{"id":113159,"date":"2026-02-11T11:27:18","date_gmt":"2026-02-11T10:27:18","guid":{"rendered":"https:\/\/industry-science.com\/?post_type=article&#038;p=113159"},"modified":"2026-06-29T20:06:27","modified_gmt":"2026-06-29T18:06:27","slug":"data-quality-expertise-ai","status":"publish","type":"article","link":"https:\/\/industry-science.com\/en\/articles\/data-quality-expertise-ai\/","title":{"rendered":"Data Quality and Domain Expertise for Resilient AI Deployment"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">AI offers significant opportunities in industrial quality control but also presents complex challenges along the human-AI interface, demanding a focus on applied ethics and human-centered design principles. In industrial settings, like those with automatic surface inspection systems (ASIS), machine learning is increasingly deployed for error detection tasks, such as steel rolling inspection. While this may promise increased efficiency and consistency, challenges concerning <a href=\"https:\/\/factory-innovation.de\/artikel\/human-centered-manufacturing\/\" target=\"_blank\" rel=\"noopener\">trust<\/a>, usability, adaptability, and the role of human workers do arise as a result. Standard AI deployment often seems to neglect these dimensions, thus hindering adoption and raising ethical concerns. A human-centered approach to AI (HCAI) is therefore required.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This paper presents findings from a science-practice collaboration with an ASIS supplier. Core obstacles to the effective integration of ASIS into production lines were identified: data inefficiency (especially for rare errors), user distrust due to unclear model behavior and variable performance, limited user AI expertise, and information overload for operators. These practical challenges can be addressed by specific HCAI-aligned technical solutions focused on data quality, model transparency, and user empowerment. Co-developed solutions presented here\u2014anomaly detection, label error detection, and drift monitoring concepts\u2014can, however, enhance worker agency, trust, and sustainable AI adoption in industrial surface inspection.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">This paper\u2019s key contributions are: (1) identifying practical ASIS deployment challenges as ethical HCAI deficits, (2) mapping these challenges onto HCAI solutions, thereby (3) demonstrating how AI ethics can provide a framework for solving practical human-AI collaboration challenges in industry, and (4) presenting technical solutions for data quality awareness and reliability management.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Background<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Effective and ethical AI integration into workplaces requires an understanding of both AI reliability and human-centered work design principles. Applied AI ethics concerns relate to fairness, accountability, transparency, and impact on work. HCAI provides a framework to address these challenges along three key dimensions [1, 2]:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><em>technology development<\/em> is concerned with ensuring consistent and robust performance (trustworthiness\/reliability) and making outputs understandable (explainability)&nbsp;<\/li>\n\n\n\n<li><em>employee development<\/em> focuses on empowering rather than replacing people (human agency &amp; augmentation)and ensuring tolerable work conditions (sustainable work design)&nbsp;<\/li>\n\n\n\n<li><em>organizational development<\/em> strives for ethical data governance by maintaining reliable organizational knowledge bases (responsible knowledge management\/accountability) and integrating user domain knowledge, thereby enabling non-experts to manage AI quality.&nbsp;<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Neglecting these dimensions creates systems that are technically functional but practically unusable or untrustworthy for the end users [3].<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">On a technical level, AI performance depends on data quality. Common industrial application issues include label errors, outliers, and insufficient data for rare classes. Anomaly detection helps identify novel or rare events, which is critical for robust error detection. Label error detection improves dataset integrity and model evaluation reliability. Domain drift poses a major challenge to model performance, requiring monitoring to detect shifts in data distributions and responding to changing conditions such as lighting or material variations. Model confidence scores can be misleading, especially under drift, necessitating drift-aware trust calibration.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Prior work on this research project employed a collaborative, interdisciplinary approach involving academic researchers (computer science, engineering, social sciences, information systems), an industrial ASIS provider, and ASIS quality engineers to identify challenges to human-centered AI design. That work utilized interdisciplinary workshops with stakeholders (system developers, project managers), end-user feedback, and structured potential and socio-technical workflow analyses [4].<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">User needs were assessed through semi-structured interviews with quality engineers and line operators, documented via audio-recorded transcriptions. Interview transcripts were analyzed using systematic qualitative methods to derive design principles for ASIS interfaces and characterize user roles through personas [5]. Through these methods, relevant workflows, pain points (e.g., commissioning time pressure), and <a href=\"https:\/\/industry-science.com\/en\/articles\/ai-work-system-human-centeredness\/\">HCAI<\/a> gaps were identified.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Method: Translating HCAI challenges into technical solutions<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The authors of the present study, who participated in the original workshops and had access to the interview transcripts, collaboratively analyzed the documented pain points through discussions with the ASIS supplier and stakeholders. Through these iterative exchanges, recurring themes emerged. These were mapped onto the HCAI framework dimensions [1, 2], allowing the translation of identified needs into concrete technical requirements.