知識結構在故障診斷維修之專家系統的應用－以台灣機車產業的技術維修員為例

簡易檢索 / 詳目顯示

回結果列表

研究生：	周裕發 Yu-Fa Chou
論文名稱：	知識結構在故障診斷維修之專家系統的應用－以台灣機車產業的技術維修員為例 Apply Knowledge Structure to the Fault Detection and Maintenance of The Expert System: Using Taiwan Motorcycle’s Maintenance workers as An Example
指導教授：	黃雪玲 Sheue-Ling Hwang 蘇國瑋 Kuo-Wei Su
口試委員:
學位類別：	碩士 Master
系所名稱：	工學院 - 工業工程與工程管理學系 Department of Industrial Engineering and Engineering Management
論文出版年：	2005
畢業學年度：	93
語文別：	中文
論文頁數：	58
中文關鍵詞：	知識結構、專家系統、人機介面、教育訓練
外文關鍵詞：	Knowledge Structure, Fault Diagnosis Assistance System (FDAS), Human-Computer Interface, Educational Training
相關次數：	點閱：2 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

隨著經濟的發展及人口流動的增加，國人對於交通工具需求量大増，道路面積的成長數目卻遠不及車輛成長的數目，使得價位不高，易於穿梭的機車數目不斷地增加。目前台灣機車的總數共約12,366,864輛，而密度已是全世界第一。台灣機車的維修業者也因此蓬勃發展，但目前機車維修業者的技術參差不齊，且其技能大都全靠師徒的相傳及經驗的累積，人員的高流動率更造成技術的無法傳承，消費者時常淪為實習的對象；爾後因加入WTO，重型機車將限量開放進口，電動機車也陸續加入市場，屆時機車專業維修的困難度及複雜性勢必提高。
本研究提出一套應用於機車維修工作上的故障診斷輔助系統，從專家的知識結構面，擷取出專家維修之豐富經驗與知識，成為知識法則庫，同時建構出具親和力的人機使用介面。本研究為以知識網路組織工具建立出專家群的知識結構圖，進而發展故障診斷輔助型的專家系統；並利用PRX、GTD和PFC三種相似性指數，以專家的指數為基準值，評估新手在使用本系統與新手在使用傳統手冊的知識網路結構間的差異，最後以可用性觀點綜合評估其系統運用的必要性。
由知識網路組織工具裡的相似性指數及可用性評估驗證了本研究所發展出的故障診斷輔助系統之可行，使用系統的新手確實比使用手冊的新手較接近專家解決問題的模式。本系統除了可以提升維修人員故障診斷之正確性，並可彌補台灣機車產業技術維修人才短缺、經驗不足及流失的危機，同時亦可作為維修人員教育訓練之工具而有效的縮短訓練時間，提供高品質的維修服務。

Today, Taiwan motorcycle is the third major exporting country and has better competition than other countries in a kind of under 150c.c motorcycle. According to the statistics of the ministry of transportation and communications, there are 12,366,864 motorcycles in Taiwan. Though Taiwan motorcycle’s density is the highest in the world, the motorcycle maintenance knowledge is not at the same level. Accordingly, the government has advanced motorcycle technique test since 2001 to promote motorcycle maintenance knowledge and service level for consumers.
This study proposes a fault diagnosis assistance system（FDAS） that applied in the motorcycle maintenance. An expert’s knowledge structure was measured, to construct a knowledge base that derived from expert’s wonderful experience and knowledge and then to build a friendly human-computer interface. This study also examined the knowledge structures of a novice group and an expert group by using knowledge network organizing tool. The similarities between novice’s and expert’s networks are assessed using the indices PRX, GTD, and PFC.
The results indicate that the knowledge structure of the using FDAS group is better than that of the using manual group. Additionally, the usability test confirms the well-used system interface. It also proved the evidence that the novice’s knowledge structure is changed better after using FDAS. It can, of course, serve as an educational training tool for maintainers and shorten learning time effectively and also make up for the shortage of Taiwan motorcycle’s maintainers.

