Applied System Innovation doi: 10.3390/asi9070137

Authors: Maurício de Oliveira Gondak Vinicius Moretti Cleiton Hluszko Diego Alexis Ramos Huarachi Fabio Neves Puglieri Antonio Carlos de Francisco

A recurring challenge in the development of information systems (ISs) across complex organizational domains is the lack of integration and alignment between strategic, tactical, and operational levels, resulting in methodological fragmentation that constrains traceability, innovation, and organizational value generation. This study proposes and applies to the Integrated Innovation Framework for Information System Development (IIF-ISD) to overcome this gap. The research was structured through a systematic literature review, following the PRISMA and ROSES protocols, and validated through an exploratory single-case study involving the development of an IS supporting the Selo Casa Azul (SCA) environmental certification process in a Brazilian construction company, a context chosen for its multi-level organizational complexity and ESG compliance requirements, representative of broader certification IS development challenges. The framework integrates DSRM, agile methodologies, Design Thinking, and Lean Startup through three governing principles—Hierarchical Embedding, Functional Complementarity, and Traceability by Design—achieving cross-level alignment between strategic objectives, tactical performance monitoring, and operational execution. Empirical evaluation (n = 9; 14 weeks) yielded SUS scores of 76.8–82.1/100, a 76% reduction in data entry error rates, and a 78% stakeholder engagement rate, providing initial support for the framework’s practical effectiveness.

Applied System Innovation doi: 10.3390/asi9070136

Authors: Supattra Puttinaovarat Aekarat Saeliw Siwipa Pruitikanee Jinda Kongcharoen Jariya Seksan Attaporn Wangpoonsarp Thidapath Anucharn Niti Iamchuen

Wildfires pose serious risks to ecosystems, air quality, and human health. Effective wildfire monitoring requires accurate detection and timely validation, but current approaches are often constrained by fragmented data sources, false alarms, and delays in field verification. This study presents WildfireGO, a multi-source wildfire detection and validation system that integrates crowdsourced observations, satellite hotspot data, and image-based classification in a geospatial monitoring environment. The system combines user-submitted images, Sentinel-2 imagery, and Moderate Resolution Imaging Spectroradiometer (MODIS) hotspot data processed through Google Earth Engine (GEE) to support wildfire detection and verification. Four classification models, namely Convolutional Neural Network (CNN), Random Forest (RF), K-Nearest Neighbors (KNN), and Gradient Boosting (GB), were evaluated using 10-fold cross-validation and an independent test dataset of 800 wildfire-related images. The CNN model produced the best result, with an accuracy of 97.5% on the independent test dataset. By combining image-based classification with crowdsourced reporting, the system helps screen user-submitted wildfire information and reduce false detections. Satellite-derived hotspot data provide spatial evidence for cross-checking reported events and improving spatial situational awareness for wildfire monitoring and response planning. WildfireGO supports near real-time data submission, automated processing, and interactive map-based visualization through a web-based interface. The findings indicate that combining crowdsourced reports, satellite observations, and image classification in a single geospatial system has the potential to support more reliable wildfire detection and provide practical support for environmental monitoring, disaster response, and spatial decision-making.

Applied System Innovation doi: 10.3390/asi9070135

Authors: Katherine Cabana-Jiménez Vladimir Sousa Santos John E. Candelo-Becerra Zaid García Sánchez Fredy E. Hoyos

Economic dispatch in grid-connected microgrids is challenged by the variability of renewable generation, the uncertainty of demand, and the need to simultaneously satisfy technical and economic constraints under different operating conditions. This study proposes an integrated predictive economic dispatch strategy for power grids with interconnected microgrids, structured as a unified optimization framework. The approach integrates nodal electrical modeling, Optimal Power Flow (OPF)-based optimization, multi-scenario analysis, and post-optimization feasibility verification based on performance indicators within a single decision-support structure. The methodology is applied to a modified 14-node power grid interconnected with a microgrid, where simulations are conducted under three representative load scenarios (100%, 70%, and 40%) and two operational configurations (hybrid and renewable-only), enabling a comprehensive assessment of system behavior. Results show that the hybrid configuration consistently outperforms the renewable-only case, achieving loss reductions of up to 7.3 MW, increases in spinning reserve exceeding 50 MW, and a transition from net power import to export of approximately 50 MW under high demand. Additionally, the microgrid plays an active operational role, dynamically switching between import and export modes based on load levels and the generation mix. The proposed framework enables identification of operationally efficient and technically feasible configurations by incorporating bidirectional power exchange, electrical constraints, and reserve requirements. The main contribution lies in integrating technical, operational, and interaction variables within a single deterministic Optimal Power Flow (OPF)-based assessment scheme to support decision-making in interconnected microgrid-based power grids.

Applied System Innovation doi: 10.3390/asi9070133

Authors: Yazeed S. Jweihan

Machine learning has become a field of growing interest in asphalt pavement engineering, spanning mix design, material characterization, performance prediction, distress detection, sustainability, quality control, and maintenance planning. However, a lack of transparency can undermine engineering trust, defensibility, and field implementation. This systematic scoping review aims to synthesize explainable artificial intelligence (XAI) and interpretable machine-learning applications for asphalt pavement materials and systems, following the PRISMA-ScR guidelines. Major scientific databases were used to identify relevant peer-reviewed studies, which were screened against a set of inclusion and exclusion criteria and categorized into seven research dimensions. A final library of 163 publications was compiled, comprising 73 core evidence studies and 90 supporting references. The review covers techniques such as SHAP, LIME, partial-dependence analysis, attention mechanisms, surrogate models, sensitivity analysis, symbolic modeling, and physically informed interpretation. The use of XAI in performance prediction, material-property interpretation, and modeling for mix design is well developed, while distress/damage analysis, life cycle sustainability, field validation, uncertainty-aware explanation, maintenance decision support, and human-centered evaluation are still relatively underdeveloped. The main contribution is a five-layer framework linking data provenance, model performance, explanation quality, physical plausibility, and decision utility. The review proposes moving from post hoc feature ranking to validated, physically centered, uncertainty-aware, and engineer-in-the-loop decision support for asphalt XAI.

Applied System Innovation doi: 10.3390/asi9070134

Authors: Tao Xu Xiang Jin Lei Liu Song Zhang Jianzhou Zhang Wei Wang

This paper proposes an intelligent time-series early warning framework based on a production LSTM–Transformer network for petrochemical refining processes. A cascaded encoder–decoder architecture is designed, where the LSTM extracts local temporal patterns and medium-term memory from noisy industrial data, while the Transformer models global dependencies and cross-unit interactions via multi-head self-attention. An adaptive feature fusion layer bridges the representational gap between the two networks. A multi-stage preprocessing pipeline tailored for refining MES data handles missing values, outliers, and mixed operating conditions. Using 120 variables from five units of a fluid catalytic cracking unit, the framework predicts the regenerator bed temperature up to 8 h (48 steps) ahead. Comparative experiments show that the production LSTM–Transformer achieves a mean MAE of 0.088, a mean RMSE of 0.113, and the lowest median MAPE of 19.91% among all models, outperforming standalone LSTM (MAE 0.095, MAPE 20.85%) and Transformer (MAE 0.088, MAPE 20.49%). Robustness analysis confirms stable performance under strong noise (down to 5 dB) and missing rates up to 50%, with a median MAE of 0.1027 across tags. This work provides an effective, end-to-end predictive early warning solution that balances accuracy, production importance coverage, and industrial robustness, offering a generalizable data-driven paradigm for process industries.

Applied System Innovation doi: 10.3390/asi9070132

Authors: Muhammed Hakan Yorulmuş Hür Bersam Sidal

Predictive quality systems are central to Industry 4.0 manufacturing, yet detecting rare defects that pass established quality gates remains an open problem. This study addresses post-quality-gate failure detection in automotive brake manufacturing, where 310 faulty units (1.20%) among 25,756 production records create a naturally occurring extreme class imbalance of 1:82. Fault labels are derived from warranty reports and linked to multi-station production line measurements, while negative samples may include latent failures, motivating a recall-focused evaluation. We propose a Quality 4.0 machine learning framework that compares five resampling methods (ADASYN, SMOTE-Tomek, KMeans-SMOTE, CTGAN, and TVAE) plus a no-resampling baseline across 24 classifiers and stacking ensembles. In total, 504 configurations are tested on a held-out test set. The proposed SVM-RBF model trained on ADASYN-augmented data achieves recall of 0.871, specificity of 0.982, balanced accuracy of 0.926, and ROC-AUC of 0.952, producing only 93 false positives (FPR = 1.8%). Stacking ensembles provide alternative operating points maximizing the detection rate (93.5%) and a separate operating point with the highest discrimination capacity (ROC-AUC = 0.975). Feature importance analysis through Permutation Importance and SHAP identifies Force Increment as the leading feature under both attribution methods. Friedman and Wilcoxon tests confirm statistically significant differences among strategies. The framework offers a practical way to add predictive capability to existing quality control systems.

Applied System Innovation doi: 10.3390/asi9060131

Authors: Yiming Cao Longchuan Li Yitong Xue Jiaxin Liu Zhongkui Wang

Snake-like robots require body adaptation during locomotion when creeping through environments with obstacles. Central Pattern Generator (CPG) provides an effective way to generate rhythmic signals through parameter modulation. During body-shape adaptation, the body wave generated by the CPG can be modified by adjusting its parameters. In this paper, a CPG network based on Hopf oscillators is adopted, and the amplitude parameter is used for body-shape adaptation. However, the influence of amplitude variation during the transition process has not been fully understood. More specifically, when the amplitude parameter changes abruptly, the attractor shifts immediately, while the oscillator state cannot follow the new attractor instantaneously. This mismatch produces transient responses and waveform distortion during the transition process. To address this issue, a linear parameter transition method is introduced. The proposed method is subsequently extended to a coupled CPG network for controlling the multi-joint snake-like robots. Simulations are conducted under different parameter transition conditions. The results demonstrate that the parameter transition method strongly affects the transient torque response. Compared with abrupt parameter variation, the proposed linear transition method significantly reduces transient torque peaks. Additionally, the results further show that even a short transition interval is sufficient to achieve most of the torque reduction. Experiment results show that the proposed method can be applied to body-shape modulation and obstacle avoidance during snake-like robot locomotion.

Applied System Innovation doi: 10.3390/asi9060130

Authors: Piotr Ściegienka Łukasz Wróbel Daniel Dąbrowski Marcin Michalak Dawid Macha Marek Sikora Tomasz Borowik Tomasz Hartwig

Fouling on ship hulls increases hydrodynamic drag, fuel consumption, and emissions. This, in turn, necessitates the development of efficient methods for side cleaning and inspection. This work focuses on the application of image-based classification to assess the cleanliness of the surface of the hull in robotic cleaning systems, with respect to the ISO 8501-4 standard. Due to limited data availability, transfer learning techniques using pre-trained convolutional neural networks (ResNet50, EfficientNetB0 and MobileNetV2) were used. Both end-to-end models and hybrid approaches that combine deep feature extraction with XGBoost (version 3.2.0) classification were evaluated. Experiments were carried out on binary classification (cleaned vs. uncleaned surfaces) and multi-class classification of cleanliness levels (WA1, WA2, WA2.5). The results show that transfer learning enables effective recognition of cleaning status, achieving high performance for binary classification despite a small dataset. However, multi-class classification remains challenging due to subtle differences between classes and data limitations. The proposed approach supports automated visual inspection of underwater robotic platforms and represents a step toward objective standards-based assessment of hull cleaning processes.

Applied System Innovation doi: 10.3390/asi9060129

Authors: Andrés Morán-Durán Guillermo Cortés-Robles Omar Juárez-Rivera Mónica Karina González-Rosas Jesús Delgado-Maciel José Roberto Grande-Ramírez

The growing environmental crisis, particularly water pollution from detergents, necessitates a shift from reactive compliance to proactive eco-innovation, as current methods often fail to systematically resolve trade-offs between performance, safety, and ecology. This study develops and illustrates the application of the Evolutionary-Driven Design Framework (EDDF), an integrated methodology that combines PESTEL analysis, historical evolutionary pattern analysis, Quality Function Deployment (QFD) with a novel contradiction index, Theory of Inventive Problem Solving (TRIZ), and environmental assessment. The framework was applied to redesign a conventional laundry detergent with the objectives of zero phosphates, superior biodegradability (>85%), maintained efficacy, and controlled cost. The quantitative contradiction index matrix prioritized critical unsustainable parameters (e.g., EDTA, Cocamide DEA) for substitution over mere optimization. Through an iterative feedback loop, the process evolved from a biobased concentrate to an “enzymatic power tablet” (Concept B). This waterless, solid formulation uses sodium citrate as a biodegradable builder and an encapsulated multi-enzyme system, achieving an estimated >90% biodegradability and zero phosphates while meeting technical and economic targets. The EDDF provides a structured, anticipatory roadmap that transforms regulatory and market pressures into drivers of innovation, offering companies a promising method for designing sustainable products by proactively resolving contradictions and avoiding historical mistakes.

Applied System Innovation doi: 10.3390/asi9060128

Authors: Hakan Gules Muhammet Garip

This paper presents the design and experimental validation of a self-contained rotating Halbach array—based demonstrator for electrodynamic suspension (EDS) systems. The proposed platform was developed to bridge the gap between conventional externally powered laboratory testbeds and large-scale EDS vehicles by enabling investigation of levitation behavior under realistic onboard mass and subsystem integration constraints. The system integrates rotating circular Halbach arrays, onboard power supply, sensing, motor control, and structural support within a single levitated architecture. Experimental validation was conducted under a constrained one-degree-of-freedom configuration allowing vertical motion only. The system achieved stable levitation of a 35 kg platform and supported additional payloads approaching a 1:2 ratio relative to the baseline mass, while maintaining air-gap stability within approximately ±0.1 mm. The experimental results further reveal that the operational limit of the system is governed by actuation power and current constraints rather than electromagnetic levitation capability, highlighting a key distinction between self-contained and externally powered EDS systems. The proposed demonstrator provides a compact and practical experimental platform for the validation and performance evaluation of Halbach-array-based EDS systems. In addition, the study presents practical engineering insights regarding payload distribution, actuator saturation, structural integration, and system-level design constraints relevant to future self-contained EDS platforms and control-oriented levitation systems.

Applied System Innovation doi: 10.3390/asi9060127

Authors: Qiuyan Zhang Yuanbiao Zhang Junye He Junyi Li

The design of cushioning packaging for flat-screen television (TV) products relies heavily on repeated simulations, resulting in high development costs and low design efficiency. In this study, we propose a hybrid framework integrating finite element (FE) simulation, data augmentation and interpretable machine learning (ML) for rapid peak acceleration prediction and optimization of TV cushioning packaging. First, a total of 216 FE drop-impact simulation samples of TV cushioning packaging systems were generated using ANSYS Workbench, covering TV dimensions, liner type, liner density, liner thickness, drop height and peak acceleration. Mixup-based data augmentation and Bayesian optimization were then employed to develop and tune six ML models. All ML models trained on the original dataset achieved coefficients of determination (R2) ranging from 0.797 to 0.990. The Mixup-augmented XGBoost model achieved the best prediction performance, yielding R2 values of 0.998 and 0.983 for the training and testing datasets, respectively. SHAP analysis revealed that liner material type, liner density and liner thickness were the dominant factors affecting the protective performance of TV cushioning packaging. In addition, a web-based platform was developed based on the proposed FE–ML strategy to support the design exploration of feasible schemes for new TV products. The predictive capability of the proposed FE-ML framework was further evaluated using 22 independent cushioning packaging schemes, achieving an R2 of 0.926 and an average prediction error of 4.490 g. These results suggest that the proposed workflow can support the performance evaluation and optimization of TV cushioning packaging.

Applied System Innovation doi: 10.3390/asi9060126

Authors: Weitao Yang Wuqing Song Huan He Zhiliang Hu Yonghong Li

This research investigates atmospheric neutron-induced single event effects (SEEs) on advanced artificial intelligence (AI) chips during natural environment operation. The RK3588 neural processing unit (NPU) is the evaluated target chip, and its SEE is assessed through a combination of irradiation testing and software fault injection. During the irradiation test, the chip was exposed to a spectrum neutron at the China Spallation Neutron Source. Upon reaching a cumulative fluence of 8.25 × 109 n·cm2, a total of 14,018 soft errors were detected, of which 99.97% manifested as variations in target recognition accuracy and network inference latency. Among these variations, both detrimental effects (reduced target recognition accuracy or prolonged network inference time) and beneficial effects (enhanced target recognition accuracy or shortened network inference time) caused by single event effects were observed. In addition, atmospheric neutron single event effects were found to cause NPU operation suspension and system crashes. Based on the irradiation test results, failure predictions for neural processing units in real-world environments were estimated, and mitigation recommendations were proposed. Furthermore, software fault injections were employed to conduct in-depth analysis of detected soft errors during irradiation testing. This research provides support and references for the reliable application of artificial intelligence chips in natural environments.

Applied System Innovation doi: 10.3390/asi9060125

Authors: Huiyong Yi Yong Qin

Complex products continue to face low iterative-design efficiency and poor cross-generation data compatibility, while existing product-gene research is still constrained by the predominance of qualitative approaches, ambiguous representations of hierarchical associations, and insufficient standardization. Based on the principles of decomposition and reconstruction and the systems thinking of genetic engineering, this study develops a generic three-level framework for product genes at the platform, assembly, and component levels. Hierarchical mapping functions and parameter-constraint equations are introduced to enable quantitative representation, and a quantitative product-gene information system is established, including a core-parameter quantification model and inter-/intra-level association-strength models. By integrating multiple international standards, the study further constructs a tripartite standardized description system covering metadata, semantics, and format, and proposes a mathematical mapping method from product information to standardized formats. A case study of Company A’s Platform B and Concept Vehicle C shows that the association-strength model achieves the required adaptation threshold, thereby validating the proposed framework. This study provides quantitative theoretical support for the platform-based and intelligent development of complex products and offers an implementable technical solution for product-gene reuse and data sharing, particularly in the new energy vehicle industry.

