poster session 3 p.124 a new ga-based rbf neural p.l25 ... · poster session 3 p.124 a new ga-based...
TRANSCRIPT
Poster Session 3 p .124 A New GA-based RBF Neural p.l25 Constructing Model of Network with Optimal Organizational Internal
12/11/2009 10:30 . 11 :00 Selection Clustering Knowledge Integration Based Room: Longchamps I & II Algorithm for SINS Fault on Cultural Algorithm
Diagnosis Sihua chen', Changqi tao', Wei he• Zhide Liu', Jiabin Chen', 'JiangXi University of Finance And
p.123 Kinematic Analysis and Yongqiang Han', Chunlei Song' Economics, China Post-Processing Algorithm 18eijing Institute ofTeclmology, China Research for S-A xis CNC p.l25 The Empirical Study on the Machine Tools with a p.124 A Smart Model for Urban Creation of Corporate Universal Head Ticketing Based on RFID Intellectual Capital from the Chun Xie', Weimin Zhang', Applications Perspective of Social Capital XinyuanHe' Maria Grazia Gnoni', Alessandra Jun-yiRen' 1Tongji University, Chirzn Rollo', Piergiuseppe Tundo' 1Yantai University, Chi>ta 'Shenyang Machine Tool !Group) CO., 1 Dept. of Engmeering for LTD, China llmovation/University of salento, Italy p .125 International Comparison on
p.124 Applying Fuzzy Ruled Based the Coordination Degree p .l23 The Simulation of Cutting Between Economic Force of Free-form Surface to Flexible Routing Problem
Development and BERD Machining with Ball-end in a Flexible Manufacturing Investment Milling Cutter System Haifeng Wang1, Yafei Luo' l.ei Shi1
, En Fu Liu', Yi Zhang', lraj Mahdavi', Amirhosein Fekri 18eijing University of Technology, Peng Chen1, Zongbin Li' Moghaddam Azar', Morteza China 'Xi' an ]iaotong University, China Bagherpour' 2Hebei University of Science and 1Mazandaran University of Science p .125 A New Classification Method Technology, China and Technology, Iran
on the Basis of a ' Islamic Azad University of Shiraz, Patent-Science Relationship p.l23 Cold Nosing Process Iran Yan-Ru Li', Tzu-Ying Li' Modeling and Simulation for
p.l24 A Multi-Agent and Extremal 1 University of Alefheia, Taiwan Manufacturing of Aluminium
Optimization System for :' Conical Milk Can "Steelmaking-Continuous p.l25 Research on Parameter Jinn-Jong Sheu1, Hsien-Hsiu Su•
Transferring Complexity of 'National Kaohsiung University of Casting-Hot Strip Mill" Assembly Variant Design Applied Sciences, Taiwan Integrated Scheduling Xinsheng Xu1, Xin Cheng', Rigeng]i', Yong-Zai Lu• Zhengxiang Li' p.123 A Manufacturing 1Zhejiang University, China 'China Jiliang University, China Performance Evaluation
p. 124 Forecasting Stock Price Using 'Zhejiang University, China Model for Notebook Computer Manufacturers Nonlinear Independent p.l25 Customer-Oriented Library Rong-Hwa Huang', Chang-Lin Component Analysis and
Services for Chinese Higher Yang', Hui-lung Shih1 Support Vector Regression Education •Fu fen Catholic U11iversity, Taiwan Chi-Jie Lu', Jui-Yu Wu2, Jiahui Jiang', Yongbing Zhang2 Cheng-Ruei Fan>, Chih-Chou Chiu' 1 Library of Southwestem University of p.l23 Noise Identification and Fault 1Ching Yun University, Taiwan Finance and Economics, China
Diagnosis for the New ' Lunghwa University of Science a11d 'University ofTsukuba, Japan Technology, Taiwan Products of ~e Automobile
'National Taipei University of p.l26 An Ubiquitous Infrastructure Gearbox
Technology, Taiwan Wenli shang'. Yigong yan2, Haibo Applications to Support shi' p.l24 Estimating the Inquiring Museum's Service ' Shenya>tg Institute of
Time Interval for the Patent Chen-Wo Kuo', Johannes K. Automation,Chi11ese ACIJdemy of Analysis by the Technology Chiang' Science, China
'National Chengchi University, Taiwan 2Changchun University of Technology, Obsolescence Cycle China Hsiao-Chung Wu1, Hung-Yi Chen'
p.l26 The Valuation of Money-Back 1Chaoyang University of Technology, Guarantees in Retailing p.l23 Analysis of the Forming Taiwan Markets: A Real Option Defects of the Trapezoidal
p .l24 Demonstration Study on Approach Inner-gear Spinning Lieh-Ming Luo1, Hui-Tzu Lee2, Qin-xiang Xia1, Ling-yan Sun•, Small and Medium High-tech Yu-Ping Hsieh' Xiu-quan Cheng', Bang-yan Ye1 Enterprises Growth: The Case 1 Fu Jen Catholic University, Tniwa11 'South Chi11a U>riversity of Technology, ofDalian 'National Chung Hsing University, China Lin Li', Pengfei Zhou2, Yan Wang' Taiwan ' Guangzhou Civil Aviation College, 'Dalian ]iaofong University, China 'Chihlee Institute of Technology, China 2Dalian University of Technology, Taiwan China
p.l23 Research on Tool Path p.l26 Research on the Planning for Five-Axis p.125 The Antecedents and
