Microscopic active particles exhibit adaptive behavior under real-world conditions with reinforcement learning.

The application of reinforcement learning often faces limitations due to the exploration phase, which can be costly and risky in various contexts. This is particularly evident in manufacturing industries, where the training phase of a reinforcement learning agent is constrained, resulting in suboptimal performance of developed strategies. To address this challenge, digital environments are typically created, allowing agents to freely explore the consequences of their actions in a controlled setting. Strategies developed in these digital environments can then be tested in real scenarios, and secondary training can be conducted using hybrid data that combines digital and real-world experiences.In this chapter, we provide an introduction to reinforcement learning and showcase its application in two different manufacturing scenarios. Specifically, we focus on the woodworking and textile sectors, which are part of ongoing research activities within two distinct European Research Projects. We demonstrate how reinforcement learning is implemented in a digital context, with the ultimate goal of deploying these strategies in real systems.

Research on sustainable business models (BMs) rooted in the circular economy is expanding in industry and academia, encouraging companies to enhance their impact on profits, people, and the planet. However, developing sustainable BMs is complex due to the conflicting objectives of sustainability dimensions and competition with traditional models. This work addresses this challenge by proposing a Sustainable Business Guide (SBG) supported by artificial intelligence (AI) to assist in decision-making. The SBG supports the design and assessment of business models by integrating inspection; business opportunity exploration; technical, economic, and environmental analyses; an AI-based decision support system; and Porter’s Five Forces, focusing on the aeronautical and wind energy sectors.

Industry 5.0 (I5.0) is an evolution of the fourth Industrial Revolution (Industrie 4.0) aiming at a more human-centric, resilient, and sustainable manufacturing. This proposed chapter aims at illustrating the decisive contribution of ADR (AI Data Robotics) technologies in the development, implementation, and demonstration of Industry 5.0 principles and in particular of its sustainable and circular manufacturing pillar. In the domain of EC-funded projects (Horizon and Digital Europe programs, HEP and DEP), several initiatives have been recently addressing the I5.0 sustainability/circularity challenge from different viewpoints: the technology perspective in HEP ADRA partnership; the application perspective in HEP, Made in Europe (MiE) partnership; and the deployment perspective in DEP initiatives such as data spaces, AI TEFs (Testing Experimentation Facilities), and EDIH (European Digital Innovation Hubs). The analysis has been conducted along three main dimensions: a Technology dimension, where data, AI, and robotics convergence is enabling unprecedented innovations regarding sustainability and circularity for EU manufacturing industries; a Project dimension, to demonstrate the fundamental role of EU-funded research-innovation-deployment actions to create, nurture, and diffuse an Industry 5.0 mindset among researchers; a Pilot dimension, where significant industrial players are showing the improvement of their global competitiveness, thanks to Industry 5.0. Main conclusions regard (1) the fundamental role in Industry 5.0 of the robotics and automation physical world and the need for edge-to-cloud data and AI continuum; (2) the confidentiality and high strategic value of the data exchanged in Industry 5.0 value networks and the need for open, secure, and interoperable data infrastructures (data spaces); and (3) the competitiveness and resilience of manufacturing supply chains as a top priority for Europe and the need for integrated ADR technologies with methods and tools for environmental impact assessment and materials-components-products circularity promotion. The BDVA (Big Data Value Association) Smart Manufacturing Industry (SMI) group is a business-technology community that studies, analyzes, and promotes the impact of data spaces, AI, and digital twins to the manufacturing industry. It encompasses more than 130 experts representing more than 60 organizations involved in dozens of EC-funded projects. The content of this chapter originates from the last years of workshops, discussions, and webinars inside the SMI group, where experts, projects, and industrial pilots actively participated in the definition of the role of ADR technologies in the several different manufacturing business processes according to Industry 5.0 principles.

The current dataset is composed of a collection of High Entropy Alloys (HEAs). It contains the alloy composition, the number of chemical elements (No), the phase in a simple form (S_Phase), where 4 classes of phases were considered, namely amorphous (AM), intermetallic (IM), solid solution (SS), and solid solution + intermetallic (SS+IM). It contains also a second phase column (Phase), where we added the type of phase present in alloys with SS and repeated the S_Phase entry for the other cases. We have calculated 13 design parameters (see their definition below) used to design HEAs, known as the parametric approach. Finally, a set of columns containing the chemical elements and their corresponding fraction in the alloy is included. This dataset was developed in the framework of the European project ACHIEF for the discovery of novel materials to be used in industrial processes. Mean atomic radius a (Å) \(a = \displaystyle\sum_{i=1}^{n} c_i r_i\) Atomic size difference δ \(\delta = \sqrt{\displaystyle\sum_{i=1}^{n} c_i \bigg(1 - \dfrac{r_i}{a} \bigg)^2}\) Average melting temperature T_m (K) \(T_m = \displaystyle\sum_{i=1}^{n} c_i T_{mi}\) Average melting temperature standard deviation (K) \(\sigma_{T_m} = \sqrt{\displaystyle\sum_{i=1}^{n} c_i \bigg(1 - \dfrac{T_{mi}}{T_m} \bigg)^2}\) Mixing enthalpy ΔH_mix (kJ/mol) \(\Delta H_{mix} = 4 \displaystyle\sum_{i \neq j} c_i c_j H_{ij}\) Mixing enthalpy standard deviation (kJ/mol) \(\sigma_{\Delta H_{mix}} = \sqrt{\displaystyle\sum_{i \neq j} c_i c_j (H_{ij} - \Delta H_{mix})^2}\) Ideal mixing entropy S_id (R)* \(S_{id} = \Delta S_{mix} = -R \displaystyle\sum_{i=1}^{n} c_i \ln c_i\) Electronegativity χ \(\chi = \displaystyle\sum_{i=1}^{n} c_i \chi_i\) Electronegativity difference in a multi-component alloy system \(\Delta\chi = \displaystyle\sqrt{\sum_{i=1}^{n} c_i(\chi_i - \chi)^2}\) Valence electron concentration VEC \(VEC = \displaystyle\sum_{i=1}^{n} c_i \cdot VEC_i\) Valence electron concentration standard deviation \(\sigma_{VEC} = \sqrt{\displaystyle\sum_{i=1}^{n} c_i (VEC_i - VEC)^2}\) Mean bulk modulus K (GPa) \(K = \displaystyle\sum_{i=1}^{n} c_i K_i\) Bulk modulus standard deviation (GPa) \(\sigma_{K} = \sqrt{\displaystyle\sum_{i=1}^{n} c_i (K_i - K)^2}\) where n is the number of components in the alloy system, c_i is the stoichiometric ratio, r_i is the atomic radius, T_mi is the melting temperature, χ_i is the Pauli electronegativity, VEC_i is the valence electron concentration, and K_i is the bulk modulus for the i-th component of the alloy. H_ij is the binary mixing enthalpy in the liquid phase, and R is the gas constant. *Note: the ideal mixing entropy S_id units in the dataset appears as kJ/mol, but they should be written in terms of the gas constant R, e.g., the compound Ag₂Al has S_id = 0.636 R, where R = 8.314 J · K⁻¹ · mol⁻¹.