Optimal inspection timing in missions with expected profit-based abort policy

• Fault location: This paper presents a framework for locating faults using inspection crews. • Bayes’ Theorem: Trouble calls are used to estimate the posterior probability of faults. • Multi-objective model: A multi-objective model is proposed for dispatching inspection crews. • Dynamic framework: A dynamic method updates the posterior fault probabilities using field information provided by crews and updates routes accordingly. • Efficiency: The use of posterior probability can reduce inspection time. The increasing frequency of natural disasters in recent years has significantly affected electrical distribution networks. A growing body of research has focused on dispatching crews to inspect and repair the electrical grid. While some studies estimate fault probabilities, others incorporate trouble calls or propose dynamic, multi-objective approaches. However, existing studies do not integrate all these aspects into a unified framework that simultaneously refines fault probabilities based on trouble calls, updates them dynamically using field inspection information, and embeds the updated probabilities into a multi-objective optimization model. Most existing approaches address these elements separately, for example by estimating prior fault probabilities without dynamic updating or by optimizing crew routing primarily based on shortest paths. As a result, such approaches may delay the inspection of the most likely fault locations and limit the effectiveness of inspection. This gap is addressed in this work and validated through numerical experiments. This paper proposes a dynamic bi-objective model for dispatching crews to inspect outaged areas, aiming to minimize routing efforts while prioritizing areas with higher fault probabilities. The problem is addressed using a lexicographic method, which first optimizes one objective and then solves the second without altering the result of the first. To support the model, posterior fault probabilities are estimated using Bayes’ theorem and updated with trouble call and field information, allowing adaptive crew routing. Numerical experiments on the IEEE-34 and IEEE-126 test systems demonstrate that dynamic refinement of fault probabilities, driven by inspection crew feedback, leads to substantial improvements in fault detection time.

A substantial proportion of esophageal adenocarcinoma (EAC) and high-grade dysplasia (HGD) cases in Barrett's esophagus (BE) are diagnosed after a dysplasia-negative endoscopy before the next recommended surveillance interval. These are classified as postendoscopy esophageal carcinoma (PEEC) or postendoscopy esophageal neoplasia (PEEN) and represent critical failures of BE surveillance. This review summarizes current definitions, epidemiology, potential etiologies, and evolving strategies to reduce PEEC/PEEN and improve the quality of BE surveillance. High-quality endoscopic examination using high-definition white-light endoscopy (HD-WLE) and virtual chromoendoscopy (CE), adherence to the Seattle biopsy protocol, adequate inspection time, and training in the recognition of visible lesions harboring prevalent dysplasia/EAC, are critical in reducing PEEC/PEEN. Initial data on the use of adjunctive tools (wide-area transepithelial sampling) and molecular biomarkers (p53, DNA methylation panels, and Tissue Systems Pathology tests) demonstrate promising results for detecting prevalent dysplasia and for reducing PEEC and PEEN. BE surveillance quality metrics, such as cancer and neoplasia detection rates, have been shown to be inversely associated with PEEN. PEEC and PEEN reflect critical gaps in the effectiveness of BE surveillance. Improving the quality of endoscopic surveillance is essential, including meticulous mucosal inspection, appropriate use of advanced imaging techniques, and adherence to systematic biopsy protocols, to minimize missed neoplasia.