Service-oriented Machining Consequences of Customer Manufacturing Model Peinan Li', Ruifeng Guo•, Pin Knowledge Development in Na Gao1, Songzheng Zhao', Xiaodi Wang', Yan Huang' New Product Development Zhang' 1Shenyang Institute of Computiltg Yen-Tsung Huang', 1-Chun Chen' 1 Northwestern Polytechnical Technology, Chinese ACIJdemy of 'Tunghai University, Taiwan University, China Sciences, China
p.l24 Stock Index Prediction: A p.125 Computer-aided p.l26 Reachability Analysis of
Classification of Patents Service Process Model Based Comparison of MARS, BPN Oriented to TRIZ on Polychromatic Sets and SVR in an Emerging Yanhong Liang', Runhua Tan', Xinqin Gao', Yan Li', Mingshun Market Chaoyang Wang', Zlti-Guang Li' Yang1, Qilong Yuan1 Clti-Jie Lu', Chih-Hsiang Chang2, 1Hebei University of Technology, China 1Xi'an UniversityofTeclmology, China Chien-Yu Chen1, Chih-Chou Cltiu', 2Hebei Polytechnic University, China Tian-Shyug Lee'
1Ching Yun University, Taiwan 'National Taipei University of Technology, Taiwa11 'Fu ]en Catholic University, Taiwan
44
p.l26
p.126
p.126
p.126
p.l27
Safety Feasibility Analysis on the Liquid Organic Heat Transfer Material Heater Used in the Production Process of Bleaching Powder Concentrate Hui Cui', Zhisheng Xu1, Wenhua Song' •Central South Universrty, Chma 'Tranjin University of Technology, China
Stochastic Analysis on Probability of Fire Scenarios in Risk Assessment to Occupant Evacuation Guanquan Chu', )inhui Wang' •Waterborne Transportation Institute, Ministry of Transport, China •Shanghai Maritime University, China
The Characteristics of Temperature Near the Ceiling of Liquid Fires in Vertical Laminar Clean Room Environments Yan Huo', Ye Gao1, Hong Mei Wu1,
jian He Zhao' 1Harbin Engineering University, China
Large Eddy Simulation of Smoke Movement in a Teaching Building jian He Zhao', Ye Gao', Hong Mei Wu', Yan Huo1
•Harbin Engineering Ut1iversity, China
Study on the Assessment Method of Agroecosystem Health Based on the Pressure-State-Response Model Bo Li', H. L. Xie', H. H. Guo', Ying Hou' 1Beijing Normal University, China 'fiangxi University of Finance & Economics, Chir1a
p.l27 Modeling and Analyzing Safety-critical Parallel-series System Safety Qing Sun', Lirong Cui', Rong Pan' 1Beijing Institute of Technology, China 'Arizona State University, United States
p.l27 Integrating Socio-technological Factors Analysis Into Nuclear Power Plant Event Report and Safety Evaluation: A Systematic Framework Ziqing Zhai1
•Shanghai Jiao Tong University, China
p.l27 Comparsion the Maintenance Between Two-Unit Parallel Standby Systems and 2-out-of-3 Stand by Systems MinWang1
1Chaoyang University of Technology, Taiwan
p.l27 Risk Analysis of the City Gas Pipeline Network Based on the Fault Tree Yi-lin Yin1, Guang-li Lin1
1 Tianji11 University, China
p.127
p.127
p.127
The Empirically Comparative Analysis of Advanced Manufacturing Paradigm of Chinese, Japanese and South Korean Enterprises }un-yi Ren' 1Yantai University. Chma
Maintenance Behavior-Based Prediction System Using Data Mining Pedro Bastos', Rui Lopes', Luis Pires1, Tiago Pedrosa' 'lnstituto Politecnico de Bragan~a, Portugal
Concept Analysis for Service Oriented Manufacturing Based on Interpretive Structural Modeling Yi-song(Lydia) Zheng', Dong Li', FengZhao' 'Nankai University, China
45
MAINTENANCE BEHAVIOUR-BASED PREDICTION SYSTEM USING DATA MINING Pedro Bastos, Rui Lopes, Luís Pires, Tiago Pedrosa,
Instituto Politécnico de Bragança, Campus de Santa Apolónia, Apartado 1038 - 5301-854 Bragança- Portugal ([email protected], [email protected], [email protected], [email protected])
In the last years we have assisted to several and deep changes in industrial manufacturing. Induced by the need of increasing efficiency, bigger flexibility, better quality and lower costs, it became more complex. The complexity of this new scenario has caused big pressure under enterprises production systems and consequently in its maintenance systems. Manufacturing systems recognize high level costs due equipment breakdown, motivated by the time spent to repair, which corresponds to no production time and scrapyard, and also money spent in repair actions. Usually, enterprises do not share data produced from their maintenance interventions. This investigation intends to create an organizational architecture that integrates data produced in factories on their activities of reactive, predictive and preventive maintenance. The main idea is to develop a decentralized predictive maintenance system based on data mining concepts. Predicting the possibility of breakdowns with bigger accuracy will increase systems reliability.
