Search Results (3,765)

Low-impact development (LID) and nonpoint-source pollution reduction facilities are increasingly expected to function as long-term components of urban infrastructure rather than as isolated environmental features. Their sustainability depends not only on technical design but also on institutional review, construction assurance, operation and maintenance, performance verification, information continuity, and adaptive policy learning. This study comparatively examines the content and structure of three major South Korean guidance documents issued in 2013, 2016, and 2020: the national LID Technology-Element Guideline, Seoul’s Guidance for the LID Prior-Consultation System, and the national Manual for the Installation and Management/Operation of Nonpoint-Source Pollution Reduction Facilities. A directed qualitative content analysis was conducted using seven literature-informed lifecycle-governance dimensions covering problem framing, planning and approval, facility selection and sizing, installation and construction control, operation and maintenance, monitoring and performance verification, and adaptive feedback. Documentary provisions were assessed using a four-level ordinal scale ranging from absent to strong coverage. The analysis indicates a functional expansion of governance provisions across the three documents, from LID principles and technology selection to pre-permit review, quantitative runoff-management requirements, installation guidance, operation and maintenance, monitoring, and performance testing. However, the documents collectively show discontinuities between approval and construction verification, construction and long-term operation, maintenance and compliance, monitoring and policy learning, and lifecycle information management. In response, this study proposes a six-stage lifecycle-governance framework linking planning, pre-permit verification, construction commissioning, asset registration, risk-based maintenance and monitoring, and adaptive policy learning. The framework is presented as an analytical and normative synthesis rather than an empirically validated governance model. Because the study is limited to documentary analysis, the findings do not demonstrate actual implementation effectiveness, regulatory compliance, maintenance quality, or environmental performance and should be validated through future empirical and comparative research. Full article

Due to the large amount of network traffic, manual network security management has become increasingly difficult. This paper proposes an open-source network anomaly detection system prototype that operates with the Security Onion 2.4 platform. The system relies on a custom Python engine that uses the machine learning (ML) algorithm Isolation Forest to detect anomalies in multiple Zeek datasets. The prototype was developed as a virtual machine (VM), which is hosted on a server running the software for virtualization VMware ESXi 6.0.0. For the experimental tests, live network telemetry from a university network was used. The results show that the system achieves a constant anomaly detection rate of 5.4% after a volumetric threshold of 5000 logs is reached within operating periods ranging from fifteen to thirty minutes. During testing, it was discovered that the detection logic has a minimal resource impact (less than 1 GB) on the running system beyond the defined baseline of average Random Access Memory (RAM) usage. This proposed solution is an open-source and zero-cost alternative to other paid network security solutions, demonstrating that abnormal behavior detection in network traffic can be effectively integrated on existing computer configurations without the need for expensive licenses. Full article

Sensitivity is a critical performance parameter of sensors used in industrial process measurement systems. Conventional sensors are typically calibrated for a predefined operating range, resulting in reduced measurement sensitivity and accuracy when users require measurements within alternate subranges of the sensor’s operating band. To address this limitation, this paper proposes an intelligent autoranging technique that dynamically enhances sensor sensitivity across user-defined measurement ranges without the need for manual recalibration. The proposed approach integrates an adaptive calibration framework based on artificial neural network (ANN) algorithms to automatically adjust the sensor response for different operating intervals. The methodology is implemented and experimentally validated using a capacitance level sensor (CLS) incorporating a 60 cm helical electrode structure for liquid-level measurement. The experiments were conducted using water as the test liquid under controlled laboratory conditions. The sensor capacitance, varying from 0.929 nF to 421 μF over a liquid-level range of 0 to 60 cm with a resolution of 0.1 cm, is converted into a measurable voltage signal ranging from 0.010939 V to 2.953 V through a dedicated signal conversion and conditioning circuit. The experimental results demonstrate that the proposed ANN-based autoranging strategy significantly improves sensor performance by increasing the full-scale sensitivity from 4.9 V/m to 58 V/m across different user-selected measurement ranges. The developed technique offers a flexible and intelligent solution for enhancing sensor sensitivity and adaptability in process industry applications, enabling accurate measurements over varying operating ranges without additional calibration procedures. Full article