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Technical approaches were selected for simplicity, comprehensibility, implementability, and model-agnostic application potential. Technical prototypes were developed based on proprietary industrial steel band datasets provided by the ASIS supplier. The datasets consisted of two labeled tabular error datasets comprising features derived from error images and steel band surface images for visual detection tasks. The prototypes were validated through internal testing with the ASIS supplier (using validation metrics and user tests) and customer feedback sessions [5, 6].<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Case study: AI in Automatic Surface Inspection Systems (ASIS)<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">ASIS products are deployed in steel rolling to automatically assess steel band quality and flag possible low-quality bands. The system is first installed by a provider. Then, during commissioning, the AI model is adapted. It can also be retrained later for changing conditions or requirements. The typical workflow for the ASIS customer involves on-site image acquisition, manual error classification for training, real-time computer vision-based error detection, and AI classification [7]. There are three key users: <em>line operators<\/em>, who run production and require immediate feedback about current process quality; <em>quality engineers<\/em>, who monitor whether product quality meets requirements; and <em>ASIS administrators<\/em>, who manage and update data, maintain system results, and optimize performance when necessary.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Applying this methodology to AI in ASIS revealed specific HCAI challenges, presented in <strong>Figure 1<\/strong>.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"798\" height=\"1024\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-798x1024.webp\" alt=\"Figure 1: Challenges for human-centered AI design, quality\" class=\"wp-image-113160\" style=\"width:840px;height:auto\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-798x1024.webp 798w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-292x375.webp 292w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-768x986.webp 768w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-227x292.webp 227w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-1196x1536.webp 1196w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-1595x2048.webp 1595w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-510x655.webp 510w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1-64x82.webp 64w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-1.webp 2000w\" sizes=\"auto, (max-width: 798px) 100vw, 798px\" \/><figcaption class=\"wp-element-caption\"><em>Figure 1: Challenges for human-centered AI design.<\/em><\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">To respond to these challenges, three technical solutions are proposed:<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Solution 1: Anomaly and novel class detection on tabular data<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Unsupervised anomaly detection methods were explored to address data scarcity for rare or novel errors using tabular data [8, 9]. Various algorithms were evaluated on the two industrial tabular datasets, including proximity-based, statistical, linear, and ensemble methods. The most promising approach identified was a combination of a proximity-based algorithm (Connectivity-based Outlier Factor (COF)) [10] and the entropy of predicted class probabilities. This method ranks the \u201cunusualness\u201d of detections relative to the model\u2019s training data based on both sample feature space isolation and classifier confidence.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Anomalies were simulated by omitting samples of rare classes from the dataset (appr. 3% of the data). The mean area under the receiver-operating characteristic (AUROC) was chosen as a metric for investigating the ranking quality and the system\u2019s ability to discriminate between normal and anomalous samples. The combined approach using COF and entropy-based scores yielded the best overall performance across the two tested datasets with AUROCs of 80.36% and 83.14%.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Given the imbalanced nature of error datasets and false positive rates of 37.58% and 35.77% at an 85% true positive rate, the proposed method still suffers from false alarms, but it identifies anomalies and novelties fairly robustly. By identifying anomalies, the system supports data quality awareness and can guide active learning workflows, prompting users to check whether unusual samples are mislabeled or belong to novel classes. This not only enhances user agency but also contributes to responsible knowledge management by reducing the likelihood that flawed or incomplete data becomes embedded in organizational knowledge bases.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Solution 2: Visual anomaly detection on image data<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Complementary to the tabular approach, visual anomaly detection was implemented directly on steel band surface images to improve error localization using zero-shot and semi-supervised few-shot learning, eliminating the need for extensive error examples. AnomalyDINO [11], a patch-based deep nearest-neighbor approach using features from the pre-trained DINOv2 vision transformer, was adapted for the purpose. This method compares visual patches from test images against a memory bank of \u2018good\u2019 surface patches to identify deviations. For ASIS, reference patches contain only error-free steel band images.