Abstract
In this study, we compared different survival models using machine learning (ML) techniques to assess the impact of enriching survival models with features extracted from topological data analysis (TDA). Using a clinical dataset on lung transplantation, we compared the Cox model, random survival forests (RSF), and the DeepSurv deep neural network, along with their TDA-enriched versions. TDA uses persistent homology and generates descriptors that capture the global and multiscale structures present in the data that do not appear in Euclidean spaces. The results show that integrating these descriptors improves predictive performance, increasing the C-index and enhancing risk discrimination. Thus, the integration of TDA and ML emerges as a promising alternative for modeling survival in complex biomedical data. © IFIP International Federation for Information Processing 2027.

Abstract
Highlights What are the main findings? Envelope methods (convex hull and alpha shape) are generally more sensitive to point density loss than voxel-based grids, which maintain a relative stability, although they were not always the closest to field-based volume estimations. Methods parameters (alpha and voxel size) influence accuracy and should be adapted to point cloud density, canopy structure, and growth stage. What are the implications of the main findings? UAV photogrammetry provides dense, low-cost 3D canopy data suitable for vineyard monitoring at the row or plant level. Multi-temporal 3D measurements can support vineyard management and integration with decision support systems.Highlights What are the main findings? Envelope methods (convex hull and alpha shape) are generally more sensitive to point density loss than voxel-based grids, which maintain a relative stability, although they were not always the closest to field-based volume estimations. Methods parameters (alpha and voxel size) influence accuracy and should be adapted to point cloud density, canopy structure, and growth stage. What are the implications of the main findings? UAV photogrammetry provides dense, low-cost 3D canopy data suitable for vineyard monitoring at the row or plant level. Multi-temporal 3D measurements can support vineyard management and integration with decision support systems.Abstract Vegetation volume is a useful indicator for assessing canopy structure and supporting vineyard management tasks such as foliar applications and canopy management. The photogrammetric processing of imagery acquired using unmanned aerial vehicles (UAVs) enables the generation of dense point clouds suitable for estimating canopy volume, although point cloud quality depends on spatial resolution, which is influenced by flight height. This study evaluates the effect of three flight heights (30 m, 60 m, and 100 m) on grapevine canopy volume estimation using convex hull, alpha shape, and voxel-based models. UAV-based RGB imagery and field measurements were collected during three periods at different phenological stages in an experimental vineyard. The strongest agreement with field-measured volume occurred at 30 m, where point density was highest. Envelope-based methods showed reduced performance at higher flight heights, while voxel-based grids remained more stable when voxel size was adapted to point density. Estimator behavior also varied with canopy architecture and development. The results indicate appropriate parameter choices for different flight heights and confirm that UAV-based RGB imagery can provide reliable grapevine canopy volume estimates.

Abstract
In the context of transport networks, synchronization techniques refer to a set of optimization approaches whose output is the timetables of different lines obtained through the minimization of transfer waiting time and vehicle bunching. Accordingly, synchronization techniques can improve the overall quality of transfers in transport systems by making them smoother for passengers. The synchronization of transport lines at strategic locations improves the system's overall connectivity and integration and, ultimately, its attractiveness to passengers. This work aimed to synchronize strategic transport lines connecting Porto to other cities, having Porto as the only focal point. The morning peak hours were selected as the analysis period. A synchronization model already proposed in the literature was adapted and applied to this case study. The results include the comparison between the real and synchronized timetables and the sensitivity analysis of the solutions obtained by changing headway and time window parameters.

Abstract
Purpose This study aims to explore how technology-enabled services influence the overall experience of participants in running events by applying a structured service design methodology. Specifically, it examined how recreational runners engage with technology-enabled services throughout the customer journey of a running event, and how the application of the MINDS method contributes to enhancing the runners' experience. Design/methodology/approach Thirty-nine running event participants were interviewed to explore their experiences. The interviews took place in Greece in 2023, across various mass-participation events from marathons to 5K city races. Using the Management and INteraction Design for Service (MINDS) method, qualitative data were thematically analyzed. Findings The study identified how recreational runners interact with technology-enabled services across the pre-, during-, and post-event stages. Using the MINDS method, participants' experiences were mapped to reveal emotional touchpoints, service gaps, and opportunities to enhance the event experience. These findings were translated into service design proposals through the MINDS method, resulting in visual outputs that illustrate how technology-enabled services could be better integrated across the event journey. Originality/value This study is among the first to examine running event experiences from the participants' perspective using a service design methodology. It also contributes to the advancement of the MINDS by introducing customer journey and emotional journey extensions, offering richer insights into how participant experiences can be optimized across the event lifecycle.

Abstract
This paper aims to clarify the characteristics of Digital Twins (DTs) in their most advanced conceptual development, Cognitive Digital Twins (CDTs), and analyze their support for the implementation of the Circular Economy (CE). A systematic literature review was conducted using a specially developed five-dimensional analytical framework to characterize DT proposals and their potential for CE based on an established framework for circularity strategies. The study indicates that cognitive and hybrid DT approaches tend to cover high levels of interoperability, data flow, system levels, and cognitive processes. However, CDT use in CE demands harmonizing different strategies to cover the complete product lifecycle, which recent research on DTs has not fully addressed. This study is the first to systematically review cognitive digital twins and their relation to circularity, offering an analytical framework that can be expanded for future research in various application areas of Industry 5.0.

Abstract
Synthetic data has become a crucial enabler for training deep learning models in robotic perception tasks. This work presents a configurable synthetic data generation pipeline, built on BlenderProc, that leverages CAD model annotations to produce part-based labels directly at the geometry level. The proposed approach enables the automatic generation of Oriented Bounding Box (OBB) annotations and supports the definition of multiple regions of interest through a vertex-level annotation tool. The pipeline is validated in an warehouse environment use case involving a storage container for automotive parts for autonomous forklift operations. Experimental results show that YOLO11m-Oriented Bounding Box model trained exclusively on synthetic data achieve strong performance on real-world images, reaching up to 0.768 mAP@0.5:0.95. Furthermore, increasing synthetic data diversity consistently improves detection performance, and hybrid training with a small amount of real data provides an effective trade-off between annotation effort and accuracy. © 2026 IEEE.