There was an error in the original publication [...]
Artificial Intelligence and Machine Learning rely on large, high-quality datasets for accurate and robust models, yet data scarcity remains a major challenge, especially in smart farming. Phenology modeling, a key application, studies how plant biological events relate to climate and seasons. Accurate phenology models improve crop quality, support climate adaptation, and guide decisions such as pesticide use and harvesting, enhancing environmental and economic sustainability. However, agricultural data are highly diverse and heterogeneous, complicating model development. This study presents a proposed georeferenced dataset for Machine Learning-based grapevine phenology prediction across 3 Protected Designations of Origin in Aragón, Spain. Developed by a multidisciplinary team, the dataset combines 9 datasets from 8 sources (including meteorological time series, field phenology observations, and Copernicus Sentinel-2 multispectral imagery) covering the period 2016–2022. It supports both physical and Machine Learning-based phenology modeling and facilitates knowledge extraction in agronomy and plant biology. Its relevance lies in its comprehensive scope, the inclusion of 9 phenological stages, and a rigorous methodology ensuring reproducibility. This framework enables the creation of similar datasets for other regions or crops, advancing smart farming through scalable, data-driven solutions. The open publication of the code further supports this objective. We further anticipate its potential contribution to developing foundation models as well as to the creation of new knowledge in biology and agronomy.
The automotive industry is subject to major transformation initiated by societal and economical pull (reducing emissions, zero fatalities, European competitiveness) and accelerated by technology push (electrification, Cooperative, Connected and Automated Mobility (CCAM), and Cooperative Intelligent Transport Systems (C-ITS)). Following this trend, the Software-Defined Vehicle (SDV) targets the integration of software (SW) development methodologies for vehicle development as well as the value delivery shift toward customers along the entire lifecycle. It promises to create benefits for the car manufacturers in terms of faster time to market, easier update – as well as for the car users (private persons, fleet operators) in terms of personalized user experience, upgradability. At the same time, SDV requires a much more integrated and continuous development framework to enable different experts to efficiently develop and validate concurrently the different parts of the vehicles, to gather information about real operation, and to support update in the field. This paper introduces the collaborative development framework introduced in the European research program Collaborative Development Framework for electric-based Software-Defined Vehicles (CODE4EV).
Metal Additive Manufacturing (AM) offers unique design opportunities but is still limited by the high cost and time of process development, where distortion, residual stress, and geometric inaccuracies remain critical challenges. This work introduces an integrated methodology that combines finite element simulations, surrogate modelling, and experimental 3D printing to enable efficient and accurate process prediction. High-fidelity thermo-mechanical simulations are validated through benchmark twin-cantilever builds, and their results are used to generate reduced-order surrogate models. These models maintain prediction accuracy while reducing computational time by several orders of magnitude compared to full simulations. The framework captures the influence of key parameters such as laser power, scan speed, and preheating temperature, providing fast and reliable predictions of part deflection. The proposed approach establishes how physics-based surrogate models can accelerate process optimization, reduce costly trial-and-error iterations, and pave the way for digital twins that support robust qualification of metal AM components.
Food loss and waste remain the most pressing global sustainability challenges with enormous environmental, social, and economic consequences. Addressing this issue requires not only technological innovation but also a deeper understanding of how innovations diffuse and interact across supply chains. This study develops a multi-agent diffusion simulation model, calibrated with empirical data and stakeholder inputs, to assess three innovations across different product chains from 2025 to 2050. Innovations such as smart packaging for meat, AI-driven demand forecasting for fish, and AI-based quality for fruit, enabling the quantification of waste, costs, and greenhouse gas emissions under varying adoption pathways.The results show that innovation's impacts are powerfully context dependent. Smart packaging in the meat chain brings steady waste reductions of 10–15% in 2050, primarily by extending shelf life and reducing expiries. AI demand forecasting for fish achieves the most transformative outcomes, reducing waste by up to 27% together with marked decreases in costs and emissions. AI quality recognition for fruit offers smaller yet valuable gains, shifting consumer acceptance and improving retailer sell-through imperfect produce. Faster and broader adoption leads to larger benefits, highlighting the need for supportive policies, incentives, and consumer engagement to improve diffusion.By integrating diffusion theory with supply chain simulation, this study offers methodological and applied insights, serving as a decision-support tool for stakeholders to strategically deploy innovations and inform policy design. It contributes to the evidence required to achieve the EU's 2030 food waste reduction targets and promote sustainable food systems.
