盧宣文

博士候選人，於成功大學取得碩士學位，研究領域包含數據分析、故障預測與健康管理 (PHM) 與遷移學習

國立成功大學, 博士候選人, 製造資訊與系統研究所

Institute of Manufacturing Information and Systems, National Cheng Kung University, Taiwan

國立成功大學, 工學院工程管理碩士在職專班

Engineering Management Graduate Program, National Cheng Kung University, Taiwan

生產力最佳化實驗室(PO-Lab)

Productivity Optimization Laboratory

指導教授：李家岩教授

Dr. Chia-Yen Lee

國立中興大學, 化學工程學系

Department of Chemical Engineering, National Chung Hsing University, Taiwan

專題指導教授：竇維平教授

Dr. Wei-Ping Dow

Skills

Machine Learning

Data Mining

Reinforcement Learning

Transfer Learning

Programming Language

Python

SQL

JAVA

Android App Development

Android Studio

Expertise

故障預測與健康管理 (Prognostic and Health Management, PHM)

預測保養 (Predictive Maintenance, PdM)

頻率訊號的轉換與特徵萃取

利用機器學習方法建構分類或預測模型

使用集成學習技巧改善模型

遷移學習 (Transfer Learning)

Work Experience

西柏容有限公司, 工程師, Aug 2020 ~ Now

Research Assistant, Seberon

群創光電, 高級工程師, Nov 2013 ~ Jun 2018

Senior Engineer, Department of Water Treatment, INNOLUX

宏全國際股份有限公司, 品保員, Jul 2011 ~ Jul 2012

Quality Control Personnel, Department of Quality Control, Taiwan Hon Chuan Group

Education

國立成功大學, 博士候選人, 製造資訊與系統研究所, 2020 ~ Now

Institute of Manufacturing Information and Systems, National Cheng Kung University, Taiwan

國立成功大學, 碩士學位, 工學院工程管理碩士在職專班, 2016 ~ 2020

Engineering Management Graduate Program, National Cheng Kung University, Taiwan

國立中興大學, 學士學位, 化學工程, 2007 ~ 2012

Department of Chemical Engineering, National Chung Hsing University, Taiwan

Publications

Lu, Hsuan-Wen (盧宣文), and Chia-Yen Lee, 2019. Kernel-Density Dynamic Ensemble Technique for Predictive Maintenance (發展核密度動態集成技術於預測保養). 2019 Chinese Institute of Industrial Engineers (CIIE) Conference & Annual Meeting, Nov. 23, 2019, National Taipei University of Technology, Taipei, Taiwan.
Lu, Hsuan-Wen, and Lee, Chia-Yen, 2022. Kernel-based dynamic ensemble technique for remaining useful life prediction. IEEE Robotics and Automation Letters, 7 (2), 1142-1149.
李家岩、楊舒惠、歐子毓、盧宣文, 2022。數據科學分析的土壤：數據品質評估，東華AI通訊報，第三期

Thesis

Kernel-Density Dynamic Ensemble Technique for Predictive Maintenance

Abstract

In recent years, prognostic and health management (PHM) has been widely used in manufacturing. One of the important issues is predictive maintenance (PdM). The purpose of PdM is to determine whether equipment or parts are in health. In the past studies, the exponential models were often used to predict the remaining useful life (RUL). However, due to the limitation of the calculation method of the index model, when the aging characteristics suddenly rise, the model may fail to respond in time and cause the prediction error. The time series model is used to construct the prediction mechanism, but when some observations are missing, there will also be prediction errors. This study proposes a dynamic adjustment of the weight mechanism and model credit indicators. Based on the data analysis framework, important statistical characteristics of the data are extracted, and machine learning is used. Different algorithms construct prediction models, give different weights to different prediction model capabilities, and dynamically adjust the weights as the time axis moves to make the prediction results more robust. At the same time, the Inference Confidence Index(ICI) is established. The judgment model predicts the similarity, and with the decision judgment module, the prediction maintenance system is improved.

Certificate

2018 Intelligent Manufacturing and Industry Practice Workshop (智慧製造與產業實務研習會)
2018 Intelligent Production Scheduling System Workshop (智慧生產排程系統研習會)
2018 中央研究院「尖端科技研習營」─ 智慧製造研習營
2019 Chinese Institute of Industrial Engineers Annual Conference(中國工業工程學會年會暨學術研討會)
2020 Intelligent Manufacturing and Industry Practice Workshop (智慧製造與產業實務研習會)
Honorable Mention of Master Thesis Award at Information System Session in Chinese Institute of Industrial Engineers (CIIE) (July 2020)