Aeroengine blades (AEBs) are critical components of aircraft engines that require consistent monitoring and inspection to ensure airworthiness and operational safety. The aerospace industry relies heavily on maintenance, repair, and overhaul (MRO) procedures, where surface inspection of AEBs plays a pivotal role. Traditional manual or borescope-based inspection methods are time consuming, labor intensive, and susceptible to human error. In this work, a deep learning enabled intelligent robotic system is proposed for the autonomous inspection of AEBs, addressing key MRO requirements. Two distinct datasets were collected, one for aeroengine blade localization and another for surface defect detection. A robust deep learning model is trained separately on each dataset to perform their respective tasks with high accuracy. The trained models were integrated into an intelligent robotic system to automate the inspection workflow. In real time operation, a container filled with aeroengine blades is placed in front of the intelligent system, which autonomously localizes each blade, picks it up, places it in an image acquisition box, detects any surface defects, and returns the blade to its original position. The system provides real time feedback and accelerates decision making processes. Experimental results demonstrate that the proposed approach significantly reduces inspection time and enhances defect detection accuracy compared to conventional methods. By seamlessly combining vision based deep learning with robotic automation, the system offers a reliable solution for controlled industrial inspection environments for modern aerospace MRO processes, overcoming the limitations associated with manual inspection techniques. Experimental results demonstrate that the proposed approach achieves an mAP of 88.2% and reduces inspection cycle time to approximately 4 seconds per blade, significantly improving efficiency compared to conventional methods.

Abstract - This Aircraft engine maintenance is a critical component of aviation safety and operational reliability. Traditional borescope inspections depend heavily on human expertise, making the process time-consuming and prone to subjective errors. This paper proposes a Smart Borescope Inspection System integrated with Artificial Intelligence (AI) and Computer Vision techniques for automated defect detection in aero-engine components. The proposed system utilizes a high-resolution borescope camera, image preprocessing techniques, and a YOLOv8-based deep learning model to identify defects such as cracks, corrosion, dents, blade burns, and foreign object damage (FOD). Experimental evaluation demonstrates improved inspection accuracy, reduced inspection time, and enhanced maintenance decision-making. The proposed framework supports predictive maintenance and contributes toward intelligent aviation maintenance systems. Keywords: Smart Borescope, Defect Detection, Computer Vision, YOLOv8, Predictive Maintenance, Aero-Engine Inspection, Deep Learning, Aviation Safety

In colorectal cancer (CRC) screening, colonoscopy quality is assessed by sporadic adenoma detection and withdrawal time (WT). In inflammatory bowel disease (IBD), withdrawal is more complex due to IBD-specific tasks, such as targeted biopsies, dysplasia surveillance, and segmental inflammatory activity assessment. We hypothesized that in screening-age IBD patients in endoscopic remission, the yield of sporadic adenomas would be similar to that observed in non-IBD patients. We aimed to test this hypothesis recognizing that current quality benchmarks are extrapolated from non-IBD screening despite higher CRC risk in IBD. We conducted a secondary analysis of prospectively collected colonoscopy data from a tertiary academic center. Our primary outcome was sporadic adenoma detection rate (ADR) in IBD, using non-IBD patients as a reference group. Secondary outcomes were mean WT, polyp detection rate, adenomas and polyps per colonoscopy, advanced adenoma detection rate, and sessile serrated lesion detection rate. We analyzed 1366 colonoscopies (155 IBD;1211 non-IBD). When adjusting for WT and other potential confounders, ADR was significantly lower in IBD than in non-IBD (18.6% vs 43.0%). The increase in detection per additional WT minute was markedly attenuated in IBD: the odds of detecting an adenoma increased by 16.6% per additional minute in non-IBD versus 2.7% in IBD. Achieving a 26% ADR required an estimated WT of 6.8min in non-IBD versus 28.2min in IBD. In a screening-age cohort without endoscopic IBD activity, our findings suggest that IBD-specific procedural demands limit effective inspection time, and other intrinsic characteristics of the surveillance context influence adenoma detection dynamics. Non-IBD quality thresholds may not be directly applicable to IBD, supporting the development of IBD-specific quality metrics.

