製程巡檢為維持製造品質與作業安全之重要機制，然而現有數位巡檢系統多著重於任務派送、表單填報與結果查詢，巡檢規劃、執行查核、異常分析、改善追蹤及知識累積之間仍缺乏完整連結。此外，巡檢資料未能充分轉化為異常判斷與作業支援資訊，相關規範、案例及處理經驗亦分散於不同資料來源，限制巡檢資料與知識的後續應用。為解決上述問題，本研究旨在建構一套結合數據分析、智慧型代理人與知識工程之智慧型製程巡檢模式，以強化巡檢流程整合、異常辨識及知識應用能力。
本研究依循戴明循環，建構涵蓋巡檢規劃、現場執行、結果查核、異常通報、績效評量、改善追蹤與知識回饋之閉環流程。巡檢分析機制整合Isolation Forest、核密度估計、ADWIN與CUSUM，分別支援多變量離群、資料密度分布與低密度區域、狀態變化及累積偏移之辨識，並將分析結果定位為供人員進一步查核之候選警示。智慧型代理人以主控協調代理人作為單一互動入口，依據使用者需求進行任務辨識與路由，調用巡檢資料、分析工具及知識服務；知識架構則整合關聯式資料庫、向量資料庫與知識圖譜，並透過人工審核機制，控管候選經驗知識納入正式可檢索知識集合。
本研究完成智慧型製程巡檢原型系統，並以模擬巡檢資料、代表性作業情境及70題代理人測試題集進行驗證。技術驗證結果顯示，系統可完成巡檢資料之跨模組流轉與預定狀態轉換；代理人任務分流之Micro F1與Macro F1分別為0.852與0.858，知識圖譜路徑正確率與參數完全匹配率分別為0.950與0.850，安全回答規則及候選知識狀態控制亦均依預定流程完成。在應用層面，本研究所建置之系統可將巡檢流程、異常分析結果、規範知識與歷史經驗加以連結，提供異常查核、原因追溯、處置參考及知識回饋等作業支援。綜合而言，本研究驗證所提出模式在巡檢流程整合、異常分析、代理人協作與知識應用方面之技術可行性，可作為後續導入實際製造場域、進行模型參數校準及評估長期應用成效之基礎。

Process inspection is an essential mechanism for maintaining manufacturing quality and operational safety. However, existing digital inspection systems primarily focus on task assignment, form completion, and result retrieval, while inspection planning, execution verification, anomaly analysis, improvement tracking, and knowledge accumulation remain insufficiently integrated. Moreover, inspection data are not fully transformed into information that supports anomaly assessment and operational decision-making, while relevant regulations, cases, and handling experience are dispersed across different data sources, limiting the subsequent use of inspection data and knowledge. To address these issues, this study aims to develop an intelligent process inspection framework that integrates data analytics, intelligent agents, and knowledge engineering to strengthen inspection process integration, anomaly identification, and knowledge utilization.
Following the Deming cycle, this study establishes a closed-loop process encompassing inspection planning, on-site execution, result verification, anomaly reporting, performance evaluation, improvement tracking, and knowledge feedback. The inspection analytics mechanism integrates Isolation Forest, kernel density estimation (KDE), ADaptive WINdowing (ADWIN), and cumulative sum (CUSUM) to identify multivariate outliers, data density distributions and low-density regions, state changes, and cumulative shifts, respectively. The resulting analytical outputs are treated as candidate alerts for further human review. The intelligent agent employs a master coordination agent as the single point of interaction, performing task identification and routing according to user requests and invoking inspection data, analytical tools, and knowledge services as needed. The knowledge architecture integrates a relational database, a vector database, and a knowledge graph. A human review mechanism is also implemented to control the admission of candidate experiential knowledge into the formal searchable knowledge repository.