CONTENTS
摘 要    I
ABSTRACT    II
CONTENTS    III
FIGURE CAPTIONS    V
TABLE CAPTIONS    VI
CHAPTER 1  INTRODUCTION    1
1.1 MOTIVATION AND BACKGROUND    1
1.2 OBJECTIVES    2
1.3 RESEARCH FRAMEWORK    3
CHAPTER 2  LITERATURES REVIEW    6
2.1 FAULT DIAGNOSIS ASSISTANCE SYSTEM (FDAS)    6
2.1.1 Case-Based reasoning (CBR)    8
2.1.2 Neural network    9
2.1.3 Expert system (ES)    10
2.2 THE CONCEPT OF KNOWLEDGE STRUCTURE    11
2.2.1 The different knowledge structure between the novice and expert    11
2.2.2 Knowledge acquisition and representation    13
2.2.3 The basic components of expert system    15
2.3 THE RELEVANT ANALYTIC DESCRIPTION    17
2.3.1 Failure modes and effects analysis (FMEA)    17
2.3.2 Pathfinder network analysis    20
2.3.3 The Knowledge network organizing tool (KNOT) Software introduction    23
2.3.4 Visual rule studio (VRS) introduction    23
2.3.5 Usability evaluation    24
CHAPTER 3  THE PROCESS OF SYSTEM DEVELOPMENT    26
3.1 TO FIND OUT KEY COMPONENTS    26
3.2 KNOWLEDGE STRUCTURES OF THE EXPERT    29
3.3 THE DEVELOPMENT OF FDAS    33
CHAPTER 4  THE INTERPRETATION AND PROOF OF FDAS    37
4.1 THE OPERATIONAL PROCEDURES AND FUNCTIONAL INTERPRETATION    37
4.2 EXPERIMENT    40
4.2.1 Subjects    40
4.2.2 Experimental material    40
4.2.3 Procedure    41
CHAPTER 5  RESULTS AND DISCUSSIONS    43
5.1 THE COMPARISON OF THE KNOWLEDGE STRUCTURE BETWEEN USING SYSTEM GROUP AND USING MANUAL GROUP    43
5.2 USABILITY TESTING    44
5.5 DISCUSSIONS    45
CHAPTER 6  CONCLUSIONS    47
6.1 CONTRIBUTIONS    47
6.2 LIMITATIONS AND FUTURE RESEARCH    48
REFERENCES    49
APPENDICES    52
APPENDIX ONE: FAULT SYMPTOM QUESTION TEST    52
APPENDIX TWO: THE HANDBOOK OF MOTORCYCLE PARTS    53
APPENDIX THREE: USABILITY TESTING RESULTS    58

FIGURE CAPTIONS
Figure 1-1. Research Framework    5
Figure 2-1. The differences between expert and novice in the model of problem solving (Revised from Ming Feng Gu, 1998.)    13
Figure 2-2. The semantic networks of a motorcycle engine system    15
Figure 2-3.  The constructional process of expert system (Revised from Nii,1993)    16
Figure 2-4. A set of hypothetical proximity data and its resulting nonhierarchical pathfinder network. (Goldsmith et al., 1991）    21
Figure 2-5. The PFC and GTD indices of network1, network2 and network3 （Goldsmith et al., 1991）    22
Figure 3-1. The order cost of different components （2002/6/1~2003/2/1）    27
Figure 3-2. The structure of a motorcycle system    27
Figure 3-3. Sample of examination questions    30
Figure 3-4. The operational procedure of the KNOT    31
Figure 3-5. The knowledge structure diagram in three experts    32
Figure 3-6. The knowledge structure figure of the experts    34
Figure 3-7. The declaring of classes and attributes    35
Figure 3-8. The edition of rule    36
Figure 3-9. The design of user interface    36
Figure 4-1. The major screen    37
Figure 4-2. The flow chart of the fault diagnosis system    38
Figure 4-3. The appearance of symptoms    38
Figure 4-4.  Symptoms occurred in the most possible system    39
Figure 4-5.  The possible causes of the system    39
Figure 4-6.  The functional introduction and the maintenance suggestion    40
Figure 4-7.  The flow chart of experiment    42
Figure 5-1.  The line plot of the usability testing    45







TABLE CAPTIONS
Table 1-1: A service factory’s the average order cost of components in pre month（2002/6/1~2003/2/1）    4
Table 2-1: Different approaches on fault diagnosis    7
Table 2-2:  The comparison of case-based reasoning, neural network and expert system    10
Table 2-3: A frame of motorcycle’s category    15
Table 3-1: FMEA report of engine device    28
Table 3-2. Some PRNs of the engine and electrical mechanism    29
Table 5-1. The results of the PRX, GTD and PFC- indexes    43
Table 5-2. The results of the PRX, GTD and PFC- indexes    43
Table 5-3. Usability Testing    44

                                

