DIAGNOSTICS SYSTEM FOR MANUFACTURING
PROCESS PROBLEMS (DSMPP)
OKFALISA
UNIVERSITI TEKNOLOGI MALAYSIA
iii
To my beloved husband.
To my honorable mother and father
iv
ACKNOWLEDGEMENT
I would like to thank my supervisor PM.Dr.Naomie Salim and my co supervisor
Dr.Wong Kuan Yew for the help and advice they have offered me during my studies. I
am also grateful for the examiners PM.Wardah and Dr.Razak for correction and
suggestions. I would like to give a very special thank to Mr.Jamaludin bin Parmin,
Mr.Zulkifli, Miss. Hayati and Mr. Firdaus Saleh from White Horse Ceramics for the data
and discussion. I also wish to thank to Universitas Islam Negeri Sultan Syarif Kasim
Riau and Pemerintahan Daerah Propinsi Riau for the sponsorships.
My thanks go as well to my family, especially for my beloved husband, my
daughter “Rara”, my father and my mother for the supporting and praying. I also wish
to thank my friends, Fikri, Imam, Hafiz, Zarnelly, Mimi and Wiwied for the discussion
and friendship.
v
ABSTRACT
Facing the market competition, fulfilling the customer satisfaction and increasing
the product quality in manufacturing firms motivated investigation and diagnosis in their
production output failure. To gain correct and accurate diagnostic, the entire process
must be recorded and controlled in every step of manufacturing. In this project, a
prototype system has been developed to record the knowledge base that was used to
diagnose the source of tiles defects and to recommend action to solve the problems. This
system consists of two main components, the knowledge base component and the
inference engines. The knowledge base has been developed by analyzing the data and
information that is related to the tiles defect, such as symptoms, probable causes, types
of defect, processes, sub processes, tile classifications and recommended actions. On the
other hand, the inference engines has been built by implementing the forward chaining
method to root the causes of defect and depth first search in searching procedures. The
analysis proves that this system can help workers in the company to diagnose tiles defect
and solve problems regarding the defect. Besides that, the system can also help share
and transfer knowledge among knowledge workers in manufacturing firm.
vi
ABSTRAK
Bagi menghadapi persaingan pasaran, memenuhi kepuasan pengguna dan
meningkatkan kualiti produk di sektor pembuatan, kajian terhadap punca kegagalan
produk perlu dilakukan. Untuk mengenal pasti kegagalan produk, keseluruhan rantaian
proses pembuatan perlu direkod dan dikawal setiap saat. Didalam project ini, satu
prototipe sistem telah dibangun untuk merekodkan pengetahuan berkenaan dengan
punca-punca kerosakan jubin dan langkah-langkah perbaikkan yang boleh diambil.
Sistem ini terdiri dari dua bahagian, yaitu pangkalan pengetahuan dan engine carian.
Pangkalan pengetahuan dibangun dengan menganalisa setiap data maupun informasi
berkaitan dengan kerosakan jubin, berupa informasi gejala kerosakan, penyebab
kerosakan, jenis-jenis kerosakan, proses pembuatan yang mengalami kerosakan,
klasifikasi jubin dan cadangan penyelesaian untuk setiap kerosakan.
Sementara itu
engines carian dibangun dengan menerapkan kaidah rantaian kehadapan dan
pencaharian pertama kedalaman. Analisa menunjukkan bahwa sistem ini dapat
membantu para pekerja untuk mendiagnosis kerusakan jubin dan mengatasi setiap
kerosakan yang dijumpai dalam proses pembuatan. Selain itu, sistem ini juga dapat
membantu proses penyebaran pengetahuan dikalangan pekerja profesional dalam suatu
perusahaan.
vii
TABLE OF CONTENTS
CHAPTER
TITLE
PAGE
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENTS
iv
ABSTRACT
v
ABSTRAK
vi
TABLE OF CONTENTS
vii
LIST OF TABLES
xi
LIST OF FIGURES
xii
LIST OF APPENDICES
xiv
1.
2.