An intelligent process inspection prototype system was developed and evaluated using simulated inspection data, representative operational scenarios, and a 70-question agent test set. The technical validation results demonstrate that the system can support cross-module inspection data flows and predefined state transitions. The Micro-F1 and Macro-F1 scores for agent task routing were 0.852 and 0.858, respectively, while the knowledge graph path accuracy and exact parameter match rate reached 0.950 and 0.850, respectively. The safety response rules and candidate knowledge state controls were also executed as designed. From an application perspective, the developed system links inspection processes, anomaly analysis results, regulatory knowledge, and historical experience to support anomaly review, cause tracing, response recommendations, and knowledge feedback. Overall, this study verifies the technical feasibility of the proposed framework in terms of inspection process integration, anomaly analysis, agent collaboration, and knowledge utilization. The findings provide a foundation for future deployment in actual manufacturing environments, model parameter calibration, and long-term evaluation of practical effectiveness.

Agrahari, Supriya, and Anil Kumar Singh. 2022. “Concept Drift Detection in Data Stream Mining: A Literature Review.” Journal of King Saud University—Computer and Information Sciences 34 (10): 9523–40. https://doi.org/10.1016/j.jksuci.2021.11.006.
Alves, Joel, Tânia M. Lima, and Pedro D. Gaspar. 2023. “Is Industry 5.0 a Human-Centred Approach? A Systematic Review.” Processes 11 (1): 193. https://doi.org/10.3390/pr11010193.
Automotive Industry Action Group. 2014. CQI-8 Layered Process Audit Guideline. 2nd ed. Automotive Industry Action Group. https://www.aiag.org/training-and-resources/manuals/details/CQI-8.
Bifet, Albert, and Ricard Gavald‘a. 2007. “Learning from Time-Changing Data with Adaptive Windowing.” Proceedings of the 2007 SIAM International Conference on Data Mining, 443–48. https://doi.org/10.1137/1.9781611972771.42.
Bousdekis, Alexandros, Dimitris Apostolou, and Gregoris Mentzas. 2020. “A Human Cyber Physical System Framework for Operator 4.0–Artificial Intelligence Symbiosis.” Manufacturing Letters 25: 10–15. https://doi.org/10.1016/j.mfglet.2020.06.001.
Breque, Maija, Lars De Nul, and Athanasios Petridis. 2021. Industry 5.0: Towards a Sustainable, Human-Centric and Resilient European Industry. European Commission, Directorate-General for Research; Innovation; Publications Office of the European Union. https://doi.org/10.2777/308407.
Buchgeher, Georg, David Gabauer, Jorge Martinez-Gil, and Lisa Ehrlinger. 2021. “Knowledge Graphs in Manufacturing and Production: A Systematic Literature Review.” IEEE Access 9: 55537–54. https://doi.org/10.1109/ACCESS.2021.3070395.
Charles, Rebecca L., Teegan L. Johnson, and Sarah R. Fletcher. 2015. “The Use of Job Aids for Visual Inspection in Manufacturing and Maintenance.” Procedia CIRP 38: 90–93. https://doi.org/10.1016/j.procir.2015.08.056.
Ding, Dong, Fugee Tsung, and Jian Li. 2016. “Directional Control Schemes for Processes with Mixed-Type Data.” International Journal of Production Research 54 (6): 1594–609. https://doi.org/10.1080/00207543.2015.1023402.
Escobar, Carlos A., Megan E. McGovern, and Ruben Morales-Menendez. 2021. “Quality 4.0: A Review of Big Data Challenges in Manufacturing.” Journal of Intelligent Manufacturing 32: 2319–34. https://doi.org/10.1007/s10845-021-01765-4.
Garcia, Cristian I., Marcus A. DiBattista, Tomás A. Letelier, Hunter D. Halloran, and Jaime A. Camelio. 2024. “Framework for LLM Applications in Manufacturing.” Manufacturing Letters 41: 253–63. https://doi.org/10.1016/j.mfglet.2024.09.030.
Garcia-Garcia, Guillermo, Guy Coulthard, Sandeep Jagtap, Mohamed Afy-Shararah, John Patsavellas, and Konstantinos Salonitis. 2021. “Business Process Re-Engineering to Digitalise Quality Control Checks for Reducing Physical Waste and Resource Use in a Food Company.” Sustainability 13 (22): 12341. https://doi.org/10.3390/su132212341.