Applied System Innovation doi: 10.3390/asi9060124

Authors: Jinbao Song Jiahui Cai Yijun Wang Kai Wang Shiwen Cui Nuo Xu

In recent years, short video platforms have rapidly developed into important media for cultural dissemination. The interactions of netizens in short video comment sections not only reflect their focus on cultural content but also contain rich emotional attitudes. However, given the vast and fragmented nature of comment data, accurately extracting keywords, identifying cultural themes, and analyzing sentiment tendencies pose significant challenges in understanding netizens’ cultural perceptions. To address these challenges, this study proposes a text analysis framework that integrates keyword extraction, clustering analysis, and sentiment analysis to explore the core topics and emotional characteristics of cultural dissemination in short video comment sections. Firstly, to address the challenge of balancing statistical information and semantic understanding in short-text keyword extraction, this paper proposes the TF-IDF-KeyBERT Integrated Algorithm (TKIA) keyword extraction algorithm, which integrates Term Frequency–Inverse Document Frequency (TF-IDF) and Key Bidirectional Encoder Representations from Transformers (BERT). Experiments on the CSL dataset demonstrate improvement in the F1@5 metric, showing its potential to enhance keyword extraction performance for short texts. Secondly, to address the difficulty of simultaneously considering semantic representation capability and clustering flexibility in short-text clustering analysis, this paper designs the Self-Supervised Contrastive Enhanced Clustering (SCEC) algorithm by integrating self-supervised contrastive learning with a soft clustering strategy. Compared to baseline methods, SCEC improves clustering accuracy (ACC) by 17.5% on AGNews and 6.8% on THUCNews, suggesting a more effective way to reveal the underlying structure of cultural topics. Finally, to address the challenge of effectively leveraging both text structural information and global semantic features in short-text sentiment analysis, this paper develops the BERT-GCN Cross-Attention (BGC) Model, integrating BERT embeddings and Graph Convolutional Network (GCN)-based structural features via a Cross-Attention mechanism. On the My_weibo_senti_100k dataset, the BGC model achieves a 2.45% increase in Macro-F1 and a 2.41% improvement in accuracy over strong baselines, offering its ability for high-precision modeling of user sentiment. This study offers effective data support and technical pathways for applications such as cultural content understanding, personalized recommendation, and user emotion guidance.

Applied System Innovation doi: 10.3390/asi9060123

Authors: Yusuf Oluwasola Omoloja Titilayo Ogunwa Romeo Marin Marian Javaan Chahl

Predictive Maintenance Digital Twin (PdMDT) is an emerging cyber-physical technology with potential to revolutionise industrial maintenance. Despite this gradual prominence, the foundational infrastructure required to stage-by-stage develop the technology is lacking. Researchers have developed reference architecture (RA), information and functional requirements and the Model Based Systems Engineering (MBSE) for systematic development. These interventions have not addressed the stage- by-stage systematic development of PdMDT. Other gaps like the conflation of PdMDT with Predictive Maintenance (PdM), lack of formal definition of the PdMDT, inconsistent nomenclature, and partial and non-scalable implementations continue to define the field. In this research, these gaps are addressed through a systematic literature review which reveal the quality of aggregated studies from three databases—IEEE Digital Explore, Scopus, and the Web of Science, and a comparative and critical analysis of PdM and PdMDT from which several gaps in literature are discovered. The Primary contribution of this investigation is the development of a 9-stage life-cycle developmental framework for the PdMDT which necessitated the clarification of the inconsistent PdMDT terminologies, clarification of the various PdMDT implementations and a formal definition for PdMDT. All these interventions are necessary for the realisation of PdMDT at an industrial scale.

Applied System Innovation doi: 10.3390/asi9060122

Authors: Mengqi Liu Hanyu Xia Xinshuo Li Ying You Leyi Zhou

At present, although there are many SCARA manipulator solutions with vertical lifting functionality, they generally suffer from high maintenance costs and complex structures. Moreover, systematic performance evaluations based on international standards are lacking, leading to unclear critical performance boundaries such as accuracy and payload in practical applications. To address these issues, this paper designs and manufactures a low-cost SCARA manipulator for educational and research demonstrations as well as light-duty electronic parts assembly scenarios. A “leadscrew + stepper motor” scheme is adopted for vertical lifting, and an Arduino Mega 2560 development board serves as the core controller, significantly reducing system cost. A three-dimensional model is established using SolidWorks 2022, and kinematic simulations are carried out with MATLAB 2024a to preliminarily verify the feasibility of the mechanism. Subsequently, a physical prototype is built and experimental tests are conducted in accordance with the ISO 9283 standard. The experimental results show that the repeatability of the manipulator is controlled within the range of 0.05–0.3 mm, the path deviation caused by vibration lies between −0.52 mm and 0.3 mm, and the maximum payload capacity is 3.91 N. These experimental data can serve as a benchmark for the design and performance comparison of similar low-cost manipulators.

Applied System Innovation doi: 10.3390/asi9060121

Authors: Hao Yuan Yiting Wang Jiawei Zhong

With the rapid development of artificial intelligence and computer technology, digital interfaces have become central to CNC press operations. However, effectively presenting complex information within limited display space while reducing cognitive load remains a key challenge. Based on visual cognition theory, this study proposes a cognition-oriented interface optimization framework for CNC press HMIs. User needs were identified through field studies, questionnaires, and interviews and prioritized using the KANO model. A cognitive model linking user perception and interface visual elements was established to guide the interface optimization. Two groups of interface layouts were developed and evaluated through eye-tracking experiments to identify the optimal layout scheme. The selected interface was further refined through visual optimization and validated using usability testing. The results indicate that the optimized interface significantly improved the visual search efficiency, information readability, task completion efficiency, and overall user satisfaction. The proposed framework provides a systematic approach for cognition-driven interface optimization in industrial HMI systems.

Applied System Innovation doi: 10.3390/asi9060120

Authors: Veena Phunpeng Kreetiwat Chaiyasin Kitsana Khodcharad Wipada Boransan Watcharapong Patangtalo Attaphon Chaimanatsakun

Ensuring the cleanliness of precision components is critical in Hard Disk Drive (HDD) manufacturing, where microscopic dust contamination on the Voice Coil Motor Assembly (VCMA) can lead to positioning errors, unstable head movement, and long-term reliability failures. However, automated inspection of such contamination remains challenging because dust particles are extremely small, visually irregular, and often appear under complex microscopic backgrounds. This study presents an explainable hybrid deep learning framework for microscopic dust inspection by integrating object detection for precise localization and image classification for defect confirmation. Three YOLO architectures, namely YOLOv5, YOLOv8, and YOLOv11, were comparatively evaluated for dust detection, while three convolutional neural network (CNN) models, ResNet50, EfficientNetB0, and MobileNetV2, were implemented using transfer learning with frozen feature extraction layers for Good (G) and Not Good (NG) image-level classification. The experimental dataset consisted of annotated microscopic VCMA images, with data augmentation applied to the training subset to mitigate limited sample size and class imbalance. Experimental results showed that YOLOv8 achieved the strongest overall aggregate detection performance, whereas YOLOv5 was selected as the preferred detector for subsequent hybrid integration because it produced fewer false positives under reflective and textured microscopic backgrounds. YOLOv11 exhibited lower detection performance in the present setting, likely due to its architectural characteristics being less suited to the limited-data and high-background-complexity conditions of this study. In the present experimental setting, YOLOv5 achieved mAP@0.5 = 0.62, precision = 0.75, and recall = 0.69. For image-level classification, EfficientNetB0 achieved the highest classification accuracy of 93.10%, with F1-score = 0.932 and AUC = 0.986. In addition, Grad-CAM visualizations demonstrated that EfficientNetB0 consistently focused on physically meaningful dust-contaminated regions, thereby enhancing the interpretability of the classification results. Overall, the proposed hybrid framework integrating YOLOv5-based localization with EfficientNetB0-based defect confirmation showed promising potential for improving inspection reliability, false-alarm control, and explainability in automated VCMA quality inspection. These findings support the feasibility of explainable deep learning for microscopic defect inspection in HDD manufacturing and suggest its potential applicability to other precision manufacturing environments.

Applied System Innovation doi: 10.3390/asi9060119

Authors: Matthias Jäger Jörg Eberhardt Douglas W. Cunningham

Shoes are increasingly being bought online without being put on in person as internet shopping gains popularity. As a result, returns have increased significantly, which has had negative effects on the economy and the environment. Numerous technologies are available to measure foot size precisely at home or in-store in order to address this problem. People can identify their perfect shoe size and avoid needless returns by taking accurate foot measurements. A single image should be enough to measure the foot in order to make the system as easy as feasible for the user. This is accomplished by using point clouds from one side of the foot, which are produced by capturing a depth image. In order to optimise the reconstruction of partial data, this study investigates the impact of the acquisition position of a single partial foot scan on reconstruction quality and measurement accuracy when a state-of-the-art network is employed for completion. To this end, task-specific partial foot datasets were created with varying camera positions and foot orientations to determine the optimal conditions for depth map acquisition. Utilising the foot dataset that has been introduced for the purposes of training and evaluation, the network was able to generate accurate reconstructions. These reconstructions allowed for the estimation of shoe size in accordance with the European sizing system. The method is accurate enough in all tested positions to reconstruct a foot with sufficient precision. However, we also identified position 5 in our multi-view setup, which is viewed from a lower angle, as the position that leads to the best reconstruction results. Additionally, advantages were found with input data that show more of the forefoot than the heel area. Therefore, the forefoot provides more information on the overall geometry and should be the focus of single-shot procedures.

Applied System Innovation doi: 10.3390/asi9060118

Authors: David Emiliano Gutiérrez-Leal Edgar León-Sandoval Eduardo Quintana Contreras Liliana Ibeth Barbosa-Santillán

This work presents a case study of a Large Language Model based system for automated classification of student survey responses. The system processes 22,286 open-text responses collected from 2062 students across 12 academic programs and 21 nationalities spanning the years 2010–2025. The system architecture has been deployed on institutional servers for security, while integrating databases, an asynchronous task queue for processing, a web-based service layer, and distributed background workers that interact with remote LLM inference services. This work provides a practical reference framework for educational institutions aiming to responsibly and effectively operationalize LLMs in real-world applications.

Applied System Innovation doi: 10.3390/asi9060117

Authors: Saronsad Sokantika Payakorn Saksuriya Siva Shankar Ramasamy Aniwat Phaphuangwittayakul

Temples are tourist attractions that represent the history and culture of Thailand, especially in Chiang Mai province—a city with a rich history that has become a prominent destination attracting visitors from around the world. Many temples remain undiscovered yet are ready for tourists to visit; however, due to unfamiliarity, tourists tend to visit only the well-known temples, as other visitors do, missing great opportunities to engage with new cultural heritage tourism experiences. To address this issue, we propose a Hybrid Three-Stage Route Planning Recommendation (HTS-RPR), a novel method for tourist route planning that delivers recommended routes based on tourists’ preferred constraints. This model contains three-stage route recommendations providing an optimal single-day route with mandatory and recommended points of interest (POIs) through a metaheuristic integrating Mixed Integer Programming (MIP), heuristic-based POI recommendation filtering, and Genetic Algorithm route optimization with Bayesian reward and peak-time awareness, ensuring that users can effectively travel cultural routes with high popularity and satisfaction while avoiding attractions during periods of high traffic. To validate the efficacy of the proposed model, experiments with three baseline methods were conducted. The results demonstrate that HTS-RPR achieves the best fitness score in 55 out of 60 scenarios and the best reward in 54 out of 60 scenarios, with a median fitness score 28.34% and 103.67% higher than the Genetic Algorithm and Multi-Start Simulated Annealing baselines, respectively, and a median total reward exceeding all three baselines by up to 40.74%. Although HTS-RPR’s median execution time is approximately 2.6 times that of the Genetic Algorithm, it remains 84.5% faster than the Multi-Start Simulated Annealing baseline, offering a favorable trade-off between solution quality and computational cost. Moreover, the framework’s pluggable reward function enables destination managers to configure recommendation priorities, including the promotion of undiscovered tourist attractions, while the peak-time-aware optimization mitigates congestion at specific POIs.

Applied System Innovation doi: 10.3390/asi9060116

Authors: Khaled Chahine Hassan N. Noura

Accurate prediction of the long-term MPPT degradation trajectory of perovskite solar cells (PSCs) from short-term measurements can significantly reduce the time required for material characterization. Although conditional diffusion models have recently been introduced for degradation prediction in energy devices, their applicability to PSC-specific maximum power point tracking (MPPT) degradation trajectory forecasting remains uncertain due to the complexity of the underlying dynamics. This study benchmarks three approaches using 2245 devices from a publicly available dataset: NHITS, a hierarchical multilayer perceptron (MLP) with direct multi-horizon regression; Probabilistic NHITS (P-NHITS), which utilizes the same architecture with multi-quantile output; and TimeDiff, a conditional diffusion model with a CSDI backbone, autoregressive initialization, mode conditioning, and classifier-free guidance. The results indicate that PSC degradation under controlled conditions is predominantly single-exponential, with device-specific decay rates identifiable within the first 30 h. Therefore, the forecasting task is most appropriately framed as a regression problem rather than a generative one. NHITS achieves a root mean squared error (RMSE) of 0.738 PCE% compared to TimeDiff’s 0.863 (a 17% increase, p < 10−15), despite TimeDiff incorporating all architectural advantages reported in the literature. P-NHITS matches deterministic accuracy (0.744 PCE%) while providing 77% coverage prediction intervals without sampling, which is closer to the nominal 80% target than TimeDiff’s 63% coverage from 50 DDPM samples. For T90 (the time at which PCE first falls below 90% of its reference value) lifetime prediction restricted to forecast-window crossings, NHITS achieves a mean absolute error (MAE) of 16.2 h, outperforming TimeDiff’s 22.5 h. For smooth, unimodal degradation processes, direct regression with quantile outputs is both sufficient and preferable to conditional diffusion. Model selection should be guided by the underlying physical processes rather than by methodological trends.

Applied System Innovation doi: 10.3390/asi9060115

Authors: Pablo López-Márquez Jessica Zaqueros-Martinez Bruno Ramos-Cruz Francisco José Quesada-Real Mercedes Rodriguez-Garcia

Traditional academic degrees often fail to capture the full range of competencies students acquire throughout their university education, particularly those developed through laboratory activities, internships, volunteering, and other extra-curricular experiences. This limitation hinders students’ ability to differentiate themselves in increasingly competitive labor markets and complicates employers’ identification of candidates with balanced technical and transversal competencies. To address this challenge, this paper presents a design-oriented research study proposing a Self-Sovereign Identity (SSI)-based framework for the decentralized issuance and verification of academic micro-credentials. The proposed approach combines a structured methodology for generating micro-credentials with a decentralized architecture supported by a prototype implementation based on SSI technologies. The framework enables universities, lecturers, and other trusted entities to issue verifiable and tamper-resistant credentials that students can securely manage, control, and share through SSI wallets. Unlike existing approaches, which typically focus either on secure credential infrastructures or on the pedagogical value of micro-credentials, the proposed framework integrates both technological and educational perspectives while explicitly supporting the certification of extra-curricular and soft skills. The system supports the creation of granular and portable competency profiles while enhancing transparency, authenticity, interoperability, and trust in credential management. Furthermore, the paper discusses key challenges associated with large-scale adoption, including trust management, governance, scalability, interoperability, and issuer credibility. The results suggest that SSI-based micro-credentialing represents a promising approach for improving the recognition of both technical and transversal competencies, contributing to better alignment between higher education outcomes and evolving labor market demands.

Applied System Innovation doi: 10.3390/asi9060114

Authors: Dinara Tamabay Nurlan Temirbekov Ayauzhan Seitova Aruzhan Seitova

In this research, surface Au concentration measurements are considered as a spatially correlated geochemical field associated with deep occurrences of disturbing masses using real geological exploration data from the Novo-Khairuzovsky gold deposit in East Kazakhstan. The approach is based on the relationship between potential-field continuation problems and reconstruction of subsurface geological anomalies from surface observations. The considered approaches include Tikhonov and Lavrentiev regularization, SVD, and TSVD. Special attention is given to regularization parameter selection using the L-curve method, Morozov discrepancy principle, and GCV. Comparative computational analysis is performed to evaluate the accuracy, stability, and efficiency of these methods in solving first-kind Fredholm integral equations. Results are assessed using error metrics and spatial visualization of reconstructed fields within a Geographic Information System (ArcGIS), enabling consistent geospatial interpretation. Results show that Lavrentiev regularization with L-curve criterion provides the most stable and reliable reconstruction across all depths, achieving high correlations (R=0.8876 at 100 m and R=0.8049 at 200 m) with low reconstruction errors. Tikhonov regularization performs acceptably at 100 m but becomes less stable at greater depths. Among spectral methods, TSVD improves stability compared with classical SVD, while standard SVD shows weak correlations and larger reconstruction errors due to high noise sensitivity.