Change blindness involves a failure to detect changes in a visual scene, despite the differences being available to visual perception. Although change detection is crucial for many every day or professional behaviours (e.g., driving, occupations such as radiology), little is known about the individual differences that contribute to change detection ability or the cognitive domains involved. Previous research has implicated attentional and visual memory processes in successful change detection. The present study used a naturalistic change detection task, the Alternate Forms Flicker Task (AFFT), paired with a battery of nine cognitive tests, to ascertain which cognitive domains are most strongly associated with change detection accuracy in a sample of 260 participants. Strongest correlates with AFFT accuracy were tasks assessing top-down attentional search (Visual Attentional Capture and Control Task), visuospatial ability and memory (Austin Maze and Object 2back Task), and visual inspection time (Subtle Cognitive Impairment Test). Principal Axis Factoring extraction with Promax rotation showed that cognitive flexibility, visuospatial working memory, and attention and processing speed collectively accounted for 41% of the variability in test battery scores. Using these three factors, a multiple linear regression model significantly accounted for 10.0% of the variability in AFFT scores. Visuospatial working memory was the principal factor, indicating that individual differences in visuospatial working memory likely contribute to differences in change detection accuracy.

New H2S colorimetric probe based on aggregation of poly(allylamine) grafted by perylene diimide radicals: mechanism and application in analysis of sulfur-containing protein spoilage.

Low and middle-income countries (LMICs) like India bear a substantial burden of gastrointestinal (GI) cancer morbidity and mortality, largely attributable to late-stage diagnosis, which limits curative treatment options and drives poor survival outcomes. Although population-based endoscopic screening programs in East Asia have demonstrated significant mortality reduction, such approaches are not currently feasible in India due to population scale, heterogeneous infrastructure and workforce constraints. Consequently, India requires a pragmatic strategy focused on earlier detection within routine clinical practice. This review proposes a quality-first, phased framework for early GI cancer detection in India, emphasizing targeted screening and risk-based surveillance rather than population-wide endoscopic screening. High-quality conventional white light endoscopy (WLE) supported by adequate inspection time, systematic examination protocols, optimal mucosal preparation and objective quality metrics forms the foundational step, as most early GI neoplasia is detectable with optimized WLE. Image-enhanced endoscopy (IEE), including chromoendoscopy and digital optical technologies, is positioned as a complementary tool that provides incremental benefit when selectively deployed in quality-assured settings for high-risk populations and lesion characterization. Emerging Indian data demonstrates the feasibility of detecting early gastric cancer following structured training and systematic inspection, with curative endoscopic sub-mucosal dissection achievable in appropriately selected cases. However, widespread implementation faces barriers including equipment costs, training gaps, high procedural volumes, short inspection times and variable procedural quality. To address these challenges, a staged implementation roadmap is outlined: (1) adoption of risk-targeted screening in high-incidence regions and high-risk individuals; (2) standardization of endoscopic quality measures, including systematic inspection and minimum examination times; (3) integration of structured diagnostic training and selective IEE exposure into gastroenterology curricula and (4) expansion of access through hub-and-spoke networks, shared equipment models and low-cost chromoendoscopy in resource-limited settings. Coordinated action across clinicians, training programs, institutions, professional societies and policymakers is essential to transitioning GI cancer care in India from late-stage palliation toward early detection and curative endoscopic therapy.

This research presents a comprehensive and automated framework for detecting surface cracks and measuring their widths in reinforced concrete (RC) members using a modified YOLO-V11 deep learning (DL) architecture. Basically, manual surface-crack inspection is subjective and labor-intensive, relying heavily on an inspector's skill and conditions on-site. This often leads to inconsistent assessments and longer inspection time. Thus, the proposed approach mitigates these limitations by integrating automated crack detection with direct quantitative crack measurement. The proposed framework presents: (1) a DL crack segmentation model trained on a diverse dataset to enhance generalization in realistic inspection conditions for crack detection and segmentation, (2) a crack width measurement algorithm using patching and stitching method, and (3) a customized image calibration and scaling approach to transfer crack dimensions from pixel-size to real-size using low-cost imaging devices and tools. Finally, the proposed framework was validated using 230 measured crack points collected from both experimental specimens and existing RC structures. The prediction accuracy reached a coefficient of variation of 16.82% and a mean relative error of 12.65%, confirming the reliability of the proposed framework for crack measurements of RC structures.