CHAPTER 1 PROJECT OVERVIEW
1.1 Introduction
1.2 The Background of the Study
1.3 Statement of the problem
1.4 Objective of the project
1.5 Scope of the project
1.6 Importance of project
1.7 Summary
1
3
4
5
5
6
7
CHAPTER 2 LITERATURE REVIEW
2.1 Introduction
2.2 Knowledge Base System
2.2.1 Knowledge Base
2.2.1.1 Definition of Knowledge Base
2.2.1.2 Type of Knowledge Base
2.2.2 Expert System
2.2.2.1 Definition of Expert System
2.2.2.2 Key component of an Expert System
8
9
9
9
11
12
12
13
viii
2.3
2.4
2.5
2.6
3.
2.2.3 Reasoning in ruled base systems
2.2.4 Comparison of chaining methods
2.2.5 Search Strategies used in Problem Solving
2.2.6 Type of Task Carried Out by Expert System
2.2.7 Advantages of Expert System
Knowledge Management System
2.3.1 Definition
2.3.2 The purpose of KMS
2.3.3 Knowledge Management Architecture
2.3.4 Impacts KMS to Organizational Effectiveness
2.3.5 Knowledge strategy goals
Manufacturing Process
2.4.1 Definition
2.4.2 Manufacturing systems
2.4.2.1 Production Characteristics
2.4.2.2 Mechanization and Automation
2.4.2.3 Assembly
2.4.2.4 Quality Control
2.4.2.5 Function of Quality Control
Benchmarking Study
2.5.1 Integrated Diagnostic system for production and service
2.5.2 Knowledge-based expert system in manufacturing
2.5.3 An On-Line Diagnostic Expert System for Intelligent
Manufacturing
Summary
CHAPTER 3 RESEARCH METHODOLOGY
3.1 Introduction
3.2 System Framework
3.3 Project Methodology
3.2.1 Planning Phase
3.2.1.1 Project Initiation
3.2.1.2 Project Management
3.2.2 Analysis Phase
3.2.2.1 Analysis Strategy
3.2.2.2 Requirements Gathering
3.2.2.3 System Proposal
3.2.3 Design Phase
3.2.3.1 Design Strategy
3.2.3.2 Architecture Design
3.2.3.3 Designing Database and file specification
3.2.3.4 The Program Design
3.2.4 Implementation Phase
3.2.4.1 System Construction
3.2.4.2 Installations
15
16
17
20
21
23
23
24
24
26
27
28
28
29
29
30
31
31
32
33
33
34
35
37
38
38
40
40
41
41
41
42
42
42
43
43
43
44
44
44
44
45
ix
3.4
3.5
3.6
3.7
4.
5.
3.2.4.3 Support Plan
System Development Methodology
Hardware and Software Requirements
Project Schedule
Summary
45
51
51
53
53
CHAPTER 4 ANALYSIS AND DESIGN
4.1 Introduction
4.2 Organization Analysis
4.3 As Is Process
4.3.1 Production Process Operation
4.3.1.1 Flowchart Operation Process
4.3.2 Detection the Tiles Defects
4.3.3 Production Classification
4.3.3.1 Grade A
4.3.3.2 Grade B
4.3.3.3 Grade C
4.3.4 The Tiles Inspection
4.3.5 Identify the knowledge base
4.3.5.1 The Tiles Defect and Probable causes
4.3.5.2 The Tiles Defect and Classification
4.3.6 Knowledge Representation
4.3.6.1 Rule Base
4.3.6.2 Decision Three
4.3.6.3 The User Interface
4.3.7 Knowledge Management
4.4 To-be-Process
4.5 System Architecture
4.6 Physical Design
4.6.1 Database Design
4.6.2 Program (Structure) Chart
4.6.3 User Interface Design
4.7 Summery
57
57
59
59
59
62
62
63
64
64
65
66
66
67
69
70
74
76
84
86
92
94
95
95
97
99
IMPLEMENTING AND TESTING
5.1 Introduction
5.2 System Implementation
5.2.1 Database Development
5.2.2 System Development
5.2.2.1
Product
5.2.2.2
Symptom
5.2.2.3
Defect
5.2.2.4
Causes