Applied System Innovation doi: 10.3390/asi9060113

Authors: Huiping Jin Chenxi Shen Tianshi Lu Yong Ling Feng Gao Kang Han Xiaojun Jin

Conventional industrial manipulators are often costly and come with steep learning curves, which limits their scalability in hands-on robotics education. This paper presents a compact and modular vision-guided sorting platform based on a 4-DOF SCARA robot, designed for rapid assembly, reconfiguration, and beginner-friendly deployment in laboratory courses. A collaborative visual perception strategy is proposed, which introduces a lightweight YOLOv8 algorithm for robust material category recognition, while HSV-based color segmentation and Hough circle localization are utilized to extract sub-pixel centroid features. The pixel measurements are mapped to the robot base frame through an integrated nine-point hand–eye calibration model, and joint commands are generated via a joint-space quintic polynomial interpolation algorithm to ensure continuity and avoid kinematic singularities. The overall system adopts a hierarchical architecture in which the vision host communicates target commands to a motion controller via TCP/IP, while joint actuators are driven through a CAN bus. Feasibility is first verified in a Webots digital prototype with synchronized conveyor and manipulator control, and is then validated on a physical platform equipped with a compliant TPU-based soft gripper to improve grasp tolerance under localization noise. Experiments demonstrate that the system achieves an average recognition accuracy of 98.1% and a mean positioning error of 0.189 mm. The proposed platform provides an extensible testbed for teaching kinematics, perception-to-control integration, and modular robotic system development.

Applied System Innovation doi: 10.3390/asi9060112

Authors: Andrzej Chydzinski

FIFO buffers are widely employed in networking devices to store packets prior to transmission. Their impact on aggregate traffic has been extensively studied and is well documented in the literature. In contrast, significantly less attention has been allocated to the impact of FIFO buffers on individual flows contributing to the aggregate traffic. In this paper, the throughput of each flow traversing a FIFO buffer supplied with complex traffic composed of numerous flows, potentially exhibiting heterogeneous statistical properties, is investigated. A full queuing model of a FIFO buffer fed by many flows with different characteristics is considered first. This model is very precise with respect to each flow, but cannot be solved in practice. Then, a simplification of the full model based on a limiting theorem is proposed. For the simplified model, exact formulae for throughput and loss ratio of each participating flow are derived. In numerical examples, the throughput of flows of diverse types in scenarios with various buffer sizes, buffer loads, and transmission time distributions is calculated. It is also examined, how these factors influence per-flow throughput. Finally, it is demonstrated that in typical scenarios, the results of the simplified model differ by only a few percent from those obtained through simulations of the full, precise model.

Applied System Innovation doi: 10.3390/asi9060111

Authors: Andrés F. Cifuentes-Perdomo Carlos A. Rodado-Grijalba Mauricio A. Vargas-Hernández Lilibeth Aguilera-Pua Rosse M. Villamil-Cañas Jaime A. Restrepo-Carmona Luis A. Fletscher Hernán Felipe García

Public procurement systems are prone to risks such as collusion, contractual concentration, and irregular subcontracting, which undermine transparency and accountability. Traditional fiscal oversight approaches remain largely retrospective, limiting their ability to anticipate irregularities and prevent potential losses. Addressing the gap between theoretical machine learning models and real-world institutional deployment, this study introduces an applied system innovation that integrates two complementary approaches at a national scale: a Contractual Network Model (Mallas Contractuales) and a Predictive Risk Model for Contractors. The first component uses graph-based analytics, employing an Entity–Link–Property schema to represent relationships among entities, contractors, and contracts, thereby enabling the detection of structural patterns associated with collusive or anomalous behavior. The second component implements supervised machine learning models, trained on more than 16 million contracts and 2.6 million contractors from sources such as SECOP, RUES, DIAN, and national sanction registries. Models, including Random Forests and Gradient Boosted Trees, were optimized via cross-validated hyperparameter search and evaluated on a separate hold-out set using ROC AUC and Gini metrics, achieving strong discriminatory performance under the available retrospective validation setting while maintaining operational interpretability. Both approaches were deployed in a modular architecture that integrated Databricks, i2 Analyst’s Notebook, and Power BI dashboards, providing interactive visualizations and risk scores at multiple levels. Together, these systems demonstrate how the convergence of graph analytics and predictive modeling enables proactive fiscal auditing, strengthens institutional capacity, and offers a replicable framework for public sector accountability.

Applied System Innovation doi: 10.3390/asi9060110

Authors: Sarisa Theera-Umpon Jarupichaya Punyakwaw Pornpailin Suwanpitak Nipon Theera-Umpon

Respiratory rate is one of the most important vital signs. It affects ventilation which relates to oxygen inhalation and carbon dioxide elimination. Currently, only a handful of prototypes are available for estimating the respiratory rate under the condition that users remain completely still. This research focuses on the development of a respiratory rate monitoring system that can detect human respiratory signals using force sensitive resistors (FSRs). The FSR sensors measure the forces from respiratory motion and then signal processing techniques are employed to minimize background noise and artifacts. Respiratory data are processed by a microcontroller and transmitted via Bluetooth to a mobile device for further processing and visualization. The system performance was evaluated in three stages. Firstly, for the proof by simulation, a mean absolute error (MAE), root mean square error (RMSE), and Pearson correlation coefficient (PCC) of 0.26, 0.37 breaths per minute (bpm), and 0.9998 are achieved, respectively, even when the noise level is very high, i.e., power signal-to-noise ratio is 0.25 or −6.02 decibel. Secondly, for the test on a robot, the MAEs are 0.25, 0.53, and 0.75 bpm; the RMSEs are 0.28, 0.64, and 0.92 bpm; the PCCs are approximately 1, 0.9993, and 0.9986, respectively, under sitting, walking, and jogging conditions. The system is further deployed on 14 human subjects yielding MAEs of 0.51, 1.24, and 1.92 bpm; RMSEs of 0.65, 1.63, and 2.22 bpm; and PCCs of 0.9893, 0.9831, and 0.9655, for human sitting, walking, and jogging, respectively. In the future, this respiratory rate monitoring system could be applied to patients, elderly individuals, or the general population who experience movement or locomotion during monitoring.

Applied System Innovation doi: 10.3390/asi9060109

Authors: Sheng Li Shihao Gu

Virtual synchronous generators (VSGs) are prone to transient power angle instability and short-circuit current overshoot under symmetrical short-circuit grid faults. To address the limitation that existing transient control strategies fail to simultaneously guarantee power angle stability and fault current limiting, a coordinated control strategy combining dynamic active power reference regulation and adaptive virtual impedance is designed. Specifically, the active power reference is dynamically adjusted in accordance with the voltage sag magnitude at the point of common coupling (PCC), which effectively narrows the acceleration area of the virtual rotor and maintains the transient power angle near its rated value to prevent the risk of system loss of synchronism. On this basis, an adaptive virtual impedance control scheme is designed to accurately calculate and implement the optimal current-limiting impedance on demand, confining the steady-state fault current within the allowable threshold. Finally, the effectiveness of the designed strategy is verified on the Matlab/Simulink simulation platform. Simulation results demonstrate that the designed strategy achieves the coordination between transient power angle stability and fault current limiting, thus improving the operational stability of the VSG grid-connected system under symmetrical short-circuit grid faults.

Applied System Innovation doi: 10.3390/asi9060108

Authors: Luong Duc Long Vo Thi Dinh Khanh Nguyen Quang Trung Truong Ngoc Son

This study proposes an intelligent decision-support framework for alternative selection in complex construction projects, where evaluation processes are affected by uncertainty, multiple decision-makers, and interdependent criteria. The framework integrates the fuzzy group best–worst method with fuzzy TOPSIS into a unified structure that explicitly captures cross-criterion influence effects. First, triangular fuzzy judgments from multiple experts are used to derive criterion weights, while interdependencies among criteria are represented through a fuzzy influence-intensity matrix and incorporated into fuzzy nonlinear optimization models. This process enables the systematic estimation of both independent and interdependency-adjusted criterion weights. Second, the resulting weights are used in a fuzzy ranking procedure to evaluate alternatives according to their relative closeness to fuzzy ideal solutions. To enhance transparency, reproducibility, and practical usability, the proposed method is implemented in Python as an automated computational workflow for decision analysis. Its applicability is demonstrated through a real-world case study on access platform system selection for mechanical, electrical, and plumbing installation in an airport terminal subject to safety, productivity, workspace, and elevation-related constraints. The results show that explicitly modeling criterion interdependencies provides a more realistic evaluation structure and enhances the robustness and reliability of alternative selection in complex construction management contexts.

Applied System Innovation doi: 10.3390/asi9060107

Authors: Jonathan Matthei Johannes Maas Maurice Wischum Sven Mackenbach Katharina Klemt-Albert

Public participation is a central component of democratic decision-making processes, particularly in planning and approval procedures. However, increasing data complexity and the growing number of submitted objections significantly raise the effort required for their review and processing. Against this background, this study developed an AI-supported objection management system that uses a large language model (LLM) to automatically pre-sort objections by topic and generate response suggestions based on historical objection texts from previous infrastructure projects. The aim is to increase efficiency in the processing workflow while maintaining consistent response quality without replacing human decision-making. The prototype development is preceded by a literature review to identify key user requirements and derive relevant use cases. Subsequently, four expert workshops with representatives from German road and rail infrastructure administrations at the state and federal level were conducted to evaluate the prototype. The results indicate significant efficiency potential, particularly through automated thematic pre-sorting of objections. However, topic structures must be adapted to the specific procedure. AI currently mainly serves as supportive pre-processing and requires human review (“human-in-the-loop”). Transparent labeling of AI use is also necessary to ensure traceability and acceptance. The findings will be incorporated into the ongoing development of the prototype within the BIM4People research project funded by the German Federal Ministry of Transport (BMV), with the aim of further improving the system’s functionality and exploring additional applications.

Applied System Innovation doi: 10.3390/asi9060106

Authors: Daniel Sanin-Villa Cristian M. Hernandez Vanessa Botero-Gómez

Automated thermal inspection supports scalable photovoltaic asset management by reducing the subjectivity and limited temporal coverage of manual surveys. This study benchmarks a lightweight machine vision framework for low-resolution infrared inspection of photovoltaic modules using native 24×40 pixel thermal images. Morphological and textural descriptors, namely HOG, LBP, and GLCM, were evaluated with optimized SVM, Random Forest, and XGBoost classifiers under a unified experimental protocol. The HOG + SVMOpt configuration achieved the best performance, with a Macro F1-score of 0.80±0.02 and an average accuracy of 0.80±0.02. The same pipeline maintained an end-to-end CPU latency of 12.45±0.85 ms per image, including preprocessing, descriptor extraction, and prediction. The results indicate that gradient-based structural descriptors provide the most favorable balance between predictive performance and computational cost among the evaluated configurations. The proposed pipeline is therefore presented as an interpretable reference for first-stage thermal screening in low-cost photovoltaic inspection workflows.

Applied System Innovation doi: 10.3390/asi9050105

Authors: Duen-Huang Huang Yu-Cheng Wang

While the use of artificial intelligence (AI) in higher education is widespread, students’ experiences with AI-supported learning in their regular courses remain underexplored. Objective: This research examines the relationships among perceived usefulness, perceived ease of use, and academic engagement among undergraduate students enrolled in AI-supported courses at a Taiwan university. It adopts the Technology Acceptance Model, where learning desire indicates perceived usefulness, and technology self-efficacy indicates perceived ease of use. Methods: The study takes a questionnaire with six dimensions of technology self-efficacy, learning desire, learning methods, learning planning, learning habits, and learning process to evaluate students’ attitudes toward AI-supported learning and their academic engagement. Results: Students’ attitudes toward AI-supported learning were moderate to positive. Multiple regression analysis showed that perceived usefulness was significantly and positively associated with academic engagement, whereas perceived ease of use showed a positive but non-significant association. Implications: Students’ academic engagement is influenced more by how useful AI tools are perceived for learning, rather than by their confidence in using AI tools. This paper enriches the literature on student-centered AI in higher education and gives insights for designing AI-supported courses that integrate AI tools with meaningful learning tasks. Future research can examine larger and more diverse samples and use longitudinal or experimental designs to test how students’ perceptions of AI tools develop over time.

Applied System Innovation doi: 10.3390/asi9050104

Authors: Fang Zheng Yanping Lu Junghee Lee Hongyan Liu Dandan Wang Myun Kim

With the rapid expansion of the smart clothing market, designers face increasing pressure to balance functional performance, material suitability, environmental impact, and development efficiency. Conventional design workflows and rule-based assistance methods often struggle to provide adaptive and data-driven support for multi-constraint decision-making. To address this issue, this study proposes an AI-driven decision support system for sustainable smart clothing design based on a multi-scale dynamic graph convolutional network (MDGCN). The proposed system integrates material properties, environmental indicators, and user-oriented design requirements into a unified decision-support framework and further enhances feature extraction through an attention mechanism. Two datasets, the Wearable Technology Material Properties Dataset (WTMPD) and the Environmental Impact Assessment Dataset (EIAD), were used to validate the model and system effectiveness. Experimental results showed that the MDGCN-based model achieved accuracies of 0.964 and 0.943, with recalls of 0.923 and 0.920 on the WTMPD and EIAD datasets, respectively. In system-level evaluation, the proposed decision support system reduced design time from 120 h to 60 h, improved material selection accuracy to 90.2%, and achieved superior operational performance in terms of resource utilization (77.45%), energy consumption (115.25 kWh), and response time (1.56 s). These results demonstrate that the proposed framework can effectively support complex design decision-making while improving efficiency, sustainability, and adaptability in smart clothing development. The study provides a practical AI-enabled system innovation approach for sustainable smart clothing design by linking flexible material selection, environmental impact prediction, and designer-oriented decision support. In addition, the prototype deployment demonstrates the feasibility of applying the proposed system as a design-stage wearable AI tool for mediating human, technological, and environmental considerations in smart clothing development.

Applied System Innovation doi: 10.3390/asi9050103

Authors: Hamza Jakha Sanae Tbaikhi Souad El Houssaini Mohammed-Alamine El Houssaini Souad Ajjaj

Sentiment analysis has become increasingly important across various domains, particularly in business intelligence, where it is crucial for improving the performance of companies by identifying the sentiments and emotions expressed in customer feedback on products and services. Despite its growing relevance, sentiment analysis still faces several challenges, including class imbalance in datasets, limitations in feature extraction techniques, and the selection of appropriate classification models. Effectively addressing these challenges requires the integration of robust representation methods, reliable data balancing strategies, and efficient classification frameworks. In this study, we propose a novel sentiment analysis approach that combines SBERT for contextual feature extraction, WGAN-based synthetic data generation for addressing class imbalance, and a soft voting ensemble classifier for improved prediction. The proposed approach is evaluated on five datasets, including two English datasets and three Arabic datasets, in order to assess its performance in a multilingual setting. We compare the effectiveness of the proposed model with several baseline machine learning classifiers, as well as with commonly used data balancing techniques such as the synthetic minority over-sampling technique (SMOTE) and adaptive synthetic (ADASYN). The evaluation is conducted using multiple performance metrics, including accuracy, precision, recall, F1-score, MCC, ROC–AUC and training and inference time, along with different validation strategies including fixed train–test splits and k-fold cross-validation. The experimental results demonstrate the effectiveness and stability of the proposed approach. In particular, they highlight the importance of capturing sentence-level contextual representations and generating realistic synthetic samples to address class imbalance.

Applied System Innovation doi: 10.3390/asi9050102

Authors: José F. Villa-Tiburcio Rodrigo Hernández-Alvarado Antonio Estrada Cristían H. Sánchez-Saquín Teresa Hernández-Díaz

This article addresses the challenge of positioning accuracy in robotic manipulators applied to automated tape placement (ATL). A hybrid control strategy is proposed that integrates a Proportional-Integral-Derivative (PID) controller with a Backpropagation Neural Network (BP-NN). The proposed approach, called PID + NN, acts as a robust control scheme designed to compensate for parametric uncertainties and unmodeled perturbations arising from the integration of high-inertia tools in the end effector, dynamic mass variation due to tape consumption, and external reaction forces during the compaction process. Within this framework, the PID controller manages the nominal dynamics of the system, while the neural network operates as an adaptive compensator that adjusts the control signal in real time to minimize trajectory tracking errors. A rigorous stability analysis based on Lyapunov theory is presented, and the results are validated through numerical simulations on a six-degree-of-freedom manipulator. In addition, experimental tests are performed in a real operating environment to verify the practical performance of the strategy. The experimental results indicate that the proposed PID + NN controller significantly improves trajectory tracking accuracy, achieving a substantial reduction in tracking error and smoother control torque profiles compared to the conventional PID controller. These findings validate the effectiveness and robustness of the method for advanced manufacturing applications that demand high precision.

Applied System Innovation doi: 10.3390/asi9050101

Authors: Xiang Li Nanfang Li Zongrong Li Lijun Yan Denghui Ma Haishan Cao Xu Wang Yu Liu

SSH honeypots serve as critical infrastructure for cyber threat intelligence, but existing LLM-based systems suffer from context window limitations causing state loss and hallucination-driven inconsistencies, making them easily detectable through simple verification tests. To address these limitations, we propose LLM-SSHH, a framework combining explicit state management with LLM generation to achieve long-term interaction consistency. The system maintains a persistent state snapshot organized as a three-component tuple capturing file system state, runtime context, and system metadata. The framework serializes the current state into LLM prompts and validates generated responses against state constraints to reject hallucinations. Validated responses update the state snapshot, forming a closed loop that ensures consistent state evolution throughout extended interactions. Experimental results demonstrate that LLM-SSHH achieves a mean detection rate of 0.150, representing a 3 to 4 times improvement over existing methods, significantly extending honeypot survivability for threat intelligence collection.