Surface defect inspection of metallic lock components remains challenging due to strong specular reflections, low-contrast defect patterns, and geometric variability, which limit the consistency of manual inspection and conventional automated optical inspection (AOI) systems. This study presents an integrated visual inspection framework that combines controlled image acquisition with deep learning-based semantic segmentation to enable reliable and repeatable defect detection. A standardized rotational fixture with ring illumination was developed to stabilize imaging geometry, reduce reflection variability, and support consistent multi-view acquisition. A region-of-interest (ROI) masking strategy was further applied to suppress background interference and isolate the effective inspection region. At the algorithmic level, a Transformer-enhanced U-Net (TransU-Net) architecture was employed to jointly model local spatial features and global contextual dependencies, thereby improving boundary delineation and the detection of irregular surface anomalies. In addition, a boundary-aware weighted evaluation scheme was introduced to provide a more robust and application-relevant assessment by accounting for annotation uncertainty near defect edges. Experimental results demonstrate that the proposed method achieved an F1-score of 85.15%, with an average inference time of 0.3357 s per image for model prediction. Considering additional processes such as multi-view image acquisition, mechanical rotation, and preprocessing, the overall system-level inspection time is expected to be on the order of seconds per component in practical deployment.

Ensuring the structural integrity and service reliability of railway wheels has become a key challenge in modern manufacturing and maintenance strategies within the railway sector. In this context, Eddy Current (EC)-based Non-Destructive Testing (NDT) provides an automated and efficient approach for detecting surface and near-surface defects, while reducing inspection time and operator dependency compared to conventional manual methods. This study presents the integration of an EC inspection system into a precision lathe, enabling in-machining evaluation during wheel turning. Experimental validation was conducted on wheels with artificial defects, yielding high signal-to-noise ratios and enabling reliable defect characterization. Furthermore, computationally efficient and easily deployable machine learning algorithms were developed to enable automatic defect detection, localization, and size estimation. The results confirm the feasibility of in-machine EC inspection during machining operations, enabling early defect detection and contributing to safer, more efficient, and higher-quality manufacturing processes in the railway sector.

ABSTRACT Elite athletic performance is linked to advanced cognitive functioning, yet cognition’s role in motor skill acquisition among novices remains underexplored. This study examined how cognition influences learning a complex field hockey skill in participants with minimal prior experience. Forty novices (mean age = 20.04 years) completed a hockey ball control task assessed across six timepoints, three before and three after viewing a coaching video of an expert performing the task. Cognitive measures included fluid intelligence (Raven’s Progressive Matrices), crystallized intelligence (Spot-the-Word), working memory (OSpan), perceptual speed (Inspection Time), and psychomotor ability (Fitts’s task). Declarative and procedural knowledge were also recorded. Performance was evaluated using positional and technical scoring systems. Gains through repetition were modeled via linear regression. No overall significant improvement occurred during uncoached repetition. However, crystallized intelligence predicted individual differences in pre-coaching gains while fluid intelligence was positively associated with immediate coaching effects. Post-intervention improvements were predicted by working memory capacity and number of practice attempts. Psychomotor ability predicted gains through repetition both before and after intervention. Distinct cognitive domains support different phases of motor learning. Cognitive profiling may inform talent identification in early skill acquisition.