This work presents a reproducible YOLO-based pipeline for bottle detection in sandy environments, emphasizing dataset integrity, leakage-free evaluation, and deployment-oriented model selection. A one-class dataset of 1585 images and 3167 annotated bottles was audited to identify annotation-format defects and near-duplicate contamination between training and validation partitions. Sequence membership was reconstructed through perceptual-image similarity and used to assign complete image components to a sequence-aware train/validation split, eliminating the near-duplicate pairs found in the initial random partition. A controlled ablation holding model, seed, and corrected labels fixed showed that the random split reports 0.040 higher mAP@0.5:0.95 than the sequence-aware split (0.787 vs. 0.747), quantifying the leakage risk directly rather than only asserting it. Five YOLO configurations were then benchmarked under three independent seeds each; the observed mAP@0.5:0.95 differences among models (0.004–0.008) were small in absolute magnitude and, given only three seeds per model, are interpreted descriptively rather than as evidence of statistical equivalence or significance, so yolo11n_bottle was selected through a joint accuracy-parity, compactness, and exportability criterion (precision 0.982, recall 0.985, mAP@0.5 0.992, mAP@0.5:0.95 0.748), using approximately ten times fewer parameters than the largest configuration and producing a 5.2 MB checkpoint. ONNX export preserved detection geometry closely (100% count agreement, mean matched IoU $\ge 0.9998$), without meeting strict metric-parity tolerances. A stratified sample of 108 frames from operational RealSense BAG footage was manually annotated by an independent reviewer and evaluated quantitatively: mAP@0.5 remained close to the internal validation figure (0.927 vs. 0.992), while mAP@0.5:0.95 fell substantially (0.483 vs. 0.747), revealing a localization gap between the curated benchmark and operational conditions that this manuscript reports transparently. Together, these results show that dataset auditing, sequence-aware partitioning, multiseed benchmarking, and manually annotated operational evidence are each necessary to interpret a detection benchmark built from continuous video acquisition, providing a traceable, reproducible workflow for selecting and evaluating compact visual-perception models for resource-constrained environmental applications. Full article

Diaphragm-type hydrogen compressors serve as core equipment in hydrogen refueling stations, yet their early-stage deployment faces critical challenges, including insufficient fault mode data accumulation and unclear health evaluation criteria. Conventional manual operation and maintenance monitoring methods suffer from low efficiency and poor reliability, failing to meet the demand for safe and stable operation. To address these limitations, this study proposes a novel abnormal state detection methodology applicable under no-fault sample conditions. The approach leverages data symmetry enhancement techniques to expand the training dataset using only normal operation records, and integrates multi-source sensor data for comprehensive equipment health analysis. Experimental results show that the proposed method achieves 96% detection accuracy, an approximately 8% missed detection rate, and millisecond computational latency, significantly outperforming traditional detection algorithms. This work provides a practical solution for early-stage equipment monitoring without fault samples, enhancing both technical robustness and operational efficiency for diaphragm hydrogen compressor maintenance. Full article

To address motion blur and the degradation of object detection performance caused by variations in industrial conveyor-belt speed, this paper proposes OIL, a closed-loop periodic batch adaptation framework. First, the framework employs a YOLOv8 detector incorporating the Global Attention Mechanism (GAM) and uses the Classification Index Weight (CIW), which consists of detection confidence and the temporal consistency of predicted bounding boxes between adjacent frames. Second, the detection results are divided into three intervals according to the CIW and its constituent metrics: interval A, with a CIW below 0.3; interval B, with a CIW between 0.3 and 0.5; and interval C, with a CIW above 0.5. Results in interval A are treated as negative samples, those in interval B as ambiguous samples, and those in interval C as positive samples. Positive samples are used to train the candidate model, ambiguous samples are stored in a buffer and re-evaluated in subsequent cycles, and negative samples are excluded from the current model update. Finally, when the accumulated number of positive samples reaches a predefined threshold, the system trains a candidate model using the baseline data and newly collected samples. The candidate model is then compared with the currently deployed model on a fixed manually annotated evaluation set. The candidate model is deployed only when its overall performance improves; otherwise, a rollback is performed. Experiments on a coal conveyor line demonstrate that, at 25× speed, incorporating OIL into YOLOv8 + GAM improves Recall by 34.3 percentage points compared with the same detector without OIL. In the variable-speed experiments ranging from 5× to 25×, Recall remained above 64.3%, indicating that the proposed framework can effectively suppress missed detections under high-speed operating conditions. Full article

