Dataset for "A Smart Manufacturing Approach to Enhancing the Reliability of Visual Inspection in the Automotive Industry"

Open Data The dataset supporting the findings of this study is openly available in the Zenodo repository under the title “A Smart Manufacturing Approach to Enhancing the Reliability of Visual Inspection in the Automotive Industry”, DOI: 10.5281/zenodo.17649175. The dataset includes anonymized production-inspection records collected during the smart-manufacturing improvement initiative in the automotive industry. Specifically, it comprises monthly data on loading time, downtime, retry time, operating time, total output, cycle time, and calculated OEE components (Availability, Performance, and Quality rates). It also contains sensor capability index (Cpk) measurements reflecting the accuracy and stability of the camera-based visual inspection system before and after the intervention. All data have been fully anonymized and do not contain product codes, machine identifiers, operational timestamps, financial indicators, or any personally identifiable information (PII). The dataset is strictly technical in nature and is statistically consistent with the reliability and performance trends reported in the study, thereby enabling transparency, reproducibility, and methodological validation while adhering to ethical and corporate data-governance requirements. Open Contributorship This research was conducted collaboratively by the author team from the Industrial Engineering Department, BINUS Online Learning, Bina Nusantara University, following the Open Contributorship principles and the CRediT taxonomy. Andryan Nugraha contributed to Conceptualization, Problem Formulation, Methodology, Data Curation, and Formal Analysis, including OEE computation, downtime evaluation, sensor capability assessment (Cpk), and pre and post improvement studies, ensuring data consistency, completeness, and anonymization for open-data publication. Hubertus Davy Yulianto provided Supervision, Methodological Validation, and Review of Analytical Rigor, ensuring alignment with industrial engineering principles, visual inspection system design, and applied manufacturing research standards. Irma Ratna Avianti was responsible for Technical Review, Quality Assurance, Refinement of Interpretation, and Writing, including Review and Editing, guiding discussion formulation, validation of inspection system improvements, and adherence to academic writing and publication standards. The author team collectively handled Visualization, Manuscript Finalization, and Dataset Documentation, ensuring coherent presentation of results, figures, and supplementary materials for Zenodo publication. All authors have read and approved the final manuscript and take full responsibility for its accuracy, integrity, and scholarly quality. The authors endorse the Open Contributorship framework and commit to making associated datasets, analytical templates, and methodological resources openly accessible to support replication, industrial application, and wider academic dissemination.开放数据集 本研究结论所依托的数据集可于Zenodo仓储中公开获取，标题为《面向提升汽车行业视觉检测可靠性的智能制造方法》，数字对象标识符（Digital Object Identifier，DOI）：10.5281/zenodo.17649175。 该数据集包含汽车行业智能制造改进项目期间收集的匿名化生产检测记录。具体而言，其涵盖装载时长、停机时长、重试时长、运行时长、总产出、循环时长以及经计算得到的设备综合效率（Overall Equipment Effectiveness，简称OEE）各组成项（可用性、性能及质量率）的月度数据。此外，数据集还包含基于摄像头的视觉检测系统在干预前后的传感器过程能力指数（Capability Process Index，简称Cpk）测量值，用以反映其精度与稳定性。所有数据均已完成完全匿名化处理，未包含产品代码、机器标识符、运营时间戳、财务指标或任何个人可识别信息（Personally Identifiable Information，简称PII）。本数据集仅包含技术类内容，且在统计层面与本研究报告的可靠性及性能趋势保持一致，可在符合伦理与企业数据治理要求的前提下，提升研究透明度、可重复性与方法学验证性。 开放贡献机制 本研究由印度尼西亚宾那纳斯塔大学（Bina Nusantara University）在线学习学院工业工程系的作者团队协作完成，遵循开放贡献原则与作者贡献分类法（Contributor Roles Taxonomy，简称CRediT）。 安德里扬·努格拉哈（Andryan Nugraha）负责概念构建、问题梳理、方法学设计、数据管理与正式分析工作，包括设备综合效率计算、停机时长评估、传感器过程能力指数（Cpk）测评以及改进前后的对照研究，确保数据的一致性、完整性与匿名化处理，以符合开放数据出版要求。 胡贝图斯·戴维·尤利安托（Hubertus Davy Yulianto）提供研究指导、方法学验证与分析严谨性审查工作，确保研究符合工业工程原理、视觉检测系统设计规范与应用制造研究标准。 伊尔玛·拉特纳·阿维安蒂（Irma Ratna Avianti）负责技术审查、质量保证、解读优化与论文撰写工作，包括审阅与编辑、指导讨论框架构建、验证检测系统改进效果，并遵循学术写作与出版规范。 全体作者共同负责可视化呈现、稿件终稿定稿与数据集文档编制工作，确保研究结果、图表及补充材料以连贯清晰的形式呈现，以供Zenodo仓储出版使用。 所有作者均已阅读并批准最终稿件，并对稿件的准确性、完整性与学术质量承担全部责任。作者们认可开放贡献框架，并承诺将相关数据集、分析模板与方法学资源公开共享，以支持研究复制、工业应用及更广泛的学术传播。