Applied System Innovation doi: 10.3390/asi9050100

Authors: Yeou-Jiunn Chen Aryanti Aryanti Qian-Bei Hong

Hand gesture recognition (HGR) is a critical area in computer vision that supports intuitive human–computer interaction and sign language communication, yet existing systems remain sensitive to lighting variations, background clutter, and diverse hand postures. This study introduces two contributions to address these limitations: a Gradient-Based Augmentation Validation (GBAV) framework that establishes structurally safe augmentation ranges before training, and a multi-backbone Convolutional Neural Network (CNN) architecture combining ResNet50 and InceptionV3 with optional attention-based pooling. GBAV uses magnitude-weighted gradient orientation histograms with Pearson correlation and Kullback–Leibler divergence thresholds to verify label invariance under spatial transformations, providing a classifier-agnostic pre-training calibration mechanism. The proposed framework is evaluated on three static gesture datasets, Indonesian Sign Language (BISINDO), American Sign Language (ASL), and Hand Gesture 14 (HG14), yielding validation accuracies of 96.87%, 99.92%, and 95.25%, respectively, with 5-fold cross-validation on HG14 confirming result stability (93.51% ± 2.31%). Quantitative attention localization, cross-dataset transfer evaluation, and computational efficiency analysis (26.8 ms per image, ~37 FPS) further support the framework’s robustness and practical deployability. These findings establish GBAV-calibrated augmentation as the principal performance driver, which complements the multi-backbone architecture for robust hand gesture recognition across diverse visual contexts.

Applied System Innovation doi: 10.3390/asi9050099

Authors: Mostafa Aboulnour Salem

This study examines Responsible Decision-Making (RADM) in AI-enabled sustainability within tertiary education under conditions of uncertainty and complex interdependence. Conventional analytical approaches are limited in such settings because they typically explain behavioural relationships without adequately modelling uncertainty. To address this limitation, the study proposes an AI-driven Decision Support System (DSS) based on a hybrid probabilistic framework integrating PLS-SEM with Bayesian Network (BN) inference. The framework combines structural analysis with probabilistic reasoning in a unified, interpretable system capable of modelling conditional dependencies among decision variables. Data were collected from 713 academic leaders in tertiary education institutions in Saudi Arabia. The model examines the effects of AI-Driven Sustainable Value (AISV), Responsible AI Ease of Use (RAIU), Institutional Sustainability Support (ISS), Ethical Leadership Norms (ELN), Responsible AI Competence (RAC), and AI Risk and Hallucination Awareness (ARHA) on Responsible Decision-Making and Sustainability Impact Performance (GGIP). The results indicate that ELN and ARHA have significant positive effects on RADM, while AISV and RAIU also contribute positively to decision quality. In contrast, ISS and RAC do not demonstrate significant direct effects on RADM. However, ISS shows indirect effects through contextual and cognitive pathways. The findings further suggest that awareness of uncertainty and AI-related risks plays a more influential role in decision quality than technical competence alone. The model demonstrates strong explanatory power (R2 = 0.64) and acceptable predictive capability (R2 = 0.48). Bayesian inference further indicates that sustainability outcomes improve under favourable institutional and cognitive conditions. Overall, the framework provides an interpretable and scalable DSS that supports scenario-based evaluation and probabilistic decision analysis under uncertainty. The findings are specific to the institutional context examined in this study. Although the framework may have relevance to other organisational environments characterised by uncertainty and complex decision structures, no external or cross-contextual validation was conducted. Therefore, the findings should be interpreted with appropriate contextual caution.

Applied System Innovation doi: 10.3390/asi9050098

Authors: Ernesto José García Fernández de Castro Leonardo Lizcano Daladier Jabba Miguel Jimeno Wilson Nieto Bernal Andrés Pedraza

Background: Maritime operations increasingly rely on integrated, secure, and resilient architectures, yet the associated body of knowledge remains fragmented across conceptual, operational, logical, methodological, and governance-oriented perspectives. Objective: Our aim is to systematically review the literature on maritime integrated systems architecture in order to identify dominant themes, methodological tendencies, enabling technologies, and unresolved research gaps. Eligibility criteria: Peer-reviewed studies published in English were included when they addressed integrated systems architecture, or closely related architectural approaches, in maritime or naval contexts. Studies centred exclusively on isolated components, non-maritime settings without clear architectural transferability, or insufficient technical or methodological detail were excluded. Information sources: ACM Digital Library, IEEE Xplore, SpringerLink, ScienceDirect, MDPI, and IMarEST. Searches were carried out between January and March 2025, with the final search update for all sources completed in March 2025. Methods: The review was conducted and reported in accordance with PRISMA 2020. Three reviewers independently screened titles, abstracts, and full texts. Two reviewers independently extracted data, assessed methodological limitations and risk of bias using a review-specific qualitative appraisal framework, and evaluated the risk of bias due to missing results at the synthesis level. Disagreements were resolved through discussion and consensus, with third-reviewer arbitration when necessary. The synthesis combined qualitative thematic analysis across eleven predefined analytical categories with descriptive bibliometric and thematic mapping procedures. Results: Of 300 identified records, 60 studies met the inclusion criteria. Across non-mutually exclusive analytical categories, the literature was concentrated in Integrated Systems Architecture (52 studies), Development Processes (42), and Conceptual Models (37), whereas Zachman-based Methodology (4) and Maturity Models (3) were only marginally represented. Three recurrent patterns were observed across the corpus: the central role of cybersecurity and risk governance in architectural design; the growing importance of information technology and operational technology convergence for resilient monitoring, coordination, and decision support; and the increasing use of model-based and model-driven approaches to address architectural complexity. Overall confidence in the principal synthesized findings was judged to be moderate. Limitations: The review was limited to six databases and English-language publications, and the included studies varied in reporting depth, methodological transparency, and degree of empirical validation. Conclusions: The review organizes the field into a multilevel taxonomy spanning conceptual and operational models, logical and layered views, development processes, reference architectures, enabling technologies, and maturity-related perspectives. Taken together, the findings suggest that research in this area has progressed more clearly in architectural representation and integration than in long-term evaluation, particularly with regard to the practical operationalization of Zachman-based approaches and the development of maritime-specific maturity assessment frameworks. Funding: This review received no external funding. Registration: The review was not prospectively registered, and no publicly accessible protocol was prepared.

Applied System Innovation doi: 10.3390/asi9050097

Authors: Panfeng Zhang Xuecheng Du Chao Ma Yiran Wu Zhenya Li Hao Yun Jiajun Wei Zhirui Zheng

Data collected by sensors plays a critical role in system decision-making. Microphone arrays enable distance measurement and fault localization, which is particularly critical in the radiation environments of nuclear facilities. Acoustic localization based on microphone arrays can effectively fulfill this requirement. This study experimentally evaluates the Total Ionizing Dose (TID) effects of 60Co γ-ray radiation on commercial MEMS (micro-electro-mechanical systems) silicon microphones. Five identical microphone units were simultaneously irradiated at a dose rate of 0.0342 Gy(Si)/s while continuously monitoring operating current and spectral response. Experimental results show that the commercial MEMS silicon microphones exhibit an average TID failure threshold of 932.6 ± 62.8 Gy(Si), with a 95% confidence interval of [875.5, 989.7] Gy(Si). Three degradation/failure levels are clearly defined: channel degradation, channel failure, and full system failure. Radiation exposure causes a progressive increase in operating current (up to 6.7 times the initial value), severe spectral distortion, and ultimately complete loss of localization function. This indicated that standard commercial MEMS silicon microphones possess a certain degree of tolerance to TID radiation. Subsequently, an annealing test was performed. However, Post-irradiation annealing restored the operating current but not the acoustic performance, indicating irreversible radiation-induced damage.

Applied System Innovation doi: 10.3390/asi9050096

Authors: Mostafa A. Mahdy A. Abdellatif Mohamed Fawzy El-Khatib

Energy-aware Adaptive Cruise Control (Eco-ACC) has become an essential approach for enhancing the energy efficiency of electric vehicles while ensuring safe and comfortable driving. This paper presents a systematic review, following the PRISMA methodology, of 60 recent studies published between 2021 and 2025. The review provides a structured analysis of control strategies, validation approaches, computational demands, and battery-related considerations in Eco-ACC systems. The results indicate that Model Predictive Control (MPC) remains the most widely adopted technique (41.7%), primarily due to its ability to handle system constraints and address multi-objective optimization problems. Reinforcement Learning (RL) approaches (33.3%) are increasingly explored for their capability to adapt to uncertain and dynamic driving conditions. In addition, hybrid MPC–AI methods (16.7%) show strong potential for balancing optimal control performance with real-time implementation requirements. A key observation is the clear imbalance in validation practices: more than 73% of the studies rely on simulation-based evaluation, whereas only 10% include real-world experiments, revealing a pronounced simulation-to-reality (Sim2Real) gap. Furthermore, two critical research gaps are identified. First, the computational energy paradox highlights the trade-off between improved control performance and increased computational cost. Second, battery-aware control remains insufficiently addressed, as most existing methods overlook long-term battery degradation effects. Based on these findings, this review proposes a deployment-oriented research framework that prioritizes hybrid control architectures, real-time feasibility, and robust validation strategies, including Hardware-in-the-Loop and field testing. The presented insights aim to support the development of practical and energy-efficient Eco-ACC systems suitable for real-world deployment in next-generation electric vehicles.

Applied System Innovation doi: 10.3390/asi9050095

Authors: Shaode Yu Xinyi Li Songnan Zhao Qian Liu

Artificial intelligence (AI) is transforming modern agriculture from experience-driven practices to data-driven production paradigms. To provide an in-depth analysis of AI technologies in intelligent agriculture, we retrieved literature from Web of Science, IEEE Xplore, Google Scholar and Scopus, covering publications from 2015 to 2025, and 85 articles remained after screening 1867 relevant publications. These articles are grouped into three stages from perception, to decision making, to execution (PDE) in a closed-loop framework. At the perception level, we highlight progress in intelligent sensing systems, such as unmanned aerial vehicle (UAV) and multi-modal monitoring platforms, for crop disease and pest detection, growth monitoring and abiotic stress assessment. At the decision making level, integration of heterogeneous data sources, including meteorological records, soil measurements, remote sensing (RS) imagery and market information, supports advanced analytics, such as yield prediction, pest and disease warning, irrigation and fertilization planning, and crop management optimization. At the execution level, agricultural robots equipped with simultaneous localization and mapping (SLAM) and deep reinforcement learning (RL) facilitate precision spraying, autonomous harvesting, and unmanned field operations. Overall, AI technologies demonstrate substantial potential in the PDE pipeline of agricultural production. However, several challenges remain, including heterogeneous data fusion, limited generalization across diverse environments, complex system integration, and high hardware and deployment costs. Future directions are discussed from the perspectives of lightweight model design, cross-platform standardization, enhanced human–machine collaboration, and a deeper integration of emerging AI paradigms to support scalable, robust, and autonomous agricultural intelligence systems.

Applied System Innovation doi: 10.3390/asi9050094

Authors: Masood Mottaghizadeh Shayan Soltani Innocent Kamwa Abbas Rabiee Seyed Masoud Mohseni-Bonab

With the fast penetration of renewable energy resources (RERs) in modern power grids, system inertia is gradually decreasing, whereby threatening frequency stability. Grid-forming (GFM) permanent magnet synchronous generator (PMSG) wind turbines have emerged as a promising solution for supporting and maintaining power system stability. Nevertheless, many studies neglect the inherent intermittency and limited power capability of RERs. As a result, the dynamic interactions between machine-side and grid-side converters are often neglected, and the DC link is commonly modeled as either an ideal voltage source or a controlled current source, which may lead to inaccurate representations of system dynamics. As a solution, this paper investigates the influence of RER intermittency and power constraints on DC-link dynamics and their impact on the frequency support performance of GFM PMSGs. First, the overall system is configured using back-to-back voltage source converters, and the system’s dynamic equations are presented. Afterwards, the impact of wind speed variations is thoroughly discussed, alongside a critical examination of the requirements specified in IEEE Standard 2800-2022. Furthermore, a supervisory curtailment strategy is proposed to ensure overall system stability under severe load disturbances when the PMSG is unable to supply the required power. Finally, detailed case studies are conducted to: (1) assess the influence of variable wind speed and DC-link voltage control on the dynamic response of PMSGs, and (2) compare the performance of the accurate DC-link dynamic model with conventional idealized and simplified models.

Applied System Innovation doi: 10.3390/asi9050093

Authors: Șener Ali Simona-Vasilica Oprea Adela Bâra

The shift to 15 min market time units (MTUs) in single-day-ahead coupling (SDAC) increases temporal granularity, but complicates the interpretation of intra-hour electricity price spikes and rapid ramps. This paper examines whether architectural decomposition improves the reliability of large language model (LLM)-based diagnostics in price-only settings, rather than causal market analytics, under severe information constraints. We compare a proposed agentic workflow featuring structured context extraction, spike/ramp detection, hypothesis generation, consistency checks, and explicit uncertainty calibration against non-agentic baselines. The paper contributes: (i) a reproducible benchmark for 15 min diagnostic question answering in day-ahead markets, (ii) an agentic architecture tailored to structured time-series reasoning with explicit uncertainty handling, and (iii) empirical evidence that decomposition and verification improve evidence grounding and trustworthiness in market analytics. The evaluation includes 360 price-only cases sampled across autumn 2025, winter 2025–2026, and early spring 2026, balanced by bidding zone, temporal period, event type, and impact tier, comprising 180 spike and 180 ramp cases from six Central and Eastern European bidding zones (Bulgaria, Czechia, Hungary, Poland, Romania, and Slovakia). Using identical inputs, we assess automatic reliability metrics and human ratings. The agentic workflow improves reliability (∆ = +0.067, 95% CI [+0.049, +0.085]) and significantly increases calibrated price-only disclaimers (∆ = +0.500) relative to the monolithic LLM baseline. Human evaluation confirms higher overall quality (+0.74), helpfulness (+1.06), and correctness (+0.94), with a 65.5% pairwise win rate. Overall, the results support a narrower conclusion: structured decomposition and verification improve calibration and perceived explanation quality relative to a simple monolithic LLM baseline, but their advantages are not uniform across stronger non-agentic baselines and remain limited by the absence of exogenous market data.

Applied System Innovation doi: 10.3390/asi9050092

Authors: John Steven Fierro-Rincón Carlos Arturo Lozano-Moncada Eduardo Gómez-Luna Luis Fernando Grisales-Noreña Daniel Sanin-Villa

The rapid growth of Distributed Energy Resources (DERs) has intensified operational challenges in modern distribution networks, especially with respect to observability, bidirectional power flow, feeder model accuracy, and fast event detection. This review critically examines the role of Distribution Phasor Measurement Units (D-PMUs) in this transition. Rather than only listing reported applications, the paper evaluates the technical and practical conditions under which D-PMUs provide meaningful value beyond conventional monitoring technologies. Particular attention is given to state estimation, event detection, ancillary operation, communication latency, synchronization vulnerability, economic viability, and the limited evidence from field deployment. The review shows that D-PMUs are especially attractive at feeder heads, DER interconnection points, switching locations, and microgrid boundaries, where synchronized phase-angle measurements improve visibility of dynamic and unbalanced phenomena. However, widespread deployment is still constrained by cost, communication infrastructure, interoperability, timing security, and the scarcity of publicly documented utility-scale results. The paper concludes by identifying the most promising research directions, including physics-aware learning, graph-based analytics, edge processing, and application-driven placement strategies for DER-rich distribution systems.

Applied System Innovation doi: 10.3390/asi9050091

Authors: Iván Abel Hernández-Robles Xiomara González-Ramírez Aldo Elizarraraz-Perez Luis Ramón Merchan-Villalba Jesús Martínez-Patiño

Approximately 60% of the world’s primary energy is dissipated as waste heat, representing a critical opportunity for energy recovery in sectors such as electro-mobility and fuel cells. Commercial thermoelectric generators (TEGs), predominantly based on bismuth telluride (Bi2Te3), face limitations due to mechanical rigidity, toxicity, and high production costs. This study proposes graphene oxide (GO) as an emerging alternative thanks to its oxygenated functional groups and layered structure as well as GO paper facilitates’ thermal and electrical transport. However, the effective integration of this nanomaterial into solid-state systems under real operating conditions remains a technical challenge. Therefore, this work presents the development, multiphysics modeling, and experimental validation of an innovative TEG cell using GO paper as an active layer. The results demonstrate that the proposed GO-ITC achieves an average of 2.75 times higher generated voltage with a lower thermal gradient as well as an improved equivalent figure of merit (ZT) compared to Bi2Te3-based TEGs. This work contributes to the evaluation of GO-doped materials for voltage generation under specific thermal gradients, providing a lightweight and flexible solution for waste heat harvesting in modern power systems.