With the expansion of broiler production and the transition toward intelligent and labor-saving management, conventional manual inspection is limited by high labor intensity, unintuitive spatial representation of abnormalities, and inefficient on-site verification. To address these limitations, this study developed an abnormality monitoring system for cage-reared broiler houses by integrating an inspection robot, a digital twin environment, and cloud-based data services. A parameterized three-dimensional model and semantic cage anchors were established according to the dimensions of the physical broiler house and the cage arrangement rules. Robot simultaneous localization and mapping (SLAM) poses, inspection aisles, camera identifiers, and cage arrangement parameters were combined to calculate the semantic locations of dead-bird events and map them within the digital twin environment. Open-mouth breathing, infrared abnormalities, and acoustic abnormalities were additionally visualized at the candidate-cage, local-region, or inspection-aisle level according to the completeness of the available localization information. The system also enabled virtual–physical synchronization of the robot’s position, orientation, and operating status, as well as remote interactive control through a WebGL-based interface. Field tests conducted over approximately 100 days showed that model optimization reduced the triangle count, vertex count, and file size by 48.34%, 30.47%, and 44.81%, respectively, while shortening the initial WebGL scene loading time from 3.84 to 3.05 s. When 500 abnormality markers were displayed simultaneously, the optimized scene maintained an average frame rate of 67.83 fps. Among 2035 dead-bird events, 1954 were correctly localized in terms of cage row, tier, and group, yielding a cage-level localization accuracy of 96.0%. A total of 6412 robot control-command records were evaluated, achieving an overall execution success rate of 99.50%, with mean feedback times ranging from 1.0 to 1.2 s. These results demonstrate that the proposed system provides an integrated workflow for abnormality event acquisition, cage-level localization, three-dimensional visualization, and inspection robot management. Full article

Grapes intended for fresh consumption, winemaking, raisin production and processing differ markedly in quality requirements and tolerance to harvest damage, making a universally suitable mechanised harvesting pathway unlikely. This review synthesises low-damage grape harvesting from an end use-oriented sustainability perspective. A structured literature search and thematic narrative synthesis covered studies mainly published from January 2000 to May 2026, with 184 studies included. The evidence indicates that harvesting suitability is determined by the compatibility among end use, grape and bunch traits, vineyard architecture, damage tolerance, labour and operating constraints, and downstream requirements. Manual and assisted harvesting remain preferable where selective whole-bunch handling and marketability dominate; vibration-based bulk harvesting provides greater capacity for wine and processing grapes in compatible vineyards; dry-on-vine (DOV) collection is closely coupled with raisin-specific drying systems; and selective robotic harvesting remains promising but requires stronger commercial vineyard validation. Across these pathways, higher automation does not inherently provide greater sustainability because reductions in labour demand may be offset by quality loss, capital and energy requirements, or insufficient system utilisation. The principal contribution of this review is an end use-specific framework linking vineyard design, harvesting pathway, damage and loss mechanisms, and environmental, economic and social sustainability considerations. Because experimental conditions, damage definitions and reporting metrics remain heterogeneous, the available literature supports context-dependent qualitative comparison rather than a unified quantitative sustainability ranking. Full article

Coarse root systems govern tree anchorage, yet remain among the least documented components of tree architecture: excavation is irreversible, and established 3D methods rely on specialist scanners and lengthy post-processing. We evaluated whether a consumer smartphone records exposed coarse root architecture metrically, and which traits agree most closely with manual measurement. Four fully exposed Scots pine (Pinus sylvestris L.) root systems in northwestern Poland were scanned with an iPhone 17 Pro running Scaniverse, at about 30 min of acquisition and 5 h of processing per tree. Clouds were registered, cleaned and oriented to magnetic north in CloudCompare; of eight architectural metrics, four were validated against manual references at 95 cross-sections on 44 roots, and four were exploratory. Visible root length (root-mean-square error, RMSE, 22.2 cm, 8.4%), azimuth (RMSE 3.58°, mean absolute error 2.47°) and depth (RMSE 3.18 cm, 14.9%) agreed most closely with the reference; 70 of 77 first-order roots were detected with no false positives. Diameter was the weakest metric and the only one dependent on the operator (RMSE 0.46 and 0.29 cm for two operators on the same clouds). Smartphone LiDAR thus turns an irreversible excavation into a permanent, measurable record of the traits relevant to anchorage, provided that centimetre-level diameters are not required. Full article