方俊明著（民82）。認知心理學與人格教育。台北巿：水牛，頁282-330。
古明峰（民87）。數學應用題的解題認知歷程之探討。教育研究資訊，6卷，3期，頁63-77。
交通部統計處（民91），http://www.motc.gov.tw./service/。
江淑卿 (民90)。概念構圖與圖示對兒童自然科學的知識結構、理解能力與學習反應之影響。科學教育學刊，第九卷第一期， 9(1)，頁35-54。
葉倩亨（民90），路徑搜尋網路分析應用於大一心理學學習效果評量之研究，教育與心理學研究，24期（下冊），頁421-446。
葉怡成（民86），應用類神經網路，儒林圖書公司，頁122-145。
曾憲雄、黃國禎、江孟峰、蔡昌均、林耀聰 (民92)，專家系統導論/工具/應用，松崗。
薛理桂。(民80)「專家系統在圖書館的應用」，國立中央圖書館台灣分館館訊 5期：頁10-12。
蘇木春、張孝德（民88），機械學習：類神經網路、模糊系統以及基因演算法則，全華科技圖書股份有限公司。
Aamodt and EPlaza, (1994), ”Case-Based Reasoning：Foundational issues, Methodological Variations, and System Approaches”, AI Communication. ISO Press, Vol.7, pp.139-39.
ARUN SHARMA,MICHAEL LEVY & AJITH KUMAR, (2000), Knowledge Structures and Retail SalesPerformance: An Empirical Examination, Journal of Retailing Vol. 76, No. 1, pp.53-69.
Ballard Dana H. (1997), “An Introduction to Natural Computation”, The MIT Press, ISBN 0-262-02420-9.
Bertram C. Brooks (1980), “The Foundations of Information Science. Part I. Philosophical Aspects”, Journal of Information Science 2, pp.131.
Bromme R. and Tillema H., (1995), “Fusing experience and theory: The structure of professional knowledge”. Learning and Instruction, Vol.5, pp.261-267.
Chow P.K.O. & Yeung D. S., (1995), “A Multidimensional Knowledge Structure”, Expert Systems With Applications, Vol. 9, No. 2, pp.177-187.
Frederick Hayes-Roth (1992), “Knowledge Systems: An Introduction,” Library Hi Tech 10, no. 1-2, pp. 25.
Haykin S., (1999), “Neural Networks: A Comprehensive Foundation”, Prentice Hall, pp.331-340.
Ikujiro N., and Hirotaka T., (1995), “The Knowledge-Creating Company”. Oxford University Press, pp. 61.
Lin Y. Q., (2002), “Apply Expert System to the Fault Detection of HVAC”, Master’s thesis, Taiwan, ROC: National Taiwan University of Science and Technology Department of Electrical Engineering.
Nussbaum M. , Rosas R. , Peirano I. , Cardenas F., (2001), “Development of intelligent tutoring systems using knowledge structures”, Computers & Education 36, pp.15-32.
Nii, H. P. (1993), http://www.wtec.org/loyola/kb/c3_s2.htm.
Riesbeck C.K , Schank R. (1989), “Inside Case-based Reasoning”, Lawrence Erlbbaum Associates, Northvale, pp.56-60.
Scarbrough H. (1998), “The management of knowledge workers”, in W. Currie and R. Galliers, Rethinking Information Systems, Oxford University Press.
Schvaneveldt, R.W. (Eds.) (1990). Pathfinder associative Networks : Studies in knowledge organization. Norwood, NJ: Ablex.
Schvaneveldt, R. W. (1994). Knowledge Network Organizing Tool ( PCKNOT version: 4.2 ). Las Cruces, NM: Interlink, Inc.
She, Y. M., (1996). The approach of user’s mental model and cognitive type in process control system, Master’s thesis, Taiwan, ROC: National Tsing-Hua University.
Shortliffe, E. H. & Buchanan, B. G. (1989). Advances in Expert Systems (A White Paper and Commentary), Working Paper.
Su, k. W., Hwang, S. L., Liu, T. H. (2000) “Reliability-Centered Maintenance Management: A Case Study on a Public Bus System,” International Journal of Industrial Engineering – Theory, Application, and Practice, volume 7, issue 3 (September 2000), pp.232-241.
Varma A. (1990), “ICARUS: Design and Deployment of a Case-Based Reasoning System for Locomotive Diagnostics”, Notes in Artificial Intelligence 1650, pages 581-596.
Visual Rule Studio Developer’s Guide (2000), Rule Machines Corporation.
XpertRule Knowledge Builder’s Guide (1988), Attar Soft Ltd..
Yang C.L., (2002), “Diagnosis of Engine Gas-path with Multiple Faults”, Master’s thesis, Taiwan, ROC: National Cheng Kung University of Department of Aeronautics and Astronautics.
Yen, C.Y., (2002), “Research of Remote Diagnosis and Maintenance of a Semiconductor Cluster Tool”, Doctoral dissertation, Taiwan, ROC: National Taiwan University of Department of Mechanical Engineering.
Zack, M.H. (1999). Developing a knowledge strategy. “California Management Review”, 41 (3), pp.344-353.

全文公開日期本全文未授權公開 (校內網路)
全文公開日期本全文未授權公開 (校外網路)

簡易檢索 / 詳目顯示

相關論文