臺灣博碩士論文加值系統

自動化機器生產為基礎的智慧工廠的概念，已儼然成為下一波製造業革命的 趨勢。然在製鞋產業中，因某些特殊因素與需求，使得機器自動化生產在短期間內， 仍無法取代現有產線上所需的大量人力。一般而言，傳統製鞋業的製程上需要經過 裁剪、削皮、手工、襪里、針車、底倉、成型、組底等許多以人工方式進行加工的 作業子流程，而在每一個子流程當中又涵蓋了許多的操作程序，整體製程作業繁雜。 這些繁瑣的流程即使在具有標準生產步驟管理的產線上，仍然會因作業過程中，基 於人工、原物料、工具等因素，間接導致不良品的產生，造成良率下降、生產成本 增加，從而引致企業獲利下降。
為此，本研究希望透過一系統化過程，來改善人工為主的製鞋產業進行生產品 質。首先，本研究以一家製鞋企業為基礎來進行個案探討:根據該企業之產線來闡 述製鞋作業流程上的各工作階段;再則針對與此個案公司主管及相關負責人員進 行訪談，瞭解個案產線上所遭遇的真實問題與需求，再輔以吾人於製鞋產線上之多 年從業經驗，及多方面的文獻探討為基礎來設計一資訊系統，以快速有效地界定不 良品發生成因。
為了可以實現上述目標，本研究以生產履歷、知識圖譜主要核心架構，以及時 發現問題的根本原因與異常排解的速度為優先原則，並採用逐步將知識本體擴充 完善之策略為導入策略。為了驗證此架構之有效性，本研究透過與個案企業合作開 發一智慧製造資訊平台之雛型系統來加以驗證:將產線上實際的生產資訊導入此 雛形系統當中，並透過視覺化操作介面來加以操作。根據測試可以發現:藉由導入 知識圖譜與資訊化系統將可以幫助鞋廠管理人員可以有效快速地地界定導致不良 品發生之成因，從而協助企業進行生產品質改善之作業目標。

The concept of Smart Factory based on automated machine production gradually became the trend of the next wave of manufacturing revolution. In the footwear industry, however, automated production of machines still could not replace the large amount of manpower required on existing production lines in a short time due to some special factors and needs. In general, the traditional footwear industry process needs to go through cutting, skiving, handwork, bootie, stitching, soling, assembling, and stock fitting and other operation sub-processes that are processed manually, and each sub-process covers a lot of operating procedures. The overall process operation is complicated. These cumbersome processes, even on a production line with standard production step management, will still indirectly lead to production of defective products due to factors such as labor, raw materials, tools, etc., during the operation, resulting in a decrease in yield and an increase in production costs. This leads to a decline in profits for enterprises.
Hence, the purpose of this research was to improve the production quality of the manually oriented footwear industry through a systematic process. First of all, this research conducted a case study on a shoe-making enterprise: explain the various working stages of the shoe-making operation process according to the production line of the enterprise. Then I proceeded interviews with the supervisor and relevant responsible personnel in this case to understand the real problems and needs encountered in the production line. Supplemented with our years of working experience in the footwear production line and based on various literature reviews, an information system was designed to quickly and effectively identify the causes of defective products.
In order to achieve the above objectives, this research took traceability and knowledge graph as the main core framework. Timely discovery of the root cause of the problem and the speed of troubleshooting were the priority principle. The introduction strategy of gradually expanding and perfecting the ontology was adopted. In this research, the effectiveness of this framework was verified by cooperating with the case enterprise to develop a prototype system of a smart manufacturing information platform: the actual production information on the production line was imported into this prototype system, and it was operated through a visualized operation interface. According to results of the test, the introduction of knowledge graph and information system was proved to help managerial staff in footwear factory to effectively and quickly identify the causes of defective products, thereby assisting the enterprise in the operation goal of improving production quality.

摘 要 i
ABSTRACT ii
誌 謝 iv
目 錄 v
圖 目 錄 vi
表 目 錄 vii
一、 緒論 1
1.1. 研究背景 1
1.2. 研究問題 1
1.3. 研究目的 2
1.4. 研究限制 3
二、 文獻探討 4
2.1. 鞋廠製程問題 4
2.2. 智慧製造 5
2.3. 知識圖譜 7
2.3.1. 知識圖譜 7
2.3.2. 知識圖譜建構技術流程 8
三、 系統架構 & 研究方法 10
3.1. 實地訪談 10
3.1.1. 實際QC問題 10
3.1.2. 製程紀錄繁雜錯亂，造成控管困難 11
3.1.3. 無數控設備，不易導入全自動化 11
3.1.4. 高單價高品質產品需要仰賴人工製作 12
3.1.5. 流程間相連性大 12
3.2. 系統架構 13
四、 研究結果 19
4.1. RFID GS1 20
4.2. 既有資料到生產履歷 25
4.3. 生產履歷到知識圖譜 32
4.3.1. Neo4j 32
4.4. 知識圖譜應用 35
4.4.1. 管理與視覺化介面 35
4.4.2. 基本資料 36
4.4.3. 標準化製程圖譜/異常製程圖譜 36
4.4.4. 製造流程搜尋 38
4.4.5. 自定義搜尋 38
五、 結論 40
5.1. 研究結論 40
5.2. 未來展望 40