Lean Construction and digital tools such as Building Information Modeling (BIM), common data environments (CDEs), and mobile applications are widely adopted in construction projects but are often implemented through parallel and disconnected workflows. Consequently, Lean production control continues to rely heavily on manual observations, meetings, and spreadsheets, while increasing volumes of digital project data remain underutilized for operational decision-making. This disconnect limits project teams’ ability to detect waste early, stabilize production flow, and learn systematically from recurring issues. This study develops a conceptual Digital Lean Construction (DLC) framework using a design-oriented methodology comprising three stages: synthesis of gaps in existing BIM–Lean integration research, examination of previously validated digital workflows for BIM Quality Control (QC), Quantity Takeoff (QTO), and digital twin monitoring, and integration of these components into a unified closed-loop architecture. The resulting framework organizes project information through four conceptual layers and six implementation components that connect BIM/Industry Foundation Classes (IFC4 × 3) models, schedules, issue and quality records, quantity data, field inputs, sensor information, and GIS-based spatial context. The framework assumes IFC4 × 3 because it provides enhanced support for infrastructure assets and linear referencing required for transportation and civil infrastructure projects. Rule-based analytical logic is formalized for seven Lean key performance indicators (KPIs): Percent Plan Complete (PPC), takt deviations, constraint age, rework cycles, waste event counts, QC status, and delay risk. The framework demonstrates how validated and location-aware project information can be transformed into actionable Lean performance intelligence and incorporated into weekly planning, daily huddles, problem-solving, and standardization routines. Several underlying data-generation components have been validated in previous studies; however, the integrated DLC framework itself remains conceptual and requires project-level empirical evaluation. As a conceptual framework grounded in prior literature and previously validated digital workflows, this study does not include empirical field validation. Instead, it proposes an operational architecture intended to guide future implementation and evaluation in real construction projects. The study contributes an implementable architectural foundation for moving from fragmented, retrospective reporting toward proactive, data-supported, and continuously improving production control. Full article

Thermal drying is a critical phase in coffee processing, significantly influencing energy consumption, moisture uniformity, storage stability, and sensory quality. Despite the superior throughput of mechanical dryers over open-sun drying, these systems often operate with limited observability and manual control. This systematic review evaluates the readiness of mechanical coffee drying for Digital Twin (DT) integration. A comprehensive search across Scopus, Web of Science, IEEE Xplore, and ScienceDirect identified 22,859 records. Following multi-stage screening, 58 studies were retained for qualitative synthesis, categorized into a primary coffee-drying corpus and a secondary transferable corpus of methods from related food-processing applications. Findings indicate that while DT-enabling components, such as CFD models, drying-kinetics models, IoT monitoring, and non-destructive sensing, are established, they remain fragmented. No fully implemented and operationally validated DT for mechanical coffee drying was identified. Based on the evidence, a hybrid reduced-order physics-based model integrated with constrained supervisory control represents the most defensible near-term architecture. Future research should prioritize standardized datasets, uncertainty-aware soft sensors, and field validation across diverse dryer topologies and operating conditions. Full article

The present study proposes a configuration of a solar PV-battery-powered knapsack sprayer (SPKS) equipped with a rear-mounted multi-nozzle boom to serve as a sustainable, reliable and decentralized spraying system for smallholder farmers. This design aims to significantly reduce the physiological strain and inconsistent performance associated with conventional manual lever sprayers (MLSs). The SPKS was evaluated against the MLS during onion crop spraying in terms of hydraulic performance, field capacity, spray deposit uniformity, and operator ergonomics and biomechanics, alongside an economic and environmental sustainability assessment. Results of hydraulic tests revealed that the SPKS achieved the optimal spray distribution uniformity of C.V = 16.67% at an operating pressure of 350 kPa and a boom height of 40 cm. Field experiments demonstrated that the SPKS more than doubled the effective field capacity to 0.36 ha/h compared to 0.16 ha/h for the MLS, achieving a field efficiency of 67.55%. Moreover, the SPKS achieved high spray deposit coverage (88%) compared to the MLS (52.6%), while the integrated PV panel effectively doubled operational runtime by maintaining >50% battery state of charge under continuous load. Biomechanics and ergonomics pilot analyses indicated that the MLS operation imposed high musculoskeletal (RULA score = 7) and cardiac strain, whereas SPKS operation was classified as low-risk (RULA score = 3) with minimal cardiac strain. Economically, the SPKS saves approximately $8.70 USD per hectare in labor costs. Environmentally, it prevents ~17.0 kg of CO₂ emissions annually and reduces pesticide application volume by 26.5%. By simultaneously addressing energy limitations, ergonomic hazards, and operational inefficiencies, the SPKS offers a holistic and superior solution for sustainable smallholder agriculture. Full article