Hou, Tung-Hsu, Li Lin, and Colin G. Drury. 1993. “An Empirical Study of Hybrid Inspection Systems and Allocation of Inspection Functions.” International Journal of Human Factors in Manufacturing 3 (4): 351–67. https://doi.org/10.1002/hfm.4530030404.
Hütten, Nils, Miguel Alves Gomes, Florian Hölken, Karlo Andricevic, Richard Meyes, and Tobias Meisen. 2024. “Deep Learning for Automated Visual Inspection in Manufacturing and Maintenance: A Survey of Open-Access Papers.” Applied System Innovation 7 (1): 11. https://doi.org/10.3390/asi7010011.
International Organization for Standardization. 2015a. ISO 9000:2015 Quality Management Systems—Fundamentals and Vocabulary. International Organization for Standardization.
International Organization for Standardization. 2015b. ISO 9001:2015 Quality Management Systems—Requirements. International Organization for Standardization.
Kang, Lan, and Susan L. Albin. 2000. “On-Line Monitoring When the Process Yields a Linear Profile.” Journal of Quality Technology 32 (4): 418–26. https://doi.org/10.1080/00224065.2000.11980027.
Kusiak, Andrew. 2018. “Smart Manufacturing.” International Journal of Production Research 56 (1–2): 508–17. https://doi.org/10.1080/00207543.2017.1351644.
Latsou, Christina, Maryam Farsi, and John Ahmet Erkoyuncu. 2023. “Digital Twin-Enabled Automated Anomaly Detection and Bottleneck Identification in Complex Manufacturing Systems Using a Multi-Agent Approach.” Journal of Manufacturing Systems 67: 242–64. https://doi.org/10.1016/j.jmsy.2023.02.008.
Leng, Jiewu, Weinan Sha, Baicun Wang, et al. 2022. “Industry 5.0: Prospect and Retrospect.” Journal of Manufacturing Systems 65: 279–95. https://doi.org/10.1016/j.jmsy.2022.09.017.
Lewis, Patrick, Ethan Perez, Aleksandra Piktus, et al. 2020. “Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks.” Advances in Neural Information Processing Systems 33: 9459–74.
Lim, Jonghan, Leander Pfeiffer, Felix Ocker, Birgit Vogel-Heuser, and Ilya Kovalenko. 2023. “Ontology-Based Feedback to Improve Runtime Control for Multi-Agent Manufacturing Systems.” 2023 IEEE 19th International Conference on Automation Science and Engineering, 1–7. https://doi.org/10.1109/CASE56687.2023.10260621.
Liu, Fei Tony, Kai Ming Ting, and Zhi-Hua Zhou. 2008. “Isolation Forest.” 2008 Eighth IEEE International Conference on Data Mining, 413–22. https://doi.org/10.1109/ICDM.2008.17.
Liu, Hu-Chen, Ran Liu, Xiuzhu Gu, and Miying Yang. 2023. “From Total Quality Management to Quality 4.0: A Systematic Literature Review and Future Research Agenda.” Frontiers of Engineering Management 10 (2): 191–205. https://doi.org/10.1007/s42524-022-0243-z.
Lu, Yuqian, Xun Xu, and Lihui Wang. 2020. “Smart Manufacturing Process and System Automation—A Critical Review of the Standards and Envisioned Scenarios.” Journal of Manufacturing Systems 56: 312–25. https://doi.org/10.1016/j.jmsy.2020.06.010.
McCracken, A. K., S. Chakraborti, and A. Mukherjee. 2013. “Control Charts for Simultaneous Monitoring of Unknown Mean and Variance of Normally Distributed Processes.” Journal of Quality Technology 45 (4): 360–76. https://doi.org/10.1080/00224065.2013.11917944.
Page, E. S. 1954. “Continuous Inspection Schemes.” Biometrika 41 (1–2): 100–115. https://doi.org/10.1093/biomet/41.1-2.100.