Perspec(ves Maintenance function is seen to the enterprise as a cost; Maintenance function, seen as a one, became more and more requested to contribute to cost
reduction, based on bigger and consistent equipment reliability Manufacturing systems recognize high level costs due equipment breakdown, related with
inspection costs, repair costs as well as costs associated with non production time or equipment non utilization;
Enterprises need to cope with market expectations, incorporating in their production philosophies new paradigms such as JIT- Just in time, MTO- Make to order, Mass Customization, Agile or Lean manufacturing.
Mo(va(on/Objec(ves In literature there are some approaches that use data mining concepts to improve
manufacturing activities.
It isn’t so common to find approaches that use it to improve the capacity of predicting behaviors based on historical data.
In fact, if the case is the possibility of a distributed collaboration of independent enterprises sharing data between them, even if they are competitors, the examples are even reduced.
Create a system that will help enterprises to collect, extract and create knowledge in a way that
enterprises will predict with more accuracy the moment to realize maintenance actions and thus improve the productivity of manufacturing process.
Pressure/support factors
► Globalization, new markets coming in sight
► Bigger exigency in products quality
$ $$
► Development and incorporation of information and communication technologies
► Products life cycle and time to market decreasing
► Technological changes and products bigger complexity.
► Mass Customization
7 56
121110
8 4
21
9 3
Remote data management
and communication
A1
P. 3
Knowledge Prediction System
A2
P. 4 Information Synthesis and
event generation
A3
Predictive Maintenance dataPreventive Maintenance dataReactive Maintenance dataManufacturers dataPartners Information
Status eventAlert event
Request Knowledge Base Event
Active Knowledge Base
Specification parameters
C1 Artificial neural networks
C6 Data Visualization
C2 Rule Induction
C3 Nearest neighbor method
C5 Genetic Algorithms
C4 Decision Trees
Behaviour Matrix
M1Prospective Agents
M2 Knowledge data engineM3 Pattern recognition AlgorithmsM4Communication chanel and tools
Behaviour Data Storing
I1Predictive Maintenance data
I2 Preventive Maintenance data
I3 Reactive Maintenance data
I4 Manufacturers data
I5 Partners Information
C2
Request Knowledge Base
C1
Event
Data request analisis and specification
A11
Data Management and Selection
A12
Knowledge Base definition
A13
Decision Support
formalization
A14O2
Alert eventO1
Status event
Selection specifications
Selection constrains
O3Active Knowledge Base
O4Specification parameters
M2Communication chanel and tools M1Prospective Agents
M3Knowledge data engine
I6 Behaviour Data Storing
O1Request Knowledge Base
I1Active Knowledge Base
I2Specification parameters
C1Artificial neural networks C2 Rule InductionC3 Nearest neighbor method
C4 Decision TreesC5 Genetic Algorithms
C6 Data Visualization
O2Behaviour Matrix
Data Mining processing
A21
Pattern Behaviour generation
A22
Proactive Faillure
detection module
A23
Organizational KnowledgeRelations IdentificationData InterpretationRules
M2Knowledge data engine
M3Pattern recognition Algorithms
M1Communication chanel and tools
Proactive faillure notification
I3Partners Information
M. Preven(va M. Predi(va M. Correc(va
M. Preven(va M. Predi(va M. Correc(va
M. Preven(va M. Predi(va M. Correc(va
Behavior Pa;ern genera(on ayer
Events dessimina(on
SYSTEM FUNCIONALITY