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Internal evaluation on two steel band datasets demonstrated strong performance, achieving AUROC scores of 98.45% and 97.0-98.52%, with performance improving from 16 to 100 reference shots. Zero-shot analysis reached 99.13% AUROC through unsupervised anomaly detection and quantile optimization on a different proprietary dataset. However, the approach has difficulty detecting very small anomalies like oil stains or scratches, which may not sufficiently elevate anomaly scores, and can profit from further hyperparameter search.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Solution 3: Label error detection&nbsp;<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">An algorithm was developed to improve dataset integrity and trust in model evaluations by identifying samples in the training and validation data that are likely to be labelled incorrectly. The Area Under the Margin (AUM) [12] ranking method for gradient-boosted decision tree models was adapted for the purpose [6]. The method calculates a score for each sample by averaging the margin between the predicted probability of its assigned class and the next most likely class. This takes place across all boosting steps during a single training run.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">AUM proved computationally efficient, requiring only one training run. On the two industrial tabular datasets, AUM achieved AUROC values of 97.1% and 96.2% for detecting 5% injected synthetic asymmetric label noise [13], comparable to more computationally expensive out-of-sample prediction methods [6]. In two real-world validation sessions with industry practitioners, 42% of the samples with the lowest AUM scores were confirmed as label errors, while others pointed to further data quality issues, validating the method\u2019s practical utility.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Presenting low-AUM samples to quality engineers thus enables targeted data label review and correction. This improves user agency and augmentation by allowing engineers to curate data more efficiently and with reduced manual burden, which in turn supports responsible knowledge management by reducing the risk that organizational knowledge bases are corrupted by mislabeled data. It also helps foster consensus on error classification criteria.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"323\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2-1024x323.webp\" alt=\"Figure 2: Integration of the solutions.\" class=\"wp-image-113162\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2-1024x323.webp 1024w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2-764x241.webp 764w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2-768x242.webp 768w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2-514x162.webp 514w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2-1536x485.webp 1536w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2-510x161.webp 510w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2-64x20.webp 64w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-2.webp 2000w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\"><em>Figure 2: Integration of the solutions.<\/em><\/figcaption><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\">Integration outlook&nbsp;<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">These solutions were integrated into existing ASIS software and workflows (<strong>Fig. 2<\/strong>). Anomaly and label error scores are displayed alongside error detections in tabular views used by quality engineers, allowing sorting and filtering to prioritize review tasks. Anomaly localization can be overlaid on steel band images or used for error localization. To make model reliability information accessible and minimize operator information overload, a simple \u201ctraffic light\u201d indicator could be used, which would synthesize information from multiple sources (e.g. anomaly score distributions and model confidence levels) into a glanceable AI reliability assessment for real-time monitoring interfaces. This could be displayed in the operator\u2019s real-time monitoring interface.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Combining detailed scores for expert review with high-level indicators for operators supports informed decision-making across roles: the traffic light visualization reduces cognitive burden by lowering the volume and complexity of information that operators must process, which contributes to sustainable work design and shifts their focus to important events (augmentation). Meanwhile, detailed anomaly and label error scores enhance the trust of engineers and equip them for responsible knowledge management in data-driven decisions.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Discussion and conclusion<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The co-developed prototypes, validated through lab simulations and real-world user feedback sessions, demonstrate that addressing data and model quality via an HCAI lens supports user agency and improves system resilience in industrial ASIS. User feedback from validation sessions confirmed the value of the label error detection solution. Ranking potential label errors clearly reduced the manual effort required for data quality assurance compared to non-targeted manual checks. This is demonstrated by the 42% identification rate based on label error ranking compared to regular label error prevalence rates (e.g., less than 5%) [6]. Data-efficient image-level anomaly detection proved effective at detecting surface defects on customer data.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Overall, technical solutions focused on data quality transparency are crucial for building trustworthy, resilient industrial AI. They bridge algorithmic capabilities and practical usability by empowering domain experts to understand, validate, and improve AI performance. In doing so, they directly enhance agency and augmentation (through more efficient workflows), trustworthiness (via transparent reliability monitoring), and responsible knowledge management (by improving data integrity). Strategies for ethical and effective HCAI are thereby shown to transfer beyond industrial ASIS, as outlined in <strong>Figure 3<\/strong>.