Despite substantial progress in understanding its pathophysiology and risk factors, gastric cancer remains a significant global health burden. Advances in endoscopic technology have improved the potential for early detection, yet variability in clinical practice persists. In this comprehensive narrative review, we summarize the most recent epidemiological trends in gastric pre-neoplastic and neoplastic lesions and critically appraise current evidence on optimizing endoscopic techniques and strategies for the detection of early gastric neoplasia, with an emphasis on emerging innovations. The relevant literature on epidemiology, risk factors, pathophysiology, and endoscopic management of GC was selectively reviewed based on the authors' expertise and appraisal of contemporary evidence. Marked global disparities persist in GC incidence, mortality, and stage at diagnosis. Interval GC-including missed lesions and so-called "true" interval cancers-remains a clinically relevant challenge and is frequently identified at advanced stages. These gaps are partly attributable to inconsistent quality in diagnostic esophagogastroduodenoscopy (EGD). High-quality EGD relies on adequate mucosal inspection time, systematic photodocumentation, optimal gastric preparation, and the use of standardized terminology, including mucosal visibility scores. Routine integration of chromoendoscopy and magnification techniques further enhances detection rates. Looking ahead, artificial intelligence holds promise as a transformative adjunct to standardize and augment real-time lesion recognition and quality assurance. High-quality endoscopic evaluation, coupled with tailored surveillance strategies, enables earlier detection of pre-neoplastic lesions and early gastric cancer, improving clinical outcomes. Future priorities include broadening access to high-quality endoscopy, harmonizing performance standards, and promoting continuous training alongside technological integration.

The article presents a modular, automated inspection system for ceramic inserts used in DPF and GPF filters. The system integrates 3D scanning based on structured light with 2D imaging using telecentric optics. The applied hybrid detection algorithms, combining deep neural networks with classical image processing methods, enable the identification of cell blockage within the ceramic insert as well as defects on its external surface. Converting point clouds into 2D representations allows for the effective use of convolutional neural networks, simplifies the defect labeling process on images, and reduces computational requirements. The conducted tests confirmed high detection accuracy while maintaining short inspection times consistent with industrial constraints. The modular architecture and full automation of the inspection process enable flexible adaptation of the system to diverse production lines.

Effective monitoring is vital for maintaining interconnected infrastructure systems, where components are prone to failure without proper servicing. However, designing inspection routes remains computationally difficult due to high complexity and inherent uncertainties of large-scale infrastructure systems. This paper investigates the deployment of multi-vehicle fleets, such as unmanned aerial vehicles (UAVs), to inspect spatially distributed components subject to uncertain travel times, inspection durations, and failure risks. Notably, the probability of component failure depends on inspection timing, creating decision-dependent (endogenous) uncertainty. We model this as a variant of a stochastic multi-vehicle routing problem and formulate a two-stage stochastic mixed-integer program based on finite samples. We propose a scenario decomposition framework that integrates column generation and random coloring techniques to accelerate subproblem resolution. We further provide theoretical analyses of the algorithm’s finite convergence and optimality guarantees under a user-specified probabilistic error tolerance. Numerical experiments on networks of varying topologies, including IEEE and EPANET systems, demonstrate the computational efficiency and effectiveness of our approaches. Across all large instances, our algorithm achieves an optimality gap below 4% and consistently outperforms state-of-the-art optimization solvers and the adaptive large neighborhood search as a heuristic benchmark.

Civil infrastructure inspection remains labor-intensive, subjective, and hazardous when relying on traditional visual assessments and manual measurements. This review examines how augmented reality (AR) addresses these challenges. The aim is to map the state-of-the-art AR applications across core inspection areas and to evaluate their operational performance. A systematic scoping literature review was conducted over peer-reviewed journals, conference proceedings, and industry reports from the past decade, focusing on AR system design, deployment, and empirical results. Results demonstrate that AR tools can reduce inspection time by up to 30 percent, improve defect measurement accuracy to within millimeter-scale tolerances, and support remote collaboration and AI-augmented defect detection. However, there are still major issues with data interoperability, multi-user scalability, spatial registration accuracy, and cognitive load management. While AR has progressed from proof-of-concept to real-world pilots, the review concludes that its broad adoption depends on improvements in tracking accuracy, standardised data frameworks, and integration with cutting-edge technologies like 5G edge computing and AI-driven analytics.