Applied System Innovation doi: 10.3390/asi9050090

Authors: Jorge Paredes Danilo Chavez Ramiro Isa-Jara Diego Vargas

The advent of Industry 4.0 has precipitated the digitization of myriad industrial processes, a feat attributable to the implementation of sophisticated digital enablers such as artificial intelligence (AI) and the Internet of Things (IoT). These technological advances have facilitated the implementation of various innovative applications, especially in the field of predictive maintenance. This approach facilitates more precise estimation of the remaining useful life (RUL) of equipment, determination of the health index (HI) of machinery, and planning of effective maintenance schedules that circumvent unexpected and costly shutdowns in industrial operations. The employment of hybrid approaches founded on machine learning algorithms in the domain of predictive maintenance signifies a perpetually evolving field of research, wherein novel techniques, methodologies, and strategies are proposed to enhance maintenance efficiency and reliability. In order to furnish a substantial and exhaustive compendium of information, a methodical literature review is hereby presented, offering a meticulous survey of the hybrid approaches utilized within this domain. The study analyzed 77 papers from the 914 papers found on the topic, to find and organize the body of knowledge, and presents a lucid taxonomy, the primary algorithms employed in hybrid approaches, the most prevalent datasets, the applicable technology architectures, and the maturity level of these solutions. This study provides a robust conceptual foundation for future research, underscoring the significance of hybrid approaches as a promising field of study, with considerable potential for advancement in the realm of industrial predictive maintenance.

Applied System Innovation doi: 10.3390/asi9050089

Authors: Nataliya Bilous Vladyslav Malko Dmytro Tkachenko Marcus Frohme

Reliable identification of deceased individuals may be difficult when conventional biometric methods such as facial recognition, fingerprint analysis, or DNA profiling cannot be applied. In such cases, medical imaging records acquired during a person’s lifetime may serve as an alternative source of identifying information. Certain anatomical structures visible in computed tomography (CT), including the sphenoid sinus, exhibit considerable inter-individual variability while remaining relatively stable within the same individual. This study investigates the feasibility of using sphenoid sinus morphology as an anatomical biometric for automated identification from head CT scans. Identification is formulated as a ranking problem in which a query CT examination is compared with a reference database using geometric descriptors derived from segmentation masks, reducing dependence on CT intensity values. The dataset consisted of CT scans from 816 individuals acquired in two patient positioning modes: Head First Supine (HFS) and Head First Prone (HFP). Several deep learning architectures, including YOLOv8 variants, YOLO11L-seg, UNet++, DeepLabV3+, HRNet, and SegFormer-B2, were evaluated for sphenoid sinus segmentation. Based on F1-score performance and cross-mode stability, YOLO11L-seg was selected and further trained to construct a database of binary masks representing individual sphenoid sinus anatomy. Identification was performed using pairwise mask comparison based on the Intersection over Union (IoU) metric. To reduce the influence of segmentation artifacts and slice-level variability, the final similarity score for each candidate was computed as the average of the four highest IoU values across slice comparisons. Individuals were ranked according to similarity, and identification was considered successful if the correct subject appeared among the top five candidates and exceeded a predefined similarity threshold. The proposed approach achieved Top-5 identification accuracies of 97.27% for HFP and 87.67% for HFS acquisitions. These results demonstrate the feasibility of using sphenoid sinus geometry as a stable anatomical biometric for automated identification. The key contribution of this study is the introduction of a ranking-based identification framework that utilizes anatomical biometrics derived from CT data for reliable patient matching.

Applied System Innovation doi: 10.3390/asi9050088

Authors: Alexandros Vrochidis Stelios Krinidis

Underwater images often suffer from degradations, including color distortion, reduced visibility, and low contrast due to light absorption and scatter in water. Numerous enhancement techniques have been proposed to improve visual quality and address these challenges. However, no single method consistently performs best across all underwater scenes. This work introduces a novel deep learning framework for the automatic selection of the most suitable enhancement technique for underwater images. A novel fused objective metric, combining the Underwater Color Image Quality Evaluation (UCIQE), Underwater Image Quality Measure (UIQM), and Underwater Image Fidelity (UIF) metrics is introduced to assess image quality effectively. The metric is then utilized to train a Shifted Window (Swin) transformer model, which predicts the best enhancement method for each image. This approach advances automatic underwater image enhancement by addressing varying image conditions with a data-driven, adaptive process. Experimental results show that the proposed model achieves an F1 score of 87.88% in selecting the optimal enhancement technique, effectively determining the best enhancement based on the characteristics of the input image.

Applied System Innovation doi: 10.3390/asi9050087

Authors: Xiao-Li Cai Yong Qin

This study investigates the problem of profit maximization in a retrial queueing-inventory system. Customers who arrive at the system when there is no stock enter a retrial orbit and are treated as retrial demands. We consider two strategies for inventory replenishment: the base stock policy and the (s, S) policy. For each strategy, we first formulate the fundamental equations needed to determine the rate matrix and the steady-state probabilities. Then, we compute the system’s performance metrics and profit function. Moreover, by leveraging particle swarm optimization (PSO) and genetic algorithm (GA), we introduce an improved hybrid optimization algorithm, Improved Hybrid Particle Swarm optimization (IHPSO), to solve the profit maximization problem. This algorithm initially uses PSO, followed by GA crossover and mutation to improve performance. In comparison to the canonical PSO algorithm (CPSO), our algorithm exhibits superior global search capabilities. Finally, we conduct a numerical analysis on the optimal decision variables and the corresponding profits utilizing the IHPSO algorithm and present several interesting findings.

Applied System Innovation doi: 10.3390/asi9050086

Authors: Hans Eguia Carlos Sánchez-Bocanegra Carlos Fernandez Llatas Fernando Alvarez López Francesc Saigí-Rubió

Objectives: Despite the widespread availability of digital clinical information, timely access to relevant biomedical evidence during routine consultations remains limited in practice. Primary care clinicians, in particular, face significant time constraints that make it difficult to integrate comprehensive literature searches into everyday workflows. This study evaluates whether an ontology-based recommender system can support routine clinical workflows by reducing information retrieval time while preserving the clinically acceptable usefulness of retrieved evidence. We assessed the performance of the HOPE (Health Operation for Personalised Evidence) system compared with realistic manual PubMed searches conducted by physicians. Materials and Methods: We conducted an observational evaluation involving 50 primary care physicians, who independently assessed 30 anonymised, rewritten clinical cases representative of common primary care scenarios. HOPE automatically extracted biomedical concepts from case descriptions using natural language processing and mapped them to Unified Medical Language System (UMLS) ontologies to generate ranked PubMed recommendations. A subset of 10 physicians also conducted manual PubMed searches in line with their usual clinical practice. Article relevance was assessed using a predefined binary criterion, and a reference relevance set was established by consensus among three senior physicians using a pooled document set. Retrieval performance was evaluated using Precision@k, relative Recall@k, and Normalised Discounted Cumulative Gain (NDCG@k). Manual search time was measured using a standardised stopwatch protocol, whereas HOPE response time was logged automatically by the system. Results: Inter-physician agreement in relevance assessment was substantial (Fleiss’ κ = 0.66; 95% CI: 0.61–0.70). HOPE achieved moderate-to-high precision within the top-ranked results (Precision@3 = 0.72), with relative recall increasing as additional documents were considered. Ranking metrics indicated that relevant articles were generally positioned early in the result lists. The mean total retrieval time for manual PubMed searches was 13.3 ± 1.7 min per case, compared with 17.4 ± 2.1 s for HOPE-assisted retrieval (p < 0.001). Conclusions: In a controlled, workflow-oriented evaluation using synthetic clinical cases, HOPE substantially reduced information retrieval time while maintaining clinically acceptable relevance in the retrieved literature. These findings support the use of ontology-based, AI-assisted systems as workflow-support tools to facilitate timely access to biomedical evidence, without replacing clinical judgment.

Applied System Innovation doi: 10.3390/asi9050085

Authors: Ritthichai Ratchapan Wanwinit Wijittemee Surasak Noituptim Theerapol Muankhaw Sawek Pratummet Boonyang Plangklang

Large-diameter axial ventilation fans are widely used in poultry houses to regulate ai flow, temperature, and air quality. However, conventional induction motors driving these fans typically operate at fixed speed and suffer efficiency degradation under low-speed, high-torque conditions due to slip-induced rotor copper losses. This study presents an experimental investigation of a manufacturing constrained conversion of a commercial induction motor platform into a direct-drive surface permanent magnet synchronous motor (PMSM). Instead of developing a completely new motor design, the proposed approach reuses the existing stator lamination, housing structure, and winding production process while redesigning the rotor electromagnetic structure to incorporate surface-mounted permanent magnets. Experimental testing was conducted using a dynamo meter-based measurement system to evaluate the performance of both the commercial induction motor and the converted PMSM prototype. The results show that the commercial induction motor exhibits significant efficiency degradation at high torque due to increased slip, whereas the PMSM eliminates slip-dependent rotor copper losses and maintains efficiencies above 88% within the typical ventilation operating range of 650–750 rpm. This study further relates airflow demand to rotational speed using fan affinity laws, highlighting the cubic relationship between speed and input power and demonstrating the energy-saving potential of variable-speed PMSM drives. The proposed conversion framework therefore provides a practical pathway for improving the energy efficiency of agricultural ventilation systems while maintaining compatibility with existing motor manufacturing infrastructure.

Applied System Innovation doi: 10.3390/asi9050084

Authors: Ioannis S. Barbounakis Ioannis V. Saradopoulos Nikolaos E. Antonidakis Erietta Vasilaki Maria S. Zakynthinaki

Wireless sensor networks increasingly support time-critical monitoring applications, where coverage optimization must often be performed under limited computational resources. This work addresses a previously underexplored WSN coverage problem involving range-free, angular-limited sensors with transmitter-induced sensing degradation and discrete sector orientation. We formulate a mixed combinatorial problem that jointly optimizes K-out-of-N sensor activation and sector assignment under strict feasibility constraints. A constraint-aware genetic algorithm with repair-based feasibility enforcement is proposed and validated against the global optimum obtained via exhaustive enumeration, enabling direct quantification of optimality. The repair mechanism corrects infeasible offspring after each genetic operation to guarantee that exactly K sensors remain active, eliminating the need for penalty-based constraint handling. A brute-force search is used to establish the global optimum of our small-scale scenario, serving as a ground-truth optimality benchmark for evaluating the proposed method. The purpose of this comparison is not to assess competitiveness against other metaheuristic algorithms, but to quantify how closely the proposed approach approximates the true optimal solution under strict problem constraints. The constraint-aware genetic algorithm is developed using an integer chromosome encoding, two initialization strategies, two crossover pairing schemes, elitism, and per-gene mutation, combined with alternative constraint-handling strategies. Two experimental series evaluate the impact of population size, crossover method, mutation probability, and constraint handling using problem-specific metrics, alongside convergence and fitness statistics. The proposed algorithm reliably reaches near-optimal solutions with significantly reduced computational cost when compared to exhaustive search. By integrating problem-specific constraints directly into the process, the proposed evolutionary optimization method effectively balances solution quality and execution time, making it well suited for scenarios requiring rapid sensor reconfiguration.

Applied System Innovation doi: 10.3390/asi9040083

Authors: Hsuan-Yu Chen Chiachung Chen

The semiconductor industry has long relied on Statistical Process Control (SPC) for yield and reliability management. In early technology nodes, classic univariate tools such as Shewhart charts, cumulative sums (CUSUM), exponentially weighted moving averages (EWMA), and the Cp/Cpk exponent could effectively monitor a finite set of key variables. However, sub-5nm and emerging 3 nm technologies have fundamentally changed the statistical environment. Advanced patterning, high-aspect-ratio etching, atomic layer deposition (ALD), chemical-mechanical polishing (CMP), and novel materials have drastically narrowed the process window. At these scales, nanometer-level deviations in critical dimensions (CD), overlay, or surface roughness can significantly impact yield. Simultaneously, modern wafer fabs generate massive amounts of high-frequency sensor data and high-dimensional metrology data. Traditional SPC assumptions—such as independence, normality, low dimensionality, and stationarity—often do not hold. Semiconductor data exhibits: (i) extremely high-dimensionality and strong intervariate correlations; (ii) a hierarchical structure encompassing fab → tooling → chamber → recipe → batch → wafer → field; and (iii) metrological delays and sampling limitations leading to incomplete and asynchronous observations. To address these challenges, this paper reviews advanced statistical methods applicable to wafer fabrication. These methods include multivariate statistical process control (MSPC) approaches such as Hotelling T2 statistics, PCA/PLS combining T2 and Q statistics, contribution diagnostics, time-series drift and change point detection, and Bayesian hierarchical modeling for uncertainty-aware monitoring in data-limited scenarios. Furthermore, we discuss how to integrate these methods with fault detection and classification (FDC), line-to-line monitoring (R2R), advanced process control (APC), and manufacturing execution systems (MES). This paper focuses on scalable, interpretable, and maintainable implementations that transform statistical analysis from a passive monitoring tool into an active component of data-driven fab control.

Applied System Innovation doi: 10.3390/asi9040082

Authors: Iryna Bondarenko

Modern standards and requirements for ensuring the reliability and safety of transport infrastructure are aimed at shifting from routine maintenance to preventive maintenance, focused on predicting technical conditions and lifecycle management. Modern engineering approaches are based on the logic of state assessment and ensuring structural strength and dimensional stability. Therefore, they focus on recording defects or deviations from acceptable values without revealing the failure mechanism, which limits the ability to identify degradation processes and predict failures. The purpose of this article is to develop a formal conceptual framework for operationalizing process-oriented reliability analysis. Within this methodological framework, state is viewed as a snapshot of a dynamic process, while process stability is defined as the ability of a system to maintain its key behavioral characteristics under changing operating conditions and the geometric and physical–mechanical properties of system elements. The proposed framework expands on classical state-based diagnostics by introducing process invariants as prognostic indicators. The transition to trajectory-based behavior analysis allows monitoring systems to evolve into lifecycle management tools.

Applied System Innovation doi: 10.3390/asi9040081

Authors: Sara Atef Ahmed Karam

Recurring socio-temporal events, such as Ramadan in Middle Eastern cities, introduce pronounced non-stationarity in urban traffic demand. During these periods, daytime traffic volumes typically decline, while congestion becomes more severe in the evening around the Iftar (fast-breaking) period and persists into late-night hours, making conventional fixed-time signal plans less effective. An additional challenge is that demand is not only time-varying, but also unevenly distributed across competing movements: attempts to prioritize high-volume phases can inadvertently cause excessive delays—or even starvation—on lower-demand approaches. To address these issues, this study presents an adaptive, regime-aware traffic signal control framework that combines predictive modeling with constrained optimization. Short-term phase-level delays are forecast using Long Short-Term Memory (LSTM) models, and a Model Predictive Control (MPC) scheme then determines the green time allocation at each control cycle through a receding-horizon strategy. The optimization explicitly represents phase interactions by including constraints that prevent excessive delay in competing movements, thereby yielding a balanced and operationally realistic control policy. The approach is validated with one-minute-resolution TomTom delay data from a signalized intersection in Jeddah, Saudi Arabia, covering both Normal and Ramadan conditions. The LSTM models show stable predictive performance, achieving root mean square errors (RMSEs) of 19.8 s under Normal conditions and 17.1 s during Ramadan. In general, the results show that the proposed framework cuts total intersection delay by about 0.3% to 2.8% compared to standard control strategies. Even though these total-delay improvements are small, they come with big drops in delay for lower-demand phases (about 12–20%) and keep the delay increases for higher-demand phases under control. This shows that the method makes the whole process more efficient by fairly spreading out the delay instead of just making one phase better on its own. The results show that combining forecasting with constrained optimization is a strong and useful way to handle changing traffic demand. This is especially true during times of high demand when flexibility, stability, and fairness across movements are all important.

Applied System Innovation doi: 10.3390/asi9040080

Authors: Han Li Dongji Li Yunxiao Liu Jinyu Ma Guangyao Wang Jianliang Ai

Cross-domain formations composed of Unmanned Aerial Vehicles (UAVs) and Unmanned Surface Vessels (USVs) are critical for maritime defense but face significant challenges in countering complex aerial threats and developing flexible, collaborative strategies. Addressing the limitations of traditional decision support systems in semantic understanding and dynamic adaptation, this paper proposes a novel Large Language Model (LLM)-driven decision support framework grounded in the Department of Defense Architecture Framework (DoDAF). By integrating Retrieval-Augmented Generation (RAG) with a domain-specific knowledge base, the framework enhances the LLM’s ability to align natural-language directives with standardized DoDAF view models, effectively mitigating hallucinations in tactical generation. The proposed framework coordinates a closed-loop process, using Petri net-based static logic verification to ensure structural consistency and Monte Carlo-based dynamic effectiveness evaluation to optimize the selection of kill chains. Experimental validations in a simulated UAV-USV maritime defense scenario demonstrate that the framework achieves 96.6% entity accuracy and 100% format compliance in model generation. In comparison, the generated cooperative kill chains significantly outperform non-cooperative methods by improving interception efficacy by approximately 26.08% under saturation attack conditions. This study develops an automated, interpretable workflow that transforms unstructured situational understanding into decision reporting, significantly enhancing the efficiency and reliability of cross-domain collaborative mission planning.