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large is-resized\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"784\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3-1024x784.webp\" alt=\"Figure 3: Transferable HCAI strategies.\" class=\"wp-image-113164\" style=\"width:606px;height:auto\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3-1024x784.webp 1024w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3-490x375.webp 490w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3-768x588.webp 768w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3-381x292.webp 381w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3-1536x1176.webp 1536w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3-510x390.webp 510w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3-64x49.webp 64w, https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Manakidis_I4S-26-1_Figure-3.webp 2000w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\"><em>Figure 3: Transferable HCAI strategies.<\/em><\/figcaption><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">These strategies are relevant to domains requiring dependable AI with robust human oversight, such as safety-critical applications where data scarcity and reliability are an issue. By focusing on human-centered principles and co-developing technical solutions for data quality and reliability monitoring, this science-practice collaboration demonstrates how AI systems can become transparent, trustworthy collaborators that augment human capabilities.&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em>This research and development project is funded by the German Federal Ministry of Research, Technology and Space (BMFTR) within the \u201cThe Future of Value Creation \u2013 Research on Production, Services and Work\u201d program (02L19C200) and managed by the Project Management Agency Karlsruhe (PTKA). The authors are responsible for the content of this publication.<\/em><\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>This is an original article. The German translation can be accessed via <a href=\"https:\/\/doi.org\/10.30844\/I4SD.26.1.128\" target=\"_blank\" rel=\"noopener\">DOI:\u00a010.30844\/I4SD.26.1.128<\/a><\/strong><\/p>\n<div id=\"download-section\" class=\"gito-pub-download-section\" style=\"text-align:center;margin:20px;\"><h2>Your downloads<\/h2><button style=\"font-size:14px;margin-right:15px;\" class=\"button gito-pub-cpt-download-button\" data-postid=\"113159\" data-userid =\"0\" data-filename=\"I4S_01-2026_DE_Rath-Manakidis2.pdf\"><span style=\"margin-top:5px !important;\" class=\"dashicons dashicons-download\"><\/span>&nbsp;&nbsp;PDF (DE)<\/button><button style=\"font-size:14px;margin-right:15px;\" class=\"button gito-pub-cpt-download-button\" data-postid=\"113159\" data-userid =\"0\" data-filename=\"I4S_01-2026_ENG_Rath-Manakidis2.pdf\"><span style=\"margin-top:5px !important;\" class=\"dashicons dashicons-download\"><\/span>&nbsp;&nbsp;PDF (EN)<\/button><\/div><br>Potentials: <span class=\"gito-pub-tag-element\"><a href=\"\/potentials\/dynamics\/\">Dynamics<\/a><\/span> \n<h2 class=\"gito-pub-frontend-post-headline\">You might also be interested in<\/h2>\n<!-- GITO_PUB_POST start flex-container -->\n<div class=\"gito-pub-flex-container\">\n   <div class=\"gito-pub-frontend-post-card gito-pub-flex-item gito-pub-flex-item-1\">\n      <a href=\"https:\/\/industry-science.com\/en\/articles\/continuing-vocational-education\/\">\n         <div class=\"gito-pub-frontend-post-card-row\">         <div class=\"gito-pub-frontend-post-card-column gito-pub-frontend-post-card-column-image\">\n            <picture>\n               <source media=\"(max-width:640px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/09\/beichter_AdobeStock_1885327612_master1305-640x325.webp\">\n               <source media=\"(min-width:641px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/09\/beichter_AdobeStock_1885327612_master1305-196x180.webp\">\n               <img decoding=\"async\" class=\"gito-pub-frontend-post-card-image\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/09\/beichter_AdobeStock_1885327612_master1305-196x180.webp\" alt=\"Audio-Immersive Learning in Continuing Vocational Education\">\n            <\/picture>\n         <\/div>\n            <div class=\"gito-pub-frontend-post-card-column\">               <div class=\"ellipsis\" style=\"height:166px !important;overflow:hidden;\" title=\"Audio-Immersive Learning in Continuing Vocational Education\">                  <table class=\"gito-pub-frontend-post-card-header\">\n            \t     <tr>\n                        <td>                  \t\t   <h4 class=\"gito-pub-frontend-post-card-title\" style=\"line-height:1.2em;\">Audio-Immersive Learning in Continuing Vocational Education<\/h4>\n                        <div class=\"gito-pub-frontend-post-card-subtitle\">From linear audio playback to AI-supported conversational learning companions<\/div>                        <div class=\"gito-pub-frontend-post-card-author\"><a href=\"https:\/\/industry-science.com\/en\/authors\/tim-beichter\/\">Tim Beichter<\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/vanessa-hartmann\/\">Vanessa Hartmann<\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/katharina-hoelzle\/\">Katharina H\u00f6lzle<\/a> <a href=\"https:\/\/orcid.org\/0000-0002-9733-4650\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/manuel-kaiser\/\">Manuel Kaiser<\/a><\/div>\n                        <\/td>\n                     <\/tr>\n                  <\/table>\n                  <div class=\"gito-pub-frontend-post-card-text\">\n                     Artificial intelligence is increasingly shaping continuing vocational education and training by enabling the personalization of individual learning experiences. At the same time, audio-based learning formats are attracting growing interest among learners because of their flexibility and suitability for workplace learning. However, a conceptual framework for AI-supported audio learning, as well as the potential of combining artificial intelligence with audio-based learning, has received little attention to date. This paper therefore presents a conceptual perspective on the design possibilities and educational potential of AI-supported audio learning formats.                  <\/div>\n               <\/div>\n               <div class=\"gito-pub-frontend-post-card-scientific\"><strong>Industry 4.0 Science<\/strong> | Volume 42 | 2026 | Edition 5 | Pages 102-108 | DOI <a style=\"font-weight:bold !important;\" href=\"https:\/\/doi.org\/10.30844\/I4SE.26.5.12\" target=\"_blank\" rel=\"noopener\">10.30844\/I4SE.26.5.12<\/a><\/div>            <\/div>\n         <\/div>\n      <\/a>\n   <\/div>\n   <div class=\"gito-pub-frontend-post-card gito-pub-flex-item gito-pub-flex-item-1\">\n      <a href=\"https:\/\/industry-science.com\/en\/articles\/ai-demonstrators-manufacturing\/\">\n         <div class=\"gito-pub-frontend-post-card-row\">         <div class=\"gito-pub-frontend-post-card-column gito-pub-frontend-post-card-column-image\">\n            <picture>\n               <source media=\"(max-width:640px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/link_AdobeStock_311608924_Gorodenkoff-640x325.webp\">\n               <source media=\"(min-width:641px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/link_AdobeStock_311608924_Gorodenkoff-196x180.webp\">\n               <img decoding=\"async\" class=\"gito-pub-frontend-post-card-image\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/link_AdobeStock_311608924_Gorodenkoff-196x180.webp\" alt=\"Explaining AI in Industrial Production in an Accessible Way\">\n            <\/picture>\n         <\/div>\n            <div class=\"gito-pub-frontend-post-card-column\">               <div class=\"ellipsis\" style=\"height:166px !important;overflow:hidden;\" title=\"Explaining AI in Industrial Production in an Accessible Way\">                  <table class=\"gito-pub-frontend-post-card-header\">\n            \t     <tr>\n                        <td>                  \t\t   <h4 class=\"gito-pub-frontend-post-card-title\" style=\"line-height:1.2em;\">Explaining AI in Industrial Production in an Accessible Way<\/h4>\n                        <div class=\"gito-pub-frontend-post-card-subtitle\">Requirements for AI demonstrators to promote acceptance<\/div>                        <div class=\"gito-pub-frontend-post-card-author\"><a href=\"https:\/\/industry-science.com\/en\/authors\/jennifer-link-en\/\">Jennifer Link<\/a> <a href=\"https:\/\/orcid.org\/0009-0005-2407-3495\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/markus-harlacher-en\/\">Markus Harlacher<\/a> <a href=\"https:\/\/orcid.org\/0009-0007-5817-2920\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/colin-srebny\/\">Colin Srebny<\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/sascha-stowasser-en\/\">Sascha Stowasser<\/a> <a href=\"https:\/\/orcid.org\/0009-0006-2725-5793\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a><\/div>\n                        <\/td>\n                     <\/tr>\n                  <\/table>\n                  <div class=\"gito-pub-frontend-post-card-text\">\n                     Artificial intelligence (AI) offers a wide range of possibilities in industrial production, but it also presents challenges regarding employee acceptance. AI demonstrators are therefore of central importance, as they enable hands-on experience with AI. However, there has been a lack of systematically identified requirements for demonstrators that specifically promote acceptance and address negative emotions. Using a multi-stage research design, 69 requirements were identified, structured into functional requirements, quality requirements, and boundary conditions.                  <\/div>\n               <\/div>\n               <div class=\"gito-pub-frontend-post-card-scientific\"><strong>Industry 4.0 Science<\/strong> | Volume 42 | 2026 | Edition 5 | Pages 6-14 | DOI <a style=\"font-weight:bold !important;\" href=\"https:\/\/doi.org\/10.30844\/I4SE.26.5.1\" target=\"_blank\" rel=\"noopener\">10.30844\/I4SE.26.5.1<\/a><\/div>            <\/div>\n         <\/div>\n      <\/a>\n   <\/div>\n   <div class=\"gito-pub-frontend-post-card gito-pub-flex-item gito-pub-flex-item-1\">\n      <a href=\"https:\/\/industry-science.com\/en\/articles\/ai-based-building-inspection\/\">\n         <div class=\"gito-pub-frontend-post-card-row\">         <div class=\"gito-pub-frontend-post-card-column gito-pub-frontend-post-card-column-image\">\n            <picture>\n               <source media=\"(max-width:640px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/sender_AdobeStock_227079093_Aisyaqilumar-640x325.webp\">\n               <source media=\"(min-width:641px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/sender_AdobeStock_227079093_Aisyaqilumar-196x180.webp\">\n               <img decoding=\"async\" class=\"gito-pub-frontend-post-card-image\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/sender_AdobeStock_227079093_Aisyaqilumar-196x180.webp\" alt=\"AI-Based Building Inspection for Large Structures\">\n            <\/picture>\n         <\/div>\n            <div class=\"gito-pub-frontend-post-card-column\">               <div class=\"ellipsis\" style=\"height:166px !important;overflow:hidden;\" title=\"AI-Based Building Inspection for Large Structures\">                  <table class=\"gito-pub-frontend-post-card-header\">\n            \t     <tr>\n                        <td>                  \t\t   <h4 class=\"gito-pub-frontend-post-card-title\" style=\"line-height:1.2em;\">AI-Based Building Inspection for Large Structures<\/h4>\n                        <div class=\"gito-pub-frontend-post-card-subtitle\">A new approach to construction progress monitoring<\/div>                        <div class=\"gito-pub-frontend-post-card-author\"><a href=\"https:\/\/industry-science.com\/en\/authors\/jan-sender-en\/\">Jan Sender<\/a> <a href=\"https:\/\/orcid.org\/0009-0003-9697-5709\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/konrad-jagusch-en\/\">Konrad Jagusch<\/a> <a href=\"https:\/\/orcid.org\/0009-0001-7454-1657\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/michael-geist\/\">Michael Geist<\/a> <a href=\"https:\/\/orcid.org\/0009-0005-2780-7538\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/david-jericho\/\">David Jericho<\/a> <a href=\"https:\/\/orcid.org\/0009-0001-8932-8701\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/christian-scharr\/\">Christian