Airside vehicles operate close to aircraft, service equipment and safety-critical surface markings, yet driver-assistance functions developed for road traffic do not directly transfer to airport aprons, taxiways and service roads. This article presents a vision-based driver-assistance and warning prototype for recognizing airport surface types and horizontal markings in video recorded at Poznan Airport. Two manually annotated segmentation datasets were prepared from GoPro HERO8 video acquired from a vehicle perspective: a four-class surface dataset covering asphalt, concrete, paving blocks and grass, and a three-class marking dataset covering red, white and yellow lines. The study compares You Only Look Once (YOLO) variants YOLOv8 and YOLOv11 with U-Net, DeepLabV3 and SegFormer under a common 512-by-512 input resolution and evaluates both model-level quality and complete video-application behavior. For semantic segmentation, SegFormer achieved the highest validation results, with Intersection over Union (IoU)/Dice of 0.7657/0.8624 for surfaces and 0.8852/0.9380 for markings. Among YOLO models, YOLOv8m obtained the highest surface mean average precision at an IoU threshold of 0.50 (mAP@50) of 0.7847, whereas YOLOv8s obtained the highest marking mAP@50 of 0.8449. On video recordings, paired YOLO configurations processed approximately 15–16 frames per second (FPS) on a personal computer (PC), while U-Net, DeepLabV3 and SegFormer processed approximately 10–11 FPS. A YOLOv8n pair compiled for Raspberry Pi 5 with Raspberry Pi AI HAT+ Hailo-8 reached 10.05 detection FPS and 18.15 processing FPS without GUI rendering. Under the adopted evaluation protocol, SegFormer achieved the highest mask quality in the conducted comparison, while the paired YOLOv8n configuration demonstrated embedded throughput feasibility. Full article

Wastewater treatment plants (WWTPs) are essential infrastructure for urban water management. Their stable operation is critical to effluent quality and public safety. However, wastewater treatment involves complex biochemical processes and extensive electromechanical equipment. Hazards such as sludge flotation in secondary clarifiers, fire, electric shock, and toxic gas poisoning may occur. These hazards can threaten worker safety and reduce treatment efficiency. Traditional inspection mainly relies on manual patrols, fixed-camera monitoring, and experience-based judgment. These methods often have low efficiency, limited coverage, and delayed responses. To address these limitations, this paper investigates robot-based hazard detection for WWTPs. A multisource hazard detection dataset is constructed for secondary clarifiers and confined spaces, including images collected by an inspection robot. Object detection models are then applied to identify typical hazards. Comparative experiments are conducted using Faster R-CNN and several YOLO-series models. YOLOv12 achieves mAP@0.5 values of 0.917 and 0.819 for sludge flotation detection and confined space hazard detection, respectively. It also provides a good balance between detection performance and inference efficiency. The results demonstrate that robot vision combined with object detection can support intelligent inspection in WWTPs. Full article

Decades-old enterprise-resource-planning (ERP) systems lock operational data inside unstructured, human-readable reports, forcing slow, costly, error-prone manual re-keying. Because multimodal large language model (MLLM) capability is uneven, deploying MLLMs for extraction means trusting outputs without a labeled reference. We test this with a within-document controlled experiment on 400 controlled-substance stock-ledger documents (2951 records, 11 fields, predominantly Thai) from a Thai pharmaceutical factory, comparing trained human double-entry against four open-weight MLLMs (2 × 2 design: vendor × architecture) via OpenRouter. Human double-entry left 14 discrepancies against the adjudicated gold standard, none common to both operators. The strongest model, Qwen3-VL-32B-Instruct (Dense), reached 93.95% cell accuracy; among these four models, field accuracy varied more across vendors, whereas structural completeness differed consistently between dense models (0 missing records) and Mixture-of-Experts models (up to 51 of 2951 dropped). Deterministic accounting invariants flagged 0.61% of its records, leaving the unflagged majority 94.1% accurate across all 11 fields; adding calendar rules flagged 4.61% and raised residual date accuracy from 92.1% to 95.9%. We report both operating points and recommend the extended level where date fidelity is regulatory-critical. The pipeline is 13.5–29.4× faster in wall-clock terms and 97.5–99.5% cheaper. Gold-free, rule-based verification thus locates where MLLM reliability holds, giving human–AI collaboration quantified, disclosed residual risk rather than an implied guarantee. Even at the more conservative operating point, unflagged records average 94.5% accuracy across all 11 fields but only 43.7% on the free-text Remarks field, which the triage cannot check; the results support risk reduction and the localization of review effort, not unrestricted regulatory reliability across all fields. Full article