Parzen, Emanuel. 1962. “On Estimation of a Probability Density Function and Mode.” The Annals of Mathematical Statistics 33 (3): 1065–76. https://doi.org/10.1214/aoms/1177704472.
Pulikottil, Terrin, Luis A. Estrada-Jimenez, Hamood Ur Rehman, Fan Mo, Sanaz Nikghadam-Hojjati, and José Barata. 2023. “Agent-Based Manufacturing—Review and Expert Evaluation.” The International Journal of Advanced Manufacturing Technology 127 (5–6): 2151–80. https://doi.org/10.1007/s00170-023-11517-8.
Qin, S. Joe. 2012. “Survey on Data-Driven Industrial Process Monitoring and Diagnosis.” Annual Reviews in Control 36 (2): 220–34. https://doi.org/10.1016/j.arcontrol.2012.09.004.
Rosenblatt, Murray. 1956. “Remarks on Some Nonparametric Estimates of a Density Function.” The Annals of Mathematical Statistics 27 (3): 832–37. https://doi.org/10.1214/aoms/1177728190.
Saihi, Afef, Mahmoud Awad, and Mohamed Ben-Daya. 2023. “Quality 4.0: Leveraging Industry 4.0 Technologies to Improve Quality Management Practices—a Systematic Review.” International Journal of Quality & Reliability Management 40 (2): 628–50. https://doi.org/10.1108/IJQRM-09-2021-0305.
Sundaram, Sarvesh, and Abe Zeid. 2023. “Artificial Intelligence-Based Smart Quality Inspection for Manufacturing.” Micromachines 14 (3): 570. https://doi.org/10.3390/mi14030570.
Tao, Fei, Qinglin Qi, Ang Liu, and Andrew Kusiak. 2018. “Data-Driven Smart Manufacturing.” Journal of Manufacturing Systems 48: 157–69. https://doi.org/10.1016/j.jmsy.2018.01.006.
Tercan, Hasan, and Tobias Meisen. 2022. “Machine Learning and Deep Learning Based Predictive Quality in Manufacturing: A Systematic Review.” Journal of Intelligent Manufacturing 33 (7): 1879–905. https://doi.org/10.1007/s10845-022-01963-8.
Wan, Yuwei, Ying Liu, Zheyuan Chen, et al. 2024. “Making Knowledge Graphs Work for Smart Manufacturing: Research Topics, Applications and Prospects.” Journal of Manufacturing Systems 76: 103–32. https://doi.org/10.1016/j.jmsy.2024.07.009.
Wang, Baicun, Fei Tao, Xudong Fang, Chao Liu, Yufei Liu, and Theodor Freiheit. 2021. “Smart Manufacturing and Intelligent Manufacturing: A Comparative Review.” Engineering 7 (6): 738–57. https://doi.org/10.1016/j.eng.2020.07.017.
Wang, Baicun, Pai Zheng, Yue Yin, Albert Shih, and Lihui Wang. 2022. “Toward Human-Centric Smart Manufacturing: A Human-Cyber-Physical Systems (HCPS) Perspective.” Journal of Manufacturing Systems 63: 471–90. https://doi.org/10.1016/j.jmsy.2022.05.005.
Wang, Shu, and Roger J. Jiao. 2024. “Smart in-Process Inspection in Human–Cyber–Physical Manufacturing Systems: A Research Proposal on Human–Automation Symbiosis and Its Prospects.” Machines 12 (12): 873. https://doi.org/10.3390/machines12120873.
Wang, Zhenyu, and Leo Chiang. 2019. “Monitoring Chemical Processes Using Judicious Fusion of Multi-Rate Sensor Data.” Sensors 19 (10): 2240. https://doi.org/10.3390/s19102240.
Zou, Changliang, Chunguang Zhou, Zhaojun Wang, and Fugee Tsung. 2007. “A Self-Starting Control Chart for Linear Profiles.” Journal of Quality Technology 39 (4): 364–75. https://doi.org/10.1080/00224065.2007.11917702.