logic control as well as the integration of cyber and physical counterparts. In parallel, self-organization allows the dynamic and automatic system re-configuration responding to condition and environment changes. Modeling and simulation assume a crucial importance in the design of such complex, distributed, and self-organized systems, in the way that the detected and debugged errors may be corrected before the deployment into the real system, as well different strategies can be tested and evaluated. Agent-based modeling tools are computational frameworks able to analyze, experiment and compare systems populated by cooperative agents, supporting the fast prototyping of agent-based solutions exhibiting self-* properties. In this paper, the NetLogo tool was used to model and simulate the agent-based control layer of a small scale CPS, which control uses self-organization principles. © 2017 IEEE.

Abstract
Industrial Cyber-Physical Systems (CPS) are promoting the development of smart machines and products, leading to the next generation of intelligent production systems. In this context, Artificial Intelligence (AI) is posed as a key enabler for the realization of CPS requirements, supporting the data analysis and the system dynamic adaptation. However, the centralized Cloud-based AI approaches are not suitable to handle many industrial scenarios, constrained by responsiveness and data sensitivity. Edge Computing can address the new challenges, enabling the decentralization of data analysis along the cyber-physical components. In this context, distributed AI approaches such as those based onMulti-agent Systems (MAS) are essential to handle the distribution and interaction of the components. Based on that, this work uses a MAS approach to design cyber-physical agents that can embed different data analysis capabilities, supporting the decentralization of intelligence. These concepts were applied to an industrial automobile multi-stage production system, where different kinds of data analysis were performed in autonomous and cooperative agents disposed along Edge, Fog and Cloud computing layers.