Scharr<\/a> <a href=\"https:\/\/orcid.org\/0009-0003-6300-4682\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a><\/div>\n                        <\/td>\n                     <\/tr>\n                  <\/table>\n                  <div class=\"gito-pub-frontend-post-card-text\">\n                     Monitoring construction progress, as required in the one-off production of large structures, is very time- and labor-intensive due to a high level of complexity and individuality. The goal of this article is to develop a sensor-based approach for capturing and evaluating multiple inspection characteristics. The use of machine learning models to detect objects and derive relevant information forms the basis for linking current condition to construction schedule. This enables a significant increase in efficiency during construction progress monitoring and a well-founded assessment of progress.                  <\/div>\n               <\/div>\n               <div class=\"gito-pub-frontend-post-card-scientific\"><strong>Industry 4.0 Science<\/strong> | Volume 42 | 2026 | Edition 5 | Pages 78-84 | DOI <a style=\"font-weight:bold !important;\" href=\"https:\/\/doi.org\/10.30844\/I4SE.26.5.9\" target=\"_blank\" rel=\"noopener\">10.30844\/I4SE.26.5.9<\/a><\/div>            <\/div>\n         <\/div>\n      <\/a>\n   <\/div>\n   <div class=\"gito-pub-frontend-post-card gito-pub-flex-item gito-pub-flex-item-1\">\n      <a href=\"https:\/\/industry-science.com\/en\/articles\/work-design-autonomous-systems\/\">\n         <div class=\"gito-pub-frontend-post-card-row\">         <div class=\"gito-pub-frontend-post-card-column gito-pub-frontend-post-card-column-image\">\n            <picture>\n               <source media=\"(max-width:640px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/AdobeStock_484184873_Ivan-Traimak-640x325.webp\">\n               <source media=\"(min-width:641px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/AdobeStock_484184873_Ivan-Traimak-196x180.webp\">\n               <img decoding=\"async\" class=\"gito-pub-frontend-post-card-image\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/AdobeStock_484184873_Ivan-Traimak-196x180.webp\" alt=\"Work Design in the Use of Autonomous Systems\">\n            <\/picture>\n         <\/div>\n            <div class=\"gito-pub-frontend-post-card-column\">               <div class=\"ellipsis\" style=\"height:166px !important;overflow:hidden;\" title=\"Work Design in the Use of Autonomous Systems\">                  <table class=\"gito-pub-frontend-post-card-header\">\n            \t     <tr>\n                        <td>                  \t\t   <h4 class=\"gito-pub-frontend-post-card-title\" style=\"line-height:1.2em;\">Work Design in the Use of Autonomous Systems<\/h4>\n                        <div class=\"gito-pub-frontend-post-card-subtitle\">Addressing the shortage of skilled workers<\/div>                        <div class=\"gito-pub-frontend-post-card-author\"><a href=\"https:\/\/industry-science.com\/en\/authors\/tim-jeske-en\/\">Tim Jeske<\/a> <a href=\"https:\/\/orcid.org\/0000-0001-8778-6824\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/sascha-stowasser-en\/\">Sascha Stowasser<\/a> <a href=\"https:\/\/orcid.org\/0009-0006-2725-5793\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/nicole-ottersboeck-en\/\">Nicole Ottersb\u00f6ck<\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/sebastian-terstegen-en\/\">Sebastian Terstegen<\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/rasmus-adler\/\">Rasmus Adler<\/a> <a href=\"https:\/\/orcid.org\/0000-0002-7482-7102\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a><\/div>\n                        <\/td>\n                     <\/tr>\n                  <\/table>\n                  <div class=\"gito-pub-frontend-post-card-text\">\n                     Companies are increasingly challenged to address shortages of skilled workers while meeting rising demands for productivity, flexibility, and innovation. Because labor supply can only be expanded to a limited extent, there is a growing focus on designing work systems with productivity in mind. Autonomous systems offer significant potential in this regard. Their implementation requires not only technical adjustments but, above all, changes in organization, skills, and work design. This article analyzes empirically grounded change requirements in existing work systems as well as associated economic potential.                  <\/div>\n               <\/div>\n               <div class=\"gito-pub-frontend-post-card-scientific\"><strong>Industry 4.0 Science<\/strong> | Volume 42 | 2026 | Edition 5 | Pages 44-50 | DOI <a style=\"font-weight:bold !important;\" href=\"https:\/\/doi.org\/10.30844\/I4SE.26.5.5\" target=\"_blank\" rel=\"noopener\">10.30844\/I4SE.26.5.5<\/a><\/div>            <\/div>\n         <\/div>\n      <\/a>\n   <\/div>\n   <div class=\"gito-pub-frontend-post-card gito-pub-flex-item gito-pub-flex-item-1\">\n      <a href=\"https:\/\/industry-science.com\/en\/articles\/complementors-digital-ecosystems\/\">\n         <div class=\"gito-pub-frontend-post-card-row\">         <div class=\"gito-pub-frontend-post-card-column gito-pub-frontend-post-card-column-image\">\n            <picture>\n               <source media=\"(max-width:640px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/Zabel_AdobeStock_260585096_radachynskyi-640x325.webp\">\n               <source media=\"(min-width:641px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/Zabel_AdobeStock_260585096_radachynskyi-196x180.webp\">\n               <img decoding=\"async\" class=\"gito-pub-frontend-post-card-image\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/08\/Zabel_AdobeStock_260585096_radachynskyi-196x180.webp\" alt=\"Cooperation Routines of Complementors in Digital Ecosystems\">\n            <\/picture>\n         <\/div>\n            <div class=\"gito-pub-frontend-post-card-column\">               <div class=\"ellipsis\" style=\"height:166px !important;overflow:hidden;\" title=\"Cooperation Routines of Complementors in Digital Ecosystems\">                  <table