Applied System Innovation doi: 10.3390/asi9040079

Authors: Chao Zhang Zhimeng Li Ziao Li

Minimizing the impacts of coupling, randomness, time variation and uncertain nonlinearity to enhance real-time performance is critical for controlling complex industrial systems. This paper proposes an optimized adaptive filtering method (LAF-MTNF) for time-varying uncertain nonlinear systems with multiple-input multiple-output (MIMO) measurement noise, which integrates the multi-dimensional Taylor network (MTN) with Lyapunov stability theory (LST). Leveraging MTN’s inherent advantages—simple structure, linear parameterization, and low computational complexity—LAF-MTNF achieves efficient real-time filtering while avoiding the exponential computation burden of neural networks. The contributions of this work are threefold: (1) A novel integration of LST and MTN is proposed for MIMO filtering, in which an energy space is constructed with a unique global minimum to eliminate local optimization traps, addressing the stability deficit of traditional MTN filters using LMS/RLS algorithms. (2) Convergence performance is systematically quantified by deriving explicit expressions for the error convergence rate (regulated by a positive constant) and convergence region (a sphere centered at the origin) while modifying adaptive gain to avoid singularity, filling the gap of incomplete performance analysis in existing Lyapunov-based filters. (3) The design is disturbance-independent, relying only on input/output measurements and requiring no prior knowledge of noise statistics, thus enhancing robustness to unknown industrial disturbances. We systematically analyze the Lyapunov stability of LAF-MTNF, and simulations on a complex MIMO system verify that it outperforms existing methods in filtering precision (mean error 0.0227 vs. 0.0674 of RBFNN) and dynamic response speed, while ensuring asymptotic stability and real-time applicability. The proposed LAF-MTNF method achieves significant advantages over traditional adaptive filtering methods in filtering accuracy, convergence speed and anti-cross-coupling capability. This method has broad application prospects in high-precision industrial servo motion control, power system state monitoring and other multi-variable nonlinear industrial scenarios with complex noise environments.

Applied System Innovation doi: 10.3390/asi9040078

Authors: Griselda Stephany Abarca-Jiménez Manuel Vladimir Vega-Blanco Jesús Mares-Carreño Juan Cruz-Castro Yunuén López-Grijalba

Path planning and formation control in leader–follower robotic systems are active areas of research, as both are highly relevant to the proper execution of the assigned task. In this work, a dual approach to the A* algorithm is applied to generate consistent trajectories for a multi-agent robotic system with a leader–follower scheme. The conventional A* algorithm aims to minimize the cost of finding the best path by minimizing distances. In this case, a modified A* algorithm is used because, although decision-making also involves choosing among eight options or cells, the goal is not to minimize distance; instead, the focus is on analyzing the direction of acceleration. The proposed algorithm is robust regarding the initial and relative pose of the leader with respect to the followers. The leader is tracked using a digital accelerometer. The algorithm is tested by simulating various patterns and implemented in two experimental test scenarios: the first with differential mobile robots, and the second with an Ackerman-type mobile robot. In both scenarios, the trajectories were achieved with deviations in x and y between the follower’s path and the leader’s path of less than 0.03, and the leader’s pose independence was maintained.

Applied System Innovation doi: 10.3390/asi9040077

Authors: Katarzyna Turoń Andrzej Kubik

Urban mobility has undergone a significant transformation in recent years, caused by rapid urbanization, environmental pressures, and technological innovation. Even though digital tools and mobility platforms are increasingly used to address transportation challenges, these challenges remain complex and multidimensional, concerning not only infrastructure, but also user behavior, institutional coordination, trust, and social acceptance. Crowdsourcing has proven effective in leveraging distributed knowledge and accelerating innovation in business and public sectors. However, its application in urban mobility contexts has not yet been sufficiently synthesized in a framework-oriented manner. To address this, the study first conducted a comprehensive literature review of existing crowdsourcing assessment frameworks and their applicability to mobility systems. The results show that current implementations in urban mobility often remain fragmented and limited to unidirectional data extraction, lacking comprehensive approaches that integrate technological, social, and organizational dimensions. In response to this, the authors developed the SMART-CROWD framework for assessing cities’ maturity in using crowdsourcing across six dimensions: Strategy & Leadership (S), Methods & Tools (M), Engagement & Representativeness (A), Responsiveness & Impact (R), Technology & Data (T), and Civic Capital & Sustainability (CROWD). Each dimension includes measurable indicators, providing a structured basis of diagnosing disparities between technological capabilities and socio-institutional readiness. The SMART-CROWD framework is intended to support a transition from one-way data acquisition toward more scalable, reciprocal, and citizen-focused innovation ecosystems. This work contributes to the field of applied systems innovation by proposing a structured framework for assessing and guiding the use of distributed intelligence in smart urban mobility.

Applied System Innovation doi: 10.3390/asi9040076

Authors: Ulises Balderrama-Rey Rafael Verdugo-Miranda Miguel Martínez-Gil Joel Carvajal-Soto Frank Romo-García Luis Medina-Zazueta Edgar Espinoza-Zallas Rolando Flores-Ochoa

This paper presents the design and implementation of a low-profile remote monitoring and control system for water level management in storage tanks located in rural communities. The system was developed to ensure a reliable water supply, prevent spills, reduce electrical energy consumption, and mitigate theft and vandalism risks posed by a previously installed, highly exposed commercial system. The proposed system employs LoRa technology to transmit water level data from the storage tank to a receiver located 6 km from the water well. When the water level drops below a predefined threshold, the system transmits an activation signal through the LoRa network to start the well pump and trigger tank refilling. In addition, an SMS monitoring module enables users to remotely verify water level and pump operational status at any time. System notifications and operational data are automatically delivered via SMS to predefined phone numbers, enabling continuous supervision without requiring internet connectivity. The implementation of the proposed system thus provides an efficient and reliable solution for water resource management in rural environments, ensuring continuous water availability and preventing supply shortages. LoRa communication enables robust long-range data transmission, while SMS-based monitoring offers real-time operational awareness for end users. The system was validated through field testing in a pilot rural community, demonstrating operational robustness, improved water management efficiency, and measurable positive impacts on residents’ water service continuity. The low-profile physical design significantly reduced theft and vandalism incidents reported by the local water authority. Experimental results showed an average monthly reduction of 41.2% in electrical energy consumption while maintaining high system reliability, physical security, and real-time monitoring capability.

Applied System Innovation doi: 10.3390/asi9040075

Authors: Chin-Chih Chang Chi-Hung Wei Hao-Chen Li Sean Hsiao

The performance of the pitcher is the cornerstone of baseball, often determining the flow and ultimate outcome of a game. Given this centrality, understanding the mechanics of an elite pitcher and decoding their strategies are paramount for both internal optimization and competitive scouting. This study proposes an end-to-end deep learning pipeline for automatically classifying five distinct pitch types from raw broadcast footage of MLB pitcher Max Scherzer between 2015 and 2020. By formulating pitch delivery as a time-series classification problem tailored to the unique biomechanics of an elite athlete, the proposed CNN–LSTM framework integrates per-frame spatial feature extraction using an advanced CNN backbone (YOLOv8s-cls) with a two-layer long short-term memory (LSTM) network to capture subtle biomechanical cues across a standardized 20-frame delivery sequence. While skeletal pose estimation primarily focuses on tracking major joints to analyze standard pitching mechanics, the proposed pixel-based method preserves fine-grained visual cues—such as finger grip and wrist rotation—that are critical for distinguishing pitch variations. The proposed framework achieved an accuracy of 91.8% under a standard Random Split and, importantly, 84.5% under a strict Chronological Split across different seasons, validating the feasibility of automated pitch “tell” detection from broadcast video. The resulting system provides coaches and analysts with an objective, data-driven tool for generating personalized scouting reports, identifying mechanical inconsistencies, and refining pitching strategies.

Applied System Innovation doi: 10.3390/asi9040074

Authors: Roberto Baeza-Serrato Jorge Manuel Barrios-Sánchez

Daily, monthly, and annual multi-product sales records are stored in databases, but due to the massive amounts of data, they are not used for decision-making when updating product catalogs. Meanwhile, the use of artificial intelligence in business is increasing across all sectors of the economy. Large-scale data handling can be achieved using artificial intelligence techniques. Specifically, ABC inventory classification currently employs artificial intelligence techniques, including neural networks, fuzzy systems, and genetic algorithms. However, a state-of-the-art review has not found any research using vision techniques to classify ABC inventories. To address this gap, this research presents a novel approach to the intelligent classification of a company’s multiple products, using ABC. Recent vision system research often uses the Otsu method or its variants to determine the optimum threshold for binary image segmentation. Unlike this approach, our research does not use a single threshold value; instead, it uses the full binary frequency histogram as an image representation. From this, eight invariant characteristics are extracted from translation, rotation, and scale. The results show that the classification is accurate, clear, and simple as a decision-making tool. The proposed method is general and can be used in any production sector and at any enterprise size.

Applied System Innovation doi: 10.3390/asi9040073

Authors: Wei-Hsiang Lin Che-Yen Wen

End-to-end encrypted (E2EE) messaging and the growing use of cryptocurrency create an attribution gap for digital investigators because message content is unavailable and wallet activity is often decoupled from subscriber identities, which makes it difficult to link communication behaviors with wallet activity. We propose a lawful and metadata-driven forensic attribution framework called the Data-Source Association Framework (DSAF). The DSAF links encrypted communication behavior with cryptocurrency wallet activity by correlating only legally obtainable network metadata that are observable under lawful interception (LI) with on-chain traces. By integrating information from communication behaviors and wallet activity, the framework aims to narrow the person–application–wallet attribution gap. The framework integrates two components, where one performs encrypted-application classification using transport-layer signals and flow-level features and the other conducts wallet–identity association by applying controlled decoding to intercepted traffic and extracting relevant transaction traces. Both components operate under a minimum-field schema that is aligned with Taiwanese LI procedures. We implemented the workflow and evaluated it using controlled experiments across multiple wallets and assets, reporting Wilson 95% confidence intervals (CIs). We achieved 91.4% accuracy (181/198) in end-to-end association under a confidence threshold, with high performance across wallet types, including Monero and TronLink.

Applied System Innovation doi: 10.3390/asi9040072

Authors: Daniel Osezua Aikhuele Shahryar Sorooshian

In classical reliability engineering, failure is a probabilistic structural failure based on lifetime distributions of Weibull models. However, in the control-critical mechanical systems, it is possible that functional failure of the system happens before material failure occurs as a result of control power loss. This paper proposes a Controllability–Reliability Coupling (CRC) model, which redefines the concept of reliability as the stabilizability in the face of progressive degradation. The actuators’ deterioration is modeled using the time-varying input effectiveness factor α(t), and the actuator is said to be in failure when the minimum singular value of the finite-horizon controllability Gramian becomes less than a stabilizability threshold ε. The performance of the simulation indicates that the functional failure is a precursor of structural failure in several degradation conditions. A baseline comparison shows that the CRC metric forecasts loss of controllability at TCRC=17.0 s, but the classical Weibull reliability never attains the structural failure threshold even in the time horizon of 20 s. The system retains margins of Lyapunov stability and H infinity robustness are not lost, and it is still stable and attenuates disturbances even when control authority is lost. In practical degradation scenarios, the forecasted CRC failure times are 21.5 s (linear wear), 13.1 s (accelerated fatigue), 23.7 s (intermittent faults), and 24.4 s (shock damage), whereas maintenance recovery abated functional failure completely. In a case study of an industrial robotic joint, at 27.0 s, functional collapse occurred, and at the same time, structural reliability was still above the failure threshold. The findings support the hypothesis that structural survival and functional controllability are distinct concepts. The proposed CRC framework is an approach to control-conscious reliability measure, which can detect early failures and offer proactive maintenance advice in the context of a cyber–physical system.

Applied System Innovation doi: 10.3390/asi9040071

Authors: Bouchra Kouach Mohcin Mekhfioui Rachid El Gouri

Nowadays, MLOps approaches are gaining popularity thanks to their ability to apply DevOps best practices to machine learning models. They enable the automation and optimization of model training, deployment, and monitoring in various environments, while ensuring effective Continuous Integration/Continuous Deployment (CI/CD). These approaches thus promote real-time applications that can react quickly and improve continuously. This paper examines the feasibility of implementing MLOps practices in embedded systems, specifically on the Zynq-7000 FPGA board. We present a comprehensive MLOps architecture that enables the automated deployment and monitoring of a convolutional neural network model for face recognition on an embedded hardware platform for datacenter physical access control scenarios. This architecture integrates GitLab CI/CD for version control and pipeline automation, MLflow for experiment tracking and model lifecycles management, Prometheus and Grafana for monitoring, and data storage in an S3 Bucket cloud connected to DVC for dataset versioning. The results demonstrate that the proposed pipeline can be effectively deployed on a Zynq-7000 FPGA board enabling automated model retraining, redeployment, and performance monitoring. This approach reduces operational complexity and supports faster adaptation to dataset changes.

Applied System Innovation doi: 10.3390/asi9040070

Authors: Mihai Emanuel Spiţă Ovidiu Andrei Schipor

Background: Hearing impairment affects a significant portion of the global population. The development of assistive technologies, particularly wearable devices, has been pivotal in mitigating these challenges. Methods: We present a systematic literature review on wearable assistive technologies for individuals with hearing impairment, analyzing 106 scientific articles identified from diverse sources (IEEE Xplore, ACM Digital Library, and Web of Science). Our comprehensive analysis is structured around device types, body locations, user study methodologies, sensory modalities, and application domains. Results: Findings reveal a strong emphasis on auditory and visual feedback, a mix of traditional hearing aids complemented by smart wearable devices, and experimental evaluations focusing on speech comprehension and usability. Visual analysis highlights a significant anatomical shift towards body-worn and wrist-worn haptic devices. While speech accuracy is rigorously reported, user-centric metrics like comfort and battery life are frequently neglected. Conclusions: Addressing these disparities, we propose the HEAR framework (Hybrid Architectures, Engaging Experiences, Adaptive Systems, Real-world Validation). This strategic roadmap advocates for a diversification of sensory outputs, more extensive longitudinal user studies, and the development of adaptive, multi-modal solutions that seamlessly integrate into users’ everyday lives.

Applied System Innovation doi: 10.3390/asi9040069

Authors: Ali Belkhiri My Abdelouahed Sabri Abdellah Aarab

Modern deep convolutional neural networks achieve remarkable performance but require substantial computational resources due to their large parameter counts, limiting their suitability for resource-constrained environments. We propose Tiny Recursive ResNet-50, a parameter-efficient architecture that reduces model complexity through recursive feature refinement with weight sharing across reasoning cycles. The proposed design combines lightweight bottleneck blocks, iterative latent state accumulation, and deep supervision to enhance representation quality without increasing parameter count. Extensive experiments are conducted on melanoma classification using the HAM10000 dataset as the primary training and evaluation benchmark. Results demonstrate that the proposed recursive architecture maintains competitive accuracy while reducing parameters by approximately 49%, confirming its efficiency under constrained settings. To assess robustness under limited data and acquisition variability, we additionally validate on the PH2 dataset (200 images). Due to the small dataset size and class imbalance, evaluation is performed using 5-fold stratified cross-validation, and performance metrics are reported as mean ± standard deviation. This validation confirms that recursive refinement with moderate cycle depth improves stability and generalization in small-data regimes.

Applied System Innovation doi: 10.3390/asi9040068

Authors: Styve L. Ndjonkin Simen Simon P. Philbin Gordon Hunter

Artificial Intelligence (AI) is increasingly embedded in project management within the financial sector, yet existing research remains fragmented and largely focused on isolated technical applications. A systemic understanding of how AI reshapes financial project management as an integrated socio-technical capability is still lacking. This study addresses this gap through a systematic literature review of 62 peer-reviewed articles (2022–2025), combined with BERTopic-based thematic analysis supported by large language model-assisted topic representation. The findings reveal the emergence of Agentic AI as a dominant theme, marking a shift from analytical support tools toward autonomous and collaborative agents embedded in project processes. While predictive analytics and automation are relatively mature, governance-oriented and human-centric dimensions remain underdeveloped and weakly integrated. This study contributes by: (1) presenting a computationally enhanced systematic mapping study that integrates a systematic literature review with BERTopic-based topic modelling to map the evolving research landscape; (2) identifying Agentic AI as a pivotal interface between technical execution and strategic governance; and (3) proposing a socio-technical target architecture that offers a structured roadmap for AI-enabled transformation in financial project management systems.

Applied System Innovation doi: 10.3390/asi9040067

Authors: Giva Andriana Mutiara Muhammad Rizqy Alfarisi Paramita Mayadewi Lisda Meisaroh Periyadi

Continuous monitoring of sitting posture is crucial for ergonomic assessment and fatigue prevention, yet many existing approaches rely on vision-based systems or single-modality sensing that are limited in capturing spatial and temporal biomechanical dynamics. This paper presents a multimodal smart-skin sensing system for spatial and temporal ergonomic fatigue analysis in sitting postures. The proposed platform integrates 42 distributed pressure, temperature, and vibration sensors arranged in 14 trimodal sensing nodes embedded across anatomical seating and back regions to enable real-time multimodal acquisition of human–chair interaction patterns. The study introduces an analytical framework combining anatomical heatmap visualization, temporal evolution analysis, delta pressure mapping, fatigue intensity estimation, and hotspot detection to characterize dynamic pressure redistribution during prolonged sitting. Experimental evaluations were conducted using a biomechanical mannequin and a single human participant with identical anthropometric characteristics (165 cm height and 62 kg body mass) across nine seated conditions, including neutral sitting, reclining, leaning, periodic shifting, and vibration-induced motion. Each posture condition was recorded as a time-series session and segmented into temporal phases to analyze fatigue evolution during prolonged sitting. Statistical analysis of pressure redistribution dynamics indicates significantly higher pressure drift in human measurements compared with the mechanically stable mannequin baseline (p < 0.001). The proposed framework provides a scalable sensing approach for ergonomic monitoring, intelligent seating systems, and human–machine interface applications.