Project on Github

Job Shop Problem by using Mathematical Programming in Python

Courses in NCKU

Operation Research Applications and Implementation (作業研究應用與實作)
Intelligent Manufacturing Systems (智慧型製造系統)
Industry 4.1: Intelligent Manufacturing with Zero Defects (工業4.1: 零缺陷的智慧製造)
Android App Development (Android 應用程式開發)
Date-base system (資料庫系統)
Optimization Theory and Statistical Data Science (最佳化理論與統計資料科學)
Operations Research (作業研究)
Data Mining (資料探勘)
Statistical Methods (統計方法)
Business Data Communication (企業資料通訊)
Product Life Cycle Management (產品生命週期管理)
企業參謀-思考的技術

Teaching Assistant in NCKU

Data Mining (資料探勘)
Operation Research Applications and Implementation (作業研究應用與實作)

Autobiography

我的名字是盧宣文，於成功大學工學院工程管理碩士在職專班取得碩士學位，更於同年進入成功大學製造資訊與系統研究所研讀博士班，目前已取得博士候選人資格。

工作經歷

大學畢業後進入宏全國際擔任品保員一職，進行產品檢驗及可靠度測試，檢查、抽樣及估量未加工之原料或加工過之部件與產品，以避免與規格不符，造成客訴， 2013年進入群創光電擔任廠務工程師，隸屬水務課，服務五年時間，平時工作職掌為維持水處理三大系統：純水、廢水以及回收水系統穩定運轉，擁有設備保養及簡易維修能力，並與供應商及製程端協調，服務的廠區位於路竹，是台灣TFT-LCD面板生產世代最大的8.5與8.6代廠，擁有水處理的先進技術，對於先進技術原理及操作深入了解，服務期間協助提出系統運轉改善案，增加系統穩定度及降低操作成本，並與同仁合作獲得科學園區節水競賽優勝。

博士學位修習期間於西柏容公司開發資料應用工具與相關演算法，與MES、ERP系統結合，協助改善客戶生產流程與數位轉型。

專業能力

大學畢業於國立中興大學化學工程學系，專題研究題目為電鍍配方中的協同效應，指導老師為竇維平教授，並於專題研究期間，前往美國拉斯維加斯參加218th ECS Meeting，展示專題研究成果；工作一段時間後，有感於自己的不足，為了使自己具備跨領域知識，故於成功大學修習碩士學位，進入生產力最佳化實驗室，學習數據科學、資料探勘與機器學習等理論方法，並參加多次智慧製造相關之學術研討會。

研究所就讀期間修習了製造資訊與數據科學相關的課程，包含智慧型製造系統、資料庫系統、企業資料通訊、最佳化理論與統計資料科學、資料探勘等等，論文研究主要與機器學習方法相關，提出改善預測效果的演算法，另外也因為實驗室的培養，除了數據科學的方法與實作外，更學習了生產排程的相關理論，藉由討論與產學合作的機會，令我具備跨領域的知識。

進入博士班後，擔任產學計畫專案管理人與聯絡窗口，協助專案進行，研究領域則是專注於故障預測與預防保養(PHM)的深化，同時加入遷移學習，研究如何突破製造現場因個別機台特性與差異造成之品質問題。

未來

從大學、工作到研究所，深知每一個專案的完成都需要團隊合作與分工，在重新進入職場後，希望能夠發揮所學，配合團隊，並不斷更新與精進自我。

My name is Hsuan-Wen Lu, graduated from Engineering Management Graduate Program, National Cheng Kung University. In the same year, I entered the PhD program of the Institute of Manufacturing Information and Systems, National Cheng Kung University, and is currently qualified as a PhD candidate.

Work experience

After getting my bachelor degree, the I joined Taiwan Hon Chuan Group as a quality control personnel at the first year, sampling and estimating the raw materials and products to ensure the spec is in control. I joined Innolux as a water treatment engineer in 2013. I had served for five years. The daily work is maintaining the three major systems of water treatment: pure water, waste water, and recycled water systems. Coordinated with suppliers and the process side. During service, I proposed multiple improvement cases to increase system stability and reduce operating costs, also cooperated with colleagues to win the science park water-saving competition.

Professional competence

I graduated from the Department of Chemical Engineering, National Chung Hsing University, the topic of research is the synergies in electroplating formulations, the advisor is Dr. Wei-Ping Dow. I participate in the 218th ECS Meeting which in Las Vegas to present the research. After working for a period of time, I studied a master ’s degree at National Cheng Kung University, entered the productivity optimization laboratory, and studied the data science and machine learning methods. Furthermore, I took courses related to manufacturing information and data science, including intelligent manufacturing systems, database systems, enterprise data communication, optimization theory and statistical data science, data mining, etc. The thesis research is mainly related to machine learning methods Relevant, proposed an algorithm to improve the prediction effect. In addition to the cultivation of the laboratory, in addition to the methods and implementation of data science, I also learned the relevant theory of production scheduling.

I served as the project manager of the industry-university plan to assist the project. The research field focused on the prognostic and health management (PHM) and the transfer learning try to break through the quality problems caused by machine characteristics and differences.

Future

I understand that the completion of each project requires teamwork and division of labor. After re-entering the workplace, I wish be able to use what I have learned, cooperate with the team, and constantly update and improve myself.