class=\"gito-pub-frontend-post-card-header\">\n            \t     <tr>\n                        <td>                  \t\t   <h4 class=\"gito-pub-frontend-post-card-title\" style=\"line-height:1.2em;\">Cooperation Routines of Complementors in Digital Ecosystems<\/h4>\n                        <div class=\"gito-pub-frontend-post-card-subtitle\">A microfoundation of integrative dynamic capability<\/div>                        <div class=\"gito-pub-frontend-post-card-author\"><a href=\"https:\/\/industry-science.com\/en\/authors\/christian-zabel-en\/\">Christian Zabel<\/a> <a href=\"https:\/\/orcid.org\/0000-0002-4636-6679\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/tahir-schmidt\/\">Tahir Schmidt<\/a> <a href=\"https:\/\/orcid.org\/0009-0004-2409-6665\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a><\/div>\n                        <\/td>\n                     <\/tr>\n                  <\/table>\n                  <div class=\"gito-pub-frontend-post-card-text\">\n                     Complementors are central to value creation in digital ecosystems yet have limited leverage and must adapt through dynamic capabilities. Building on the Profiting From Innovation Framework, this study examines how integrative capabilities manifest for complementors through cooperative routines. Based on a systematic literature review of Scopus-indexed studies from 2020 to mid-2025 focusing on the microfoundation \u201corchestrating ecosystem actors\u201d, we identify two routine clusters. Complementors cooperate with other complementors via partner sensing, scouting, coalitions, resource sharing, and risk allocation while protecting critical assets. They cooperate with platform owners via multichannel boundary spanning, quality signaling, governance compliance, boundary resource integration, and co-development, while facing the risk of owner entry. Research gaps concern the formalization of cooperation routines, taxonomy, and B2B contexts.                  <\/div>\n               <\/div>\n               <div class=\"gito-pub-frontend-post-card-scientific\"><strong>Industry 4.0 Science<\/strong> | Volume 42 | 2026 | Edition 4 | Pages 22-28 | DOI <a style=\"font-weight:bold !important;\" href=\"https:\/\/doi.org\/10.30844\/I4SE.26.4.3\" target=\"_blank\" rel=\"noopener\">10.30844\/I4SE.26.4.3<\/a><\/div>            <\/div>\n         <\/div>\n      <\/a>\n   <\/div>\n   <div class=\"gito-pub-frontend-post-card gito-pub-flex-item gito-pub-flex-item-1\">\n      <a href=\"https:\/\/industry-science.com\/en\/articles\/b2b-collaboration-platforms-smes\/\">\n         <div class=\"gito-pub-frontend-post-card-row\">         <div class=\"gito-pub-frontend-post-card-column gito-pub-frontend-post-card-column-image\">\n            <picture>\n               <source media=\"(max-width:640px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/07\/Schaefer_AdobeStock_205809895_YiuCheung-640x325.webp\">\n               <source media=\"(min-width:641px)\" srcset=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/07\/Schaefer_AdobeStock_205809895_YiuCheung-196x180.webp\">\n               <img decoding=\"async\" class=\"gito-pub-frontend-post-card-image\" src=\"https:\/\/industry-science.com\/wp-content\/uploads\/2026\/07\/Schaefer_AdobeStock_205809895_YiuCheung-196x180.webp\" alt=\"Platform Adoption as a Dynamic Capability\">\n            <\/picture>\n         <\/div>\n            <div class=\"gito-pub-frontend-post-card-column\">               <div class=\"ellipsis\" style=\"height:166px !important;overflow:hidden;\" title=\"Platform Adoption as a Dynamic Capability\">                  <table class=\"gito-pub-frontend-post-card-header\">\n            \t     <tr>\n                        <td>                  \t\t   <h4 class=\"gito-pub-frontend-post-card-title\" style=\"line-height:1.2em;\">Platform Adoption as a Dynamic Capability<\/h4>\n                        <div class=\"gito-pub-frontend-post-card-subtitle\">How SMEs overcome barriers to adoption of B2B collaboration platforms<\/div>                        <div class=\"gito-pub-frontend-post-card-author\"><a href=\"https:\/\/industry-science.com\/en\/authors\/nikolai-schaefer\/\">Nikolai Sch\u00e4fer<\/a> <a href=\"https:\/\/orcid.org\/0009-0009-9218-4890\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a>, <a href=\"https:\/\/industry-science.com\/en\/authors\/marcel-huelsbeck\/\">Marcel H\u00fclsbeck<\/a> <a href=\"https:\/\/orcid.org\/0000-0003-4846-3533\" target=\"_blank\" title=\"ORCID eintrag \u00f6ffnen.\" rel=\"noopener\">\n        <img decoding=\"async\" src=\"https:\/\/orcid.org\/assets\/vectors\/orcid.logo.icon.svg\" alt=\"ORCID Icon\" style=\"width:16px;height:16px;vertical-align:middle;\"><\/a><\/div>\n                        <\/td>\n                     <\/tr>\n                  <\/table>\n                  <div class=\"gito-pub-frontend-post-card-text\">\n                     Digital collaboration and innovation platforms offer SMEs significant potential to compensate for structural resource weaknesses and to participate in innovation ecosystems. Nevertheless, adoption in the B2B context remains low. This paper examines adoption barriers based on a systematic literature review using Teece\u2019s dynamic capabilities approach. The analysis suggests that recurring obstacles can be structured along three dimensions: Sensing\u2014lack of ecosystem awareness, absence of scanning routines; Seizing\u2014IP concerns, governance uncertainty, adoption fatigue; Reconfiguring\u2014closed-innovation culture, lack of absorptive capacity. Building on this, a practice-oriented capability-building framework is developed with recommendations for action.                  <\/div>\n               <\/div>\n               <div class=\"gito-pub-frontend-post-card-scientific\"><strong>Industry 4.0 Science<\/strong> | Volume 42 | 2026 | Edition 4 | Pages 42-48 | DOI <a style=\"font-weight:bold !important;\" href=\"https:\/\/doi.org\/10.30844\/I4SE.26.4.5\" target=\"_blank\" rel=\"noopener\">10.30844\/I4SE.26.4.5<\/a><\/div>            <\/div>\n         <\/div>\n      <\/a>\n   <\/div>\n<\/div>\n<!