[1] 陳民議, "導入物聯網技術的智能生產線建置模式," 2015.
[2] G.Oliver, P.Gil, and J. F. G. F.Torres, "Towards footwear manufacturing 4.0: shoe sole robotic grasping in assembling operations," The International Journal of Advanced Manufacturing Technology volume vol. 114, pp. 811–827, 2021, doi: https://doi.org/10.1007/s00170-021-06697-0.
[3] E. M. M.Marconi , A.Papetti &M.Germani, "Traceability as a means to investigate supply chain sustainability: the real case of a leather shoe supply chain," International Journal of Production Research, vol. 55, no. 22, pp. 6638- 6652, 2017, doi: https://doi.org/10.1080/00207543.2017.1332437.
[4] 陳建鈞, "智慧造鞋卡關、結束兩間機器人工廠，Adidas 升級工業 4.0 遇到什麼阻礙?," 數位時代, 2019. [Online]. Available:https://www.bnext.com.tw/article/55494/adidas-speedfactories.
[5] Y. C. T.Wang, M.Qiao & H.Snoussi "A fast and robust convolutional neural network-based defect detection model in product quality control," The International Journal of Advanced Manufacturing Technology volume vol. 94, pp. 3465–3471, 2018, doi: https://doi.org/10.1007/s00170-017-0882-0.
[6] B. V. S. M.Varsha Magar, "Application of 7 Quality Control (7 QC) Tools for Continuous Improvement of Manufacturing Processes " International Journal of Engineering Research and General Science, vol. 2, no. 4, pp. 364-371, 2014.
[7] N. M. S. H.Sharma, "Implementation of Quality Control Tools and Techniques in Manufacturing Industry for Process Improvement," 2017.
[8] X. X. R.Y. Zhong, E. Klotz, S.T. Newman, "Intelligent Manufacturing in the Context of Industry 4.0: A Review," Engineering, vol. 3, no. 5, pp. 616-630, 2017.
[9] B.-c. H. B-h.Li, W-t. Yu, X-b. Lu & C-w. Yang "Applications of artificial intelligence in intelligent manufacturing: a review," Frontiers of Information Technology & Electronic Engineering, vol. 18, pp. 86-96, 2017.
[10] D. W. T.Wuest, C.Irgens, K.-D Thoben, "Machine learning in manufacturing: advantages, challenges, and applications," Production & Manufacturing Research, vol. 4, no. 1, pp. 23-45, 2016,doi:https://doi.org/10.1080/21693277.2016.1192517.
[11] G. Z. C.Xu, "Intelligent manufacturing Lie Group Machine Learning: real-time and efficient inspection system based on fog computing," Journal of Intelligent Manufacturing, vol. 32, pp. 237–249, 2021.
[12] P. Z. J.M. Rožanec, K.Kenda, I.Novalija, B.Fortuna, D.Mladenić, "XAI-KG:Knowledge Graph to Support XAI and Decision-Making in Manufacturing," CAiSE 2021: Advanced Information Systems Engineering Workshops, pp. 167- 172, 2021.
[13] 漆桂林 , 王昊奮 , 陳華鈞, "人類智慧的神殿:AI 知識圖譜實作," 2020.
[14] Z. M. Q. Wang, B. Wang, and L. Guo, "Knowledge graph embedding: A survey of approaches and applications," IEEE Transactions on Knowledge and Data Engineering, vol. 29, 12, pp. 2724-2743, 2017, doi: 10.1109/TKDE.2017.2754499.
[15] 于娟, 黄恒琪, 廖晓, 席运江, "知识图谱研究综述," 计算机系统应用, vol. 28, no. 6, pp. 1-12, 2019.
[16] 赵倩, "数控设备故障知识图谱的构建与应用," 2020.
[17] J. Y. a. J. W. H. Yan, "KnowIME: A System to Construct a Knowledge Graph for Intelligent Manufacturing Equipment," IEEE Access, vol. 8, pp. 41805-41813, 2020.
[18] 郝克刚, 刘海, "軟件缺陷原因分析方法," 計算機科學, vol. 36, no. 1, 2009.
[19] P. N. H.Wang, J.Li, S.Kopru, G.Zhang, S.Katariya, and S.Ben-Romdhane, "GRANO: Interactive graph-based root cause analysis for cloud-native distributed data platform," Proceedings of the VLDB Endowment, vol. 12, no. 12, pp. 1942–1945, 2019, doi: https://doi.org/10.14778/3352063.3352105.
[20] 李杨, 刘峤, 段宏,刘瑶,秦志光, "知识图谱构建技术综述," 计算机研究与发展, vol. 53, no. 3, pp. 582-600, 2016, doi: 10.7544/issn1000-1239.2016.20148228.
[21] "GS1 General Specifications," 2021[Online].Available:https://www.gs1us.org/DesktopModules/Bring2mind/DMX/Download.aspx?Command=Core_Download&EntryId=171&language=en-US&PortalId=0&TabId=134.
[22] 呂惠娟, "EPC 標籤資料標準懶人包." [Online]. Available:https://www.gs1tw.org/twct/web/downloads/A-2.pdf.
[23] "EPC Tag Data Standard," 2019. [Online]. Available:https://www.gs1.org/sites/default/files/docs/epc/GS1_EPC_TDS_i1_13.pdf.
[24] "全球交易品項識別碼 GTIN."[Online].Available:https://www.gs1tw.org/twct/web/trade.jsp?MINDEN=01.