PT XYZ operates in the manufacturing industry with a focus on producing wind musical instruments. The Incoming Quality Assurance (IQA) inspection process at PT XYZ faces challenges related to the effectiveness and efficiency of quality control, as indicated by the high number of material defects recorded in April 2025, totaling 8,583 defective items. The incomplete implementation of the Acceptable Quality Level (AQL) method at PT XYZ has resulted in some inspection processes still being conducted through 100% inspection, leading to inefficiencies in inspection time, particularly given the limited number of available inspectors. This study aims to evaluate the application of the AQL method based on the ANSI/ASQ Z1.4 standard and to analyze sample performance as an effort to improve inspection efficiency. The methods applied in this study include the use of the Seven Tools approach to identify types of defects, the application of the AQL method in the sampling process and lot acceptance decision-making, and the measurement of sample performance. The Seven Tools analysis identified scratch defects as the most dominant type, accounting for 35% of the total defects. The implementation of the AQL method resulted in an AQL value of 2.5 under the normal inspection category, with a sample size of 20 units and acceptance and rejection numbers of 1 and 2, respectively. The sample performance evaluation showed an Operating Characteristic (OC) curve value of 0.9198, an Average Outgoing Quality (AOQ) of 1.916%, and an Average Total Inspection (ATI) of 29.7 units. These findings demonstrate that the application of the AQL method is effective in reducing defect rates while improving the efficiency of the IQA process without the need for 100% inspection.

Evidence about the effect of artificial intelligence (AI) on upper endoscopy in multicenter, randomized controlled trials is lacking. We aimed to explore whether AI can enhance gastric neoplasm detection. Participants from 24 hospitals in China from December 21, 2021, to November 11, 2023, were randomized to AI-assisted or nonassisted esophagogastroduodenoscopy. Primary outcome was detection rate of gastric neoplasms after pathologic review. Secondary outcomes included detection rate of gastric neoplasms before review, relative early gastric cancer detection ratio, detection rate of intestinal metaplasia and/or gastric atrophy before or after review, biopsy rate, number of blind spots, and procedure/inspection time. We did intention-to-treat (ITT), per-protocol, and exploratory subgroup analysis. In the ITT cohort, 29,514 patients were enrolled. AI did not improve detection rate of gastric neoplasm after pathological review (RR, 1.13; 0.92-1.38; 1.42 vs 1.25%; P = .25). However, based on original pathology, an improvement with AI was observed (RR, 1.14; 1.0-1.28; 4.06 vs 3.57%; P = .03). AI reduced blind spots number from 2.52 to 1.07 (P < .001) and prolonged procedure and inspection time. No significant differences were observed for relative early gastric cancer detection ratio or detection rate of intestinal metaplasia and/or gastric atrophy before/after pathologic review in ITT cohort. Subgroup analysis suggested potential benefit among less experienced endoscopists and during fatigue periods. In the experimental group, AI diagnosed 100%, 91.9%, and 57.1% of pathologically confirmed gastric adenocarcinoma, high-grade, and low-grade intraepithelial neoplasia, respectively. AI did not improve the detection rate of gastric neoplasms. Further real-world studies are needed to fully address the adaptability of AI. (Chinese Clinical Trial Registry, ChiCTR2100054449.).

This study presents the development of a 3D vision–based quality inspection platform applied to Printed Circuit Heat Exchangers (PCHE) and Shell & Plate heat exchangers (SPHE), which are among the core components of LNG carrier systems. The proposed system aims to overcome the limitations of conventional manual inspection methods and to establish a standardized inspection framework compliant with international quality standards. The inspection performance and economic feasibility were analyzed through a comparative evaluation before and after system implementation. As a result, the average inspection time was reduced by more than 60%, and the time required for report generation decreased by 95%. The defect detection rate improved from 79.2% to 96.4%, and the rework rate decreased from 6.1% to 1.3%. Economic analysis confirmed an annual cost reduction of approximately KRW 44 million, with an ROI of 32.5% and a BEP of 3.06 years. Furthermore, the system demonstrated conformity with international standards including ISO 19087, ISO 9001, and classification society regulations (KR, ABS, DNV, LR). This study not only verified the applicability of 3D vision inspection in the shipbuilding equipment manufacturing process but also provided an empirical foundation for smart factory transformation and 3D-based quality management in small and medium-sized shipbuilding industry.