-- GITO_PUB_POST end flex-container -->\n","protected":false},"excerpt":{"rendered":"<p>AI implementation transforms work and worker-technology relationships in industrial quality control. This paper explores how approaches to data quality and model transparency support ethical AI deployment, fostering worker agency, trust, and sustainable work design in automatic surface inspection systems (ASIS). Recurring problems like data inefficiency, variable model confidence, and limited AI expertise point to key challenges of human-centered AI: user trust, agency and responsible data management. A solution co-developed with an ASIS supplier demonstrates that the challenges extend beyond the purely technical, underscoring the value of AI design that augments human capabilities. Technical solutions such as anomaly, label error, and domain drift detection are proposed to enhance data quality and model reliability. The insights emphasize the following generalizable strategies for resilient AI integration: understanding user-reported problems through a human-AI interaction lens, focusing on data quality transparency, providing intuitive reliability monitoring, and ensuring user-centric integration into existing workflows.<\/p>\n","protected":false},"featured_media":113095,"menu_order":0,"template":"","categories":[79167,79298],"tags":[],"product_cat":[79304],"topic":[67617,79489],"technology":[67790],"knowhow":[],"industry":[],"writer":[85739,85779,85822],"content-type":[83932],"potential":[68923],"solution":[],"glossary":[],"class_list":["post-113159","article","type-article","status-publish","has-post-thumbnail","category-design-en","category-typeset","product_cat-articles","topic-adaptability","topic-quality","technology-artificial-intelligence","writer-bjoern-kraemer","writer-henry-huick","writer-laurenz-wiskott","content-type-article","potential-dynamics","product","first","instock","downloadable","virtual","sold-individually","taxable","purchasable","product-type-article"],"uagb_featured_image_src":{"full":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg.jpeg",1400,788,false],"thumbnail":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-150x150.jpeg",150,150,true],"medium":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-666x375.jpeg",666,375,true],"medium_large":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-768x432.jpeg",768,432,true],"large":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-1024x576.jpeg",1020,574,true],"front-page-entry":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-1032x320.jpeg",1032,320,true],"post-entry":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-764x376.jpeg",764,376,true],"post-teaser":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-392x320.jpeg",392,320,true],"post-teaser-mobile":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-608x496.jpeg",608,496,true],"post-custom-size":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-640x325.jpeg",640,325,true],"whitepaper-teaser":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-274x376.jpeg",274,376,true],"card-big":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-514x292.jpeg",514,292,true],"card-portrait":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-320x440.jpeg",320,440,true],"card-big-company":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-514x289.jpeg",514,289,true],"gp-listing":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-196x180.jpeg",196,180,true],"1536x1536":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg.jpeg",1400,788,false],"2048x2048":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg.jpeg",1400,788,false],"woocommerce_thumbnail":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-510x510.jpeg",510,510,true],"woocommerce_single":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-510x287.jpeg",510,287,true],"woocommerce_gallery_thumbnail":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-100x100.jpeg",100,100,true],"dgwt-wcas-product-suggestion":["https:\/\/industry-science.com\/wp-content\/uploads\/2026\/02\/Rath_AdobeStock_1861900994_Framestock.jpg-64x36.jpeg",64,36,true]},"uagb_author_info":{"display_name":"Florian Goldmann","author_link":"https:\/\/industry-science.com\/en\/author\/"},"uagb_comment_info":0,"uagb_excerpt":"AI implementation transforms work and worker-technology relationships in industrial quality control. This paper explores how approaches to data quality and model transparency support ethical AI deployment, fostering worker agency, trust, and sustainable work design in automatic surface inspection systems (ASIS). Recurring problems like data inefficiency, variable model confidence, and limited AI expertise point to key&hellip;","_links":{"self":[{"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/article\/113159","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/article"}],"about":[{"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/types\/article"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/media\/113095"}],"wp:attachment":[{"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/media?parent=113159"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/categories?post=113159"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/tags?post=113159"},{"taxonomy":"product_cat","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/product_cat?post=113159"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/topic?post=113159"},{"taxonomy":"technology","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/technology?post=113159"},{"taxonomy":"knowhow","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/knowhow?post=113159"},{"taxonomy":"industry","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/industry?post=113159"},{"taxonomy":"writer","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/writer?post=113159"},{"taxonomy":"content-type","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/content-type?post=113159"},{"taxonomy":"potential","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/potential?post=113159"},{"taxonomy":"solution","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/solution?post=113159"},{"taxonomy":"glossary","embeddable":true,"href":"https:\/\/industry-science.com\/en\/wp-json\/wp\/v2\/glossary?post=113159"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}