Applied System Innovation doi: 10.3390/asi9030066

Authors: Natthakan Rueangnetr Satra Tor. Wattanaphol Kittipon Kankhunthod Simon J. Greaves Chanon Warisarn

As the demand for digital storage capacity continues to grow, bit-patterned magnetic recording (BPMR) has emerged as a promising technology to overcome the superparamagnetic limit of conventional recording methods. Nevertheless, the extremely close spacing of magnetic islands in BPMR can result in significant signal corruption, particularly due to inter-track interference. This paper presents robust signal-processing schemes for a two-reader, three-track detection system in a staggered BPMR configuration to address these challenges. The first proposed method employs a sum-soft-information technique, which combines log-likelihood ratios from two detectors to maximize mutual information. This approach significantly improves the reliability of middle-track detection. We also propose the inter-track interference subtraction technique, in which the highly reliable data recovered from the middle track are used to reconstruct the interference signal, which is then subtracted from the upper and lower tracks using an optimized weighting factor. Simulation results at an areal density of 3.0 Tb/in2 demonstrate that an optimized weighting factor of 1.78 effectively cancels interference. Moreover, the results indicate that our proposed scheme achieves a bit-error rate (BER) comparable to that of the three-reader, one-track detection BPMR systems. Furthermore, our method also demonstrates a lower BER for both adjacent tracks when compared to the conventional single-reader, two-track reading system, even in the presence of 10% media noise.

Applied System Innovation doi: 10.3390/asi9030065

Authors: Juan Carlos Castro-Dominguez Oscar Alejandro López-Núñez Jorge O. Aguilar Karla G. Cedano-Villavicencio Oscar A. Jaramillo

Photovoltaic thermal concentration has emerged as a method to enhance the energy efficiency and performance of photovoltaic installations. This approach addresses the growing demand for renewable energy aimed at reducing emissions and mitigating climate change. It represents a significant solution for applications requiring both thermal and electrical energy under constraints of a limited available area for solar energy harvesting. However, currently developed devices rely on expensive photovoltaic cells, incorporate complex geometries that are difficult to manufacture and maintain, and employ tracking systems that complicate interconnection with similar units. The objective of this study is to design and numerically evaluate a hybrid thermal–photovoltaic modular linear Fresnel solar concentrator (H-MLFRC) based on commercial silicon cells. The proposed system allows series and parallel interconnection and is suitable for both islanded and grid-connected configurations. Its development was guided by integrated optical, photovoltaic, and thermal analyses, which defined the system geometry, characteristic parameters, and operating conditions. The results indicate that the maximum operating temperature of the device is 70 °C under a nominal operating mass flow rate of 0.45 kg/s. Additionally, the thermal and photovoltaic efficiencies are 49% and 16%, respectively, resulting in a combined efficiency of 65%.

Applied System Innovation doi: 10.3390/asi9030064

Authors: Shirong Tan Yuchun Liu Lei Wang

Traditional approaches to university-level history education often fail to provide immersive and interactive environments that foster deep cognitive engagement. To address these limitations, we developed an AI-enabled virtual museum system that integrates AI-generated content with knowledge graphs through a multi-layered architecture. The system architecture follows a three-tier framework: a front-end interaction layer (Unity/Unreal Engine) for real-time user engagement, a core service layer for intelligent event scheduling and response control (Chat General Language Model/Stable Diffusion), and a data and model layer (My Structured Query Language/MongoDB) to provide structured knowledge. To evaluate the system’s effectiveness, a four-week controlled experiment was conducted with 83 university students. The experimental group using the AI virtual museum showed a significantly higher mean post-test score (84.5 ± 6.8) than that of the control group (71.6 ± 7.9), with statistical significance at p < 0.001, starting from nearly identical baseline scores (61.2 and 60.4 for the experimental and control groups). Correlation analysis was conducted to identify scenario simulations (r = 0.59) and deep inquiry tasks (r = 0.54) as key drivers of learning mastery. By aligning advanced system engineering with educational theory, the results of this study offer a solution for high-fidelity, intelligent digital educational platforms, proposing a validated model for integrated system innovation in education.

Applied System Innovation doi: 10.3390/asi9030063

Authors: Ashraf Labib Coralia Tǎnǎsuicǎ (Zotic) Turuna S. Seecharan Mihai-Daniel Roman

The most impactful factors for the cost of fleet management are maintenance expenses and fuel consumption. Traditional ways of monitoring fleet performance fail to connect raw operational data with driving habits. The current study addresses this challenge by developing an architecture of frameworks, consisting of unsupervised and supervised machine learning algorithms, statistical testing, simulation and survival analysis to discover insights that lead to key behavioral predictors. The nucleus of this complex architecture is the decision-making grid (DMG), a two-dimensional matrix that groups vehicles based on their frequency of entering the service and the cost of their repairs. It is the first integration of DMG with ML for prescriptive fleet management. The objective of the study is twofold: firstly, to build a system that classifies vehicles according to their risk profile, and secondly, to offer clear directions for changing driver patterns that most affect vehicle costs or for keeping good practices. The framework proposed by this study not only drives the optimization of operational efficiency but also contributes to a methodology that links driver profiles to costs, offering a scalable methodology for similar business contexts.

Applied System Innovation doi: 10.3390/asi9030062

Authors: Alberto Loffredo Yunting Liu Zhengdao Chen Yifei Fu Joerg Ahrens Yifeng Lu Dong Chen

Large insurance claims handling is a complex, knowledge-intensive process that requires the analysis of heterogeneous information sources and the reuse of past experience distributed across multiple organizational data sources. Consequently, a significant portion of decision-making knowledge is embedded in historical claims records and internal documents, making systematic access and reuse challenging. This paper presents Enterprise Knowledge Assistant (EKA), a collaborative multi-agent AI system designed to act as a sparring partner for large claims handlers. EKA integrates claims structured and unstructured data with an archive of more than five thousand historical cases related to claims management, enabling retrieval, interpretation, and synthesis of relevant past cases and decision patterns. The system is organized as a set of specialized AI agents, each responsible for distinct tasks including claim context analysis, knowledge extraction, document synthesis, and interaction with human users. Through agent collaboration, EKA provides decision support by analyzing comparable historical cases, uncovering hidden correlations, and extracting insurance wisdom, while keeping the human expert firmly in control. The paper describes the system architecture and reports an industrial case study evaluating EKA in a real insurance environment. Results indicate improved knowledge reuse and reduced analysis effort in large claims handling.

Applied System Innovation doi: 10.3390/asi9030061

Authors: Andrey Nechesov Janne Ruponen

This work presents INTELLECTUM as a reference architecture and design-time evaluation framework for multi-entity XR–AI–digital twin systems. Rather than optimizing a specific implementation, the paper formalizes architectural invariants, event semantics, and coordination mechanisms that precede and inform system realization. INTELLECTUM provides a conceptual framework for structuring interactions across physical and virtual environments, emphasizing human-centered design, immersive digital twins, and collaborative extended-reality workspaces. The technical specification defines core architectural components, human integration modalities via WebXR and heterogeneous sensor networks, and representative usage scenarios within smart city ecosystems. By enabling AI-assisted urban planning, interactive simulation, and multi-actor coordination, INTELLECTUM positions itself as an XR-based architectural foundation for next-generation smart city platforms.

Applied System Innovation doi: 10.3390/asi9030060

Authors: Mario Alejandro Benavides Álvarez Fredy E. Hoyos John E. Candelo-Becerra

Mitigating pollution in cities where transportation powered by fossil fuels has a significant impact on human health is a public health priority. Although electric vehicles are one solution to this problem, their high acquisition and maintenance costs have limited their rapid adoption; therefore, other solutions may be useful in supporting reduction efforts. Therefore, this paper proposes a power control system for an Anion Exchange Membrane Fuel Cell (AEMFC) generator powered by hydrogen with the capacity to supply a direct current (DC) motor of 0.75 kW. A mathematical model of the AEMFC was proposed, and the parameters were adjusted to obtain polarization and power curves defining safe operating ranges (12.45–17.9 V). A boost converter was designed to increase the voltage of the cell output to 48 V to meet the requirements of the DC motor. The performance of the power converter was studied by analyzing its small-signal ripple, operating modes, and efficiency. The models and simulations were implemented using MATLAB and PSIM. A cascaded control system with proportional–integral (PI) and proportional–integral–derivative (PID) controllers was implemented to maintain voltage stability in the presence of input and load variation. The results show that the AEMFC is reliable and that the boost converter presents an efficiency higher than 98% in continuous mode. The robustness of the model was validated through simulations and using a prototype.

Applied System Innovation doi: 10.3390/asi9030059

Authors: Md Asif Shaharear Chengyu Zhou Shahin Shaikh Md Mehedy Hasan Faruk

Maintaining frequency stability in modern multi-area interconnected power systems has become increasingly challenging due to the stochastic nature of wind power and reduced effective system inertia. Under these dynamic conditions, traditional fixed-gain PID controllers frequently fail to provide robust regulation. To address this limitation, this study proposes and evaluates a practical model-free secondary control strategy for multi-area Load Frequency Control (LFC). The proposed hybrid MFAC–PID framework integrates an incremental model-free adaptive control (MFAC) law with a low-gain incremental PID damping term. This combination leverages real-time input–output data to determine primary control actions without relying on an explicit plant model, while the PID component supplies supplementary damping based on recent control errors. Furthermore, the controller utilizes online pseudo-gradient estimation to dynamically adapt to stochastic wind fluctuations and ±5% parametric uncertainty. Simulation results demonstrate that the hybrid design substantially enhances Area Control Error (ACE) regulation. Under wind-disturbed conditions, it reduces the aggregated Integral Absolute Error (IAEtotal) from 92.76 to 41.10, representing an improvement of over 50% compared with the fixed-gain PID baseline. Additionally, the controller maintains a low computational overhead of 0.306 milliseconds per control cycle. These findings indicate that the hybrid MFAC–PID structure provides a robust, computationally efficient solution for real-time Automatic Generation Control (AGC) in renewable-integrated multi-area power grids.

Applied System Innovation doi: 10.3390/asi9030058

Authors: Carla Gómez-Monroy Alejandro C. Ramírez-Reivich Vicente Borja José Luis Jimenez-Corona Victor Gonzalez

More than 80% of young people (11–17 years) do not meet recommended levels of physical activity, while excessive sedentary smartphone use increases rapidly, highlighting the need for accessible tools that promote active and kinesthetic learning. This study investigates whether smartphones can function as wearable devices capable of tracking movement, detecting biomechanical errors, and providing real-time corrective feedback. Using a user-centered design approach, we developed a gamified Exertion Trainer in which children practiced a straight punch (boxing jab) while wearing a smartphone on their wrist. Embedded accelerometer data were processed on board to deliver immediate, task-specific feedback on arm orientation, using gravity as a fixed reference frame. A randomized crossover trial was conducted with 40 children, comparing a feedback condition with a no-feedback control across two test orders. Quantitative results showed that real-time feedback produced a statistically significant improvement in punch accuracy (p < 0.001) and reduced performance variability, with the strongest effects observed after initial practice and partial retention following feedback removal. Qualitative findings indicated higher engagement and stronger perceptions of kinesthetic learning when feedback was available. These results demonstrate that smartphones can serve as practical wearable devices for delivering biomechanical guidance and supporting movement skill acquisition in children.

Applied System Innovation doi: 10.3390/asi9030057

Authors: Sara Atef Siraj Zahran Ahmed Karam

Accurate short-term traffic flow prediction is vital for effective signal control and sustainable urban mobility. Deep learning models, such as the Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) networks, have demonstrated strong capability in modelling temporal traffic dynamics. However, the influence of their architectural and hyperparameter configurations remains underexplored. This study proposes a systematic methodology to assess the impact of hyperparameter optimization on GRU and LSTM models for predicting traffic flow at a signalized intersection. The methodology is evaluated using minute-level traffic data from the Globe Roundabout in Jeddah, Saudi Arabia. Bayesian optimization is applied to identify the best-performing hyperparameters. The results show that the optimized GRU model achieves a Root Mean Square Error (RMSE) of 0.0953, representing a 90.2% improvement compared to the baseline GRU (RMSE ≈ 0.969). Likewise, the optimized LSTM model attains an RMSE of 0.0960, corresponding to an 85.2% improvement relative to its baseline (RMSE ≈ 0.648). Similar gains are observed for the Mean Absolute Error. Visual analysis further shows that optimized models reduce smoothing bias, enhance the tracking of transient fluctuations, and produce stable, low-variance residuals. The findings demonstrate that hyperparameter optimization substantially improves predictive accuracy while preserving computational efficiency, enabling lightweight recurrent architectures to perform at a level comparable to more complex models.

Applied System Innovation doi: 10.3390/asi9030056

Authors: Nicholas Kaliszewski Romeo Marian Javaan Chahl

This paper presents a simulation-based methodology for evaluating maintainable asset criticality in production systems modelled as complex repairable flow networks (CRFNs). The proposed Flow-Based Asset Criticality Evaluation Methodology (FACE) adopts a consequence-based perspective, assessing criticality according to network-level economic impact rather than probability-weighted risk. FACE introduces two profitability-oriented metrics, the Minimum Consequence of Failure (MCoF) at the maintainable item (MI) and failure mode (FM) levels, computed using multilayered network simulation integrating topology, capacity, failure behaviour, and profitability-driven flow allocation. By directly linking asset unavailability to system-wide gross profitability, the methodology enables objective, data-driven criticality assessment without reliance on subjective inputs, such as guided scoring processes. The approach supports both strategic and operational maintenance decisions by identifying assets and failure modes most consequential to production throughput and profitability.

Applied System Innovation doi: 10.3390/asi9030055

Authors: Iman Ghalehkhondabi

Supply Chain Management (SCM) has received considerable attention from the industrial community in recent decades. SCM continues to be an interesting and relevant research topic in many business areas such as revealing supply chain integration benefits, uncertainty and risk mitigation methods, decision-making and optimization methodologies, etc. In current supply chain management, huge volumes of data are being developed each second, and emerging technologies such as Radio Frequency Identification (RFID) have amplified the availability of online data. Using Artificial Intelligence (AI) methods that go beyond simply using the huge volume of online data enables Supply Chain (SC) managers to monitor everything in a timely fashion. There are several aspects of an SC that AI—and specifically Artificial Neural Networks (ANNs)—can be applied to better help them manage and optimize. This study aims to review state-of-the-art ANNs and Deep Neural Networks (DNNs) in the field of supply chain management. One hundred high-quality research studies that applied ANNs in supply chain management are reviewed and categorized into four classes: performance optimization, supplier selection, forecasting, and inventory management studies. Our study shows that there is a significant possibility that we could use ANNs and DNNs to better manage supply chains. Across the reviewed studies, neural networks are frequently reported to improve predictive performance and support monitoring/control in complex, nonlinear supply chain settings, often complementing traditional operations research approaches. Finally, the limitations of ANN models and the possibilities for future studies are presented at the end of this study.

Applied System Innovation doi: 10.3390/asi9030054

Authors: Sanderson Barbalho Mariannys Rodríguez-Gasca

This article presents a study that combined theoretical and empirical methods in a longitudinal approach to develop and validate the Mechatronic Reference Model for Innovation (MRM4i), a detailed framework for designing and developing mechatronic products. The text aims to present the model in terms of cycles and revisions and to compare it with the V- and W-models for mechatronic design, as well as with previous reference models in new product development (NPD). The primary characteristic of the MRM4i is to connect traditional concepts of new product development reference models—such as phases, decisions, documents, and prototypes—with the core principles of mechatronic design, as outlined in the V-Model and W-Model. The concepts and their implementation were exemplified through a longitudinal case study at a company, in which technical artifacts for four mechatronic products were presented and discussed, and compared to V/W-Models. Validation issues are outlined, and future research directions are presented.

Applied System Innovation doi: 10.3390/asi9030052

Authors: Hui-Juan Xiao Tsung-Nan Chou Jian-Fa Li Kuei-Kuei Lai

This study investigates the predictability of Environmental, Social, and Governance (ESG) performance using financial fundamentals within the context of Taiwan, a prominent small open economy integrated into global value chains. As global markets transition toward mandatory sustainability reporting, identifying the financial ante-cedents of ESG outcomes is critical for risk management and regulatory oversight. Uti-lizing a decade of firm-level data (2014–2023) from the Taiwan Economic Journal (TEJ), we employ supervised machine learning (ML) architectures-including Decision Tree, Random Forest, and Extreme Gradient Boosting (XGBoost)-to classify firms into ESG performance tiers based on indicators such as profitability, valuation, and scale. Our empirical results provide robust support for the Slack Resources Hypothesis, identifying Return on Assets (ROA) and Firm Size (SIZE) as the most consistent predictors of ESG excellence across the semiconductor, cement, and steel sectors. Conversely, mar-ket-based indicators (Tobin’s Q) dominate predictive models for the financial industry. Methodologically, XGBoost delivers superior predictive calibration for the financial sector, while Decision Trees offer highly interpretable threshold-based logic for risk screening. Our study contributes a transparent “early-warning” framework, enabling investors and regulators to identify sustainability risks through auditable financial benchmarks. The findings suggest that while financial latitude is a structural prerequisite for ESG engagement, it is not its sole determinant, pointing toward a “virtuous circle” of financial health and managerial quality.

Applied System Innovation doi: 10.3390/asi9030053

Authors: Jana Wendt Umut Volkan Kizgin Dirk Clasen Thomas Vietor

This study examines model-based systems engineering (MBSE) within the context of vehicle development at the subsystem level. The investigation encompasses the examination of the transfer of requirements from the overarching system level—the vehicle level—to its constituent subsystems, the subsequent implementation of these requirements within the subsystems, and the generation of model-based test cases for the purpose of verification. A systematized literature review according to the key principles of PRISMA 2020 was conducted to address this research question. To this end, a set of criteria for a systematic analysis were developed and applied to the identified studies.

Applied System Innovation doi: 10.3390/asi9030051

Authors: Ilya Galaktionov

Zernike polynomials constitute an essential mathematical basis for representing functions defined over the unit disk. They are widely used in a diverse range of scientific and engineering disciplines, including adaptive optics for characterizing atmospheric distortions, ophthalmology for quantifying ocular aberrations, microscopy for instrument characterization and aberration correction, and optical metrology for surface profiling. This paper introduces ZernikeViewer, a software framework developed for the rapid calculation and visualization of fringe, phase, and point spread function (PSF) maps from Zernike coefficients. The framework leverages CPU multicore and multithreading capabilities through the .NET Task Parallel Library (TPL), augmented by codebase optimizations and the preloading of precomputed Zernike polynomial matrices. These optimizations reduce computation time by a factor of 7 to 10 compared to a conventional approach; for instance, from 1 ms to 0.1 ms for a radial order of n = 10 and from 700 ms to 80 ms for n = 100. Numerical error analysis confirms the accuracy of the computation, with an average root-mean-square (RMS) error of 0.11 ms observed in the timing measurements. Furthermore, it is demonstrated that implementing Jacobi recursion relations could potentially reduce the numerical calculation error by up to 5 orders of magnitude.

Applied System Innovation doi: 10.3390/asi9030050

Authors: Andrei Militaru Emanuel-Valentin Buica Horia Andrei

This paper presents the design and implementation of a low-cost microcontroller-based dual-channel fan controller optimized for high-power audio amplifiers, yet adaptable to power supplies, electronic loads, and other thermally intensive systems. Unlike conventional designs that drive all fans uniformly, the proposed solution provides fully independent cooling via dual I2C temperature sensors, predictive trend analysis, and multi-stage hysteresis. The controller incorporates advanced features including an anti-dust startup sequence, predictive boost with latching, active cross-cooling, anti-heat-soak protection, and stall detection via tachometer monitoring, complemented by LED-based fault signaling and automatic channel muting during overheating or fan failure. Hardware support for 12 V and 24 V fans, dual power-input options, and a compact PCB layout enhance integration flexibility. The firmware employs temperature-driven PWM mapping with EMA filtering and multi-level hysteresis. The experimental results confirm that all implemented features operate as intended, with each function demonstrating clear practical relevance, whether in improving responsiveness, preventing heat accumulation, or enhancing system reliability under a wide range of operating conditions.

Applied System Innovation doi: 10.3390/asi9030049

Authors: Deemah AlFuraih Lotfi Mhamdi Abdullah S. Karar

The rapid deployment of Internet of Things (IoT) devices has increased exposure to a diverse array of evolving cyberattacks, motivating the need for accurate and interpretable intrusion detection systems (IDS). In this work, we develop an explainable hybrid Convolutional Neural Network–Extreme Gradient Boosting (CNN–XGBoost) framework for multi-class IoT attack classification using the CIC IoT-DIAD 2024 dataset. Network-traffic records are preprocessed and standardized using a scalable, chunk-wise workflow, after which a compact top-k subset of features is selected via Random Forest importance ranking. To reduce selection bias, a leakage-prone feature-ranking strategy is compared with a leakage-aware strategy in which features are ranked using only the training data within each split. Subsequently, a one-dimensional Convolutional Neural Network (CNN) learns a 128-dimensional representation from the selected predictors, and XGBoost performs the final multi-class classification. Under the leakage-aware protocol, the proposed model achieves 0.9324 accuracy with 0.5910 macro-F1. Results indicate that leakage-aware selection provides a more defensible estimate of generalization while maintaining competitive detection performance. Finally, SHapley Additive exPlanations (SHAP) is used to interpret the model’s decisions in the learned latent space. The analysis shows that only a small number of embedding dimensions contribute most of the decision evidence, which can aid analyst triage, although the explanations remain indirect with respect to the original traffic features.

Applied System Innovation doi: 10.3390/asi9030048

Authors: Linyu Tao Changchun Hua Wei Ma Gang Lu Zhenhua Wei Shijia Wei

In this paper, a nonlinear integral sliding-mode controller (SMC) based on a radial basis function (RBF) neural network is proposed to address the challenges of high nonlinearity, parameter uncertainty, and unmodeled dynamics in the electro-hydraulic servo system of a robotic excavator. The controller design incorporates adaptive RBF neural networks to compensate for system perturbations and uncertain nonlinearities, while an integral sliding surface is employed to eliminate steady-state error. This approach not only compensates for uncertainties but also reduces the traditional SMC’s high dependency on precise system parameters. The mathematical model of the bucket electro-hydraulic servo system is established without linear approximation. Based on this model, the sliding-mode controller with RBF neural networks (SMC-RBF) is designed, and its asymptotic stability is proven using the Lyapunov method. Simulation and experimental results are compared with a traditional PID controller to verify the proposed controller’s superiority. The simulations show that the SMC-RBF controller meets the requirements for tracking performance and demonstrates robustness, improving sinusoidal tracking performance by 46% compared to the PID controller. Experimental results further demonstrate that the SMC-RBF controller improves the trajectory accuracy for a two-meter straight line by 52.46% in comparison to the traditional PID controller.

Applied System Innovation doi: 10.3390/asi9020047

Authors: Mohammed Essouabni Jamal El Mhamdi Abdelilah Jilbab

Water utilities continue to lose a lot of Non-Revenue Water (NRW) because of leaks that go undetected. This makes it necessary to find accurate but easy-to-use monitoring solutions. This paper presents FiT-WST+, a wavelet-guided Frequency-Informed Transformer (FiT) designed for the classification of five distinct leak types utilising accelerometer measurements. The proposed architecture combines the spectral modelling ability of a FIT with the stable translation-invariant representation of the Wavelet Scattering Transform (WST). The model uses a guided attention mechanism to combine spectral and scattering cues that work well together to make classes more distinct, especially for fault types that are similar. On the held-out test set, FiT-WST+ achieves 99.6% accuracy, 99.6% balanced accuracy, and a 99.6% macro-averaged F1-score. Comparative benchmarking against recent methods tested on the same dataset shows that this method works at a low sampling rate (1 kHz), which greatly lowers bandwidth needs and allows for scalable deployment on edge devices with limited resources for real-time monitoring of important water infrastructure.

Applied System Innovation doi: 10.3390/asi9020046

Authors: Victória Fernanda Dionizio Marcelo Tsuguio Okano Irenilza de Alencar Nääs

Precision Livestock Farming (PLF) relies on accurate, high-frequency data to optimize production efficiency. Traditional assessments of feeding behavior remain manual and invasive, lacking the kinematic resolution required for automated control systems. This study developed and validated a novel computer vision framework integrating YOLOv8 and the Segment Anything Model (SAM) to address this gap. The objective was to engineer a non-invasive, automated pipeline to quantify high-speed broiler biomechanics in real time. The system was validated using video data from broilers across three growth stages and varying feed granulometries (fine mash, coarse mash, and pellets) to test its robustness in detecting subtle kinematic variations. The hybrid YOLO-SAM pipeline achieved high performance, with a precision of 0.95 and a recall of 0.91, confirming its reliability as a scalable sensor for smart farming platforms. Biomechanical analysis demonstrated the system’s sensitivity, showing that larger feed particles induce greater beak gape and displacement while significantly improving ingestion efficiency (0.6 effort ratio for pellets vs. 3.0 for mash). This research provides a validated technical foundation for digital phenotyping in poultry, offering a hands-free, quantitative tool that supports data-driven decision-making in feed formulation and production management.

Applied System Innovation doi: 10.3390/asi9020045

Authors: My Abdelouahed Sabri Ali Belkhiri Abla Rahmouni Abdellah Aarab

Melanoma remains one of the most aggressive forms of skin cancer, and its early detection is critical for improving patient survival. Vision Transformers (ViTs) have recently shown strong performance in dermoscopic image analysis; however, their effectiveness often relies on stacking multiple transformer encoder blocks, resulting in large numbers of trainable parameters and increased model complexity. In this study, we propose TRM-ViT, a parameter-efficient recursive Vision Transformer designed for binary melanoma classification. Instead of using multiple independent encoder blocks, TRM-ViT applies a single transformer encoder block recursively with shared weights, enabling effective depth while substantially reducing the number of trainable parameters. Experiments conducted on the HAM10000 dataset demonstrate that TRM-ViT achieves a ROC–AUC of 0.7952, comparable to a standard Vision Transformer (0.7951), while using approximately seven times fewer parameters (2.15 M vs. 14.57 M). Notably, the proposed model maintains high melanoma sensitivity, making it particularly suitable for screening-oriented applications. These results indicate that recursive weight sharing can provide an effective trade-off between diagnostic performance and model compactness, supporting the development of efficient decision-support tools for melanoma screening in resource-constrained environments.

Applied System Innovation doi: 10.3390/asi9020044

Authors: Eloina Lugo-del-Real Jorge A. Soto-Cajiga Antonio Ramirez-Martinez Edmundo Guerra Paradas Antoni Grau

Pipeline integrity is crucial for ensuring the safe and efficient transportation of hydrocarbons. One of the essential methods for maintaining pipeline integrity is periodic inspection using Pipeline Inspection Gauges (PIGs). These PIGs traverse extensive pipeline networks, collecting critical data related to inertial navigation and inspection technologies, such as geometric, ultrasonic, or magnetic flux inspection. Following an inspection, data is downloaded for post-processing to identify and accurately locate pipeline anomalies. Accurate positioning of indications is crucial for effective repair or maintenance of the identified pipeline section. Thus, ongoing efforts aim to improve the precision of indication positioning. This study introduces an innovative method and model for deriving pipeline trajectory characteristics to enhance positioning accuracy. The method is based on distance sampling of odometers, improving the PIG displacement measurement by implementing multiple odometries. Using the method described in this work can compensate for odometer slip, since the distance measurement error was reduced from 15.67% to 1.38%. The model simulates (three and four) odometer trajectories in curvature and calculates the curvature along the pipeline based on odometer data. The curvature model is evaluated with real data obtained from a test circuit, demonstrating that the proposed method and model technique can yield trajectory characteristics such as curvature detection; we can differentiate linear sections from bend sections in the test circuit. However, the curvature measurement error remains considerable due to odometer slippage. Therefore, future work proposes using additional odometers to improve measurement accuracy.

Applied System Innovation doi: 10.3390/asi9020043

Authors: Dandan Cui Zili Jiang Xiangning Zhang Wenchao Yang Zihong He

In elite sports, discovering interdisciplinary causal relationships from public data is critical for gaining a competitive edge. However, the causal knowledge required for these practices is difficult to obtain through either existing intervention-based sports science methods or computational techniques focused on statistical association. This paper formalizes a multi-domain collaborative framework, which involves three roles: (1) the elite sports team; (2) the sport science expert; and (3) the causal inference expert. Our nine-step workflow, which processes three core elements of problem, data, and computing, guides these experts through a cycle that systematically transforms practical problems into computational models and, crucially, translates complex analytical outputs back into actionable strategies. The framework also introduces a dual-dimensional “field evaluation” method, encompassing both process and outcome, to quantify the trustworthiness of knowledge in practical settings where a “gold standard” is absent. This framework was applied in an illustrative case study prior to the Paris 2024 Olympics, providing one additional evidence-informed input for the national team. The success was observed and interpreted as contextual consistency rather than causal validation. This framework ensures the practical application of causal discovery in elite sports, offering a repeatable and explainable pathway for generating credible, evidence-based insights from public data for elite sports decision-making.

Applied System Innovation doi: 10.3390/asi9020042

Authors: Valentin Lang Qichen Zhu Malgorzata Kopycinska-Müller Steffen Ihlenfeldt

Additive manufacturing is increasingly adopted for the industrial production of small series of functional components, particularly in thermoplastic strand extrusion processes such as Fused Filament Fabrication. This transition relies on technological advances addressing key process limitations, including dimensional instability, weak interlayer bonding, extrusion defects, moisture sensitivity, and insufficient melting. Process monitoring therefore focuses on early defect detection to minimize failed builds and costs, while ultimately enabling process optimization and adaptive control to mitigate defects during fabrication. For this purpose, a data processing pipeline for monitoring Optical Coherence Tomography images acquired in Fused Filament Fabrication is introduced. Convolutional neural networks are used for the automatic classification of tomographic cross-sections. A dataset of tomographic images passes semi-automatic labeling, preprocessing, model training and evaluation. A sliding window detects outlier regions in the tomographic cross-sections, while masks suppress peripheral noise, enabling label generation based on outlier ratios. Data are split into training, validation, and test sets using block-based partitioning to limit leakage. The classification model employs a ResNet-V2 architecture with BottleneckV2 modules. Hyperparameters are optimized, with N = 2, K = 2, dropout 0.5, and learning rate 0.001 yielding best performance. The model achieves 0.9446 accuracy and outperforms EfficientNet-B0 and VGG16 in accuracy and efficiency.

Applied System Innovation doi: 10.3390/asi9020041

Authors: Juan C. Olivares-Galvan Manuel A. Corona-Sánchez Rodrigo Ocon-Valdez Jose L. Hernandez-Avila Rafael Escarela-Perez David A. Aragon-Verduzco

Accurate prediction of the zero-sequence impedance (Z0) of three-legged zigzag grounding transformers is essential for ground-fault protection and power-quality performance, yet manufacturer analytical estimations often have limited accuracy. This paper investigates how accurately Z0 can be predicted using 3D finite element method (FEM) models based on the stored magnetic energy approach and how modeling the metallic tank and nonlinear core B–H behavior affects Z0 relative to analytical calculations and laboratory measurements. Two 3D FEM models are developed for a three-legged zigzag grounding transformer, incorporating the nonlinear core characteristic; impedance boundary conditions are used to efficiently account for tank-induced currents while reducing computational cost. The FEM results are compared with laboratory tests and with the analytical method used by manufacturers. The proposed models achieve errors below 4% with respect to the nominal Z0 and outperform the analytical approach. The contributions are a validated 3D FEM methodology that resolves zero-sequence flux paths under fault conditions and a practical modeling tool that improves grounding transformer design and ground-fault protection settings in modern power systems.

Applied System Innovation doi: 10.3390/asi9020040

Authors: Lifeng Yin Chenglong Li Yaohan Peng Hao Tang Ningruo Wang Huayue Chen

In modern intelligent manufacturing, spindle thermal errors are critical to machining accuracy. To address this, we propose a two-stage prediction framework. First, for feature selection, an enhanced Red-Billed Magpie Optimization algorithm (RBMO-X) optimizes the parameters of a hybrid convolutional neural network (DLTK). Concurrently, PSO-optimized HDBSCAN clustering combined with Pearson correlation selects optimal temperature-sensitive points. The DLTK network integrates LSTM, deformable convolution, Transformer, and Fourier KAN modules for robust spatiotemporal feature extraction. The experimental results demonstrate significant improvements. The proposed feature selection method improves the Silhouette index by 32.39% and increases BWP by 49.16%. Using the selected points reduces prediction RMSE by 31.89% compared to random selection. The final RBMO-X-DLTK model achieves an RMSE of 0.181 μm, an MAE of 0.128 μm, and an R2 score of 0.9978, outperforming seven benchmark models (e.g., BP, LSTM, CNN-LSTM). In practical validation, the model enabled an average thermal error reduction of 89%. This integrated approach provides a robust and accurate solution for spindle thermal error prediction, demonstrating strong generalization capability.

Applied System Innovation doi: 10.3390/asi9020039

Authors: Yunhui Gu Jingjing Yan Yunliang Ma

The problem of the dynamic quantization stabilization of the networked switched systems affected by denial-of-service (DoS) attacks is investigated. Firstly, a quasi-periodic logarithmic quantization strategy is proposed, which ensures the quantization accuracy of the quantizer under the premise of limited quantization levels. Secondly, the adjustment time and the update period of the quantizer are designed to avoid the saturation of the quantizer under DoS attacks. Subsequently, a quantized feedback controller is designed for the switched system under the influence of DoS attacks, and the sufficient conditions are obtained to ensure the global asymptotic stability of the closed-loop system. Finally, the effectiveness of the theoretical analysis is verified through a dual-tank system.

Applied System Innovation doi: 10.3390/asi9020038

Authors: Po-Lei Lee Shih-Ting Hung Pao-Hsin Chang Chun-Yen Chang Lei Bao Ting-Kuang Yeh Li-Ching Lee

This study investigated how gifted and regular high school students employ different cognitive strategies and integrate information during scientific problem solving, using eye-tracking techniques. Eighteen multiple-choice items were selected from the Investigating Scientific Thinking and Reasoning (iSTAR) assessment developed at The Ohio State University, including nine text-only questions (tMCQs) and nine picture-embedded questions (pMCQs). The items were chosen to ensure clear spatial separation among text, image, and answer areas, allowing reliable region-based eye-movement analysis. Eye-tracking data were analyzed using two indices: fixation time ratio (FTR), reflecting relative attention allocation, and saccade count ratio (SCR), capturing cross-region information integration. The results revealed clear group differences. Gifted students devoted a larger proportion of attention to pictorial information (0.38 vs. 0.32) and showed more frequent transitions between picture and answer regions (0.15 vs. 0.12), indicating more integrative processing and mental model construction. In contrast, regular students spent more time focusing on textual regions and exhibited higher within-text saccade activity, consistent with a direct translation strategy. Furthermore, SCR-based machine learning classification using a Random Forest model demonstrated meaningful discriminative capability between the two groups, particularly for picture-embedded questions, achieving an accuracy of 77.5%. Overall, the findings provide empirical evidence that question format influences students’ cognitive strategies during scientific reasoning. Methodologically, this study combines a validated reasoning assessment, a carefully defined ROI-based eye-tracking design, and interpretable behavioral indicators, offering practical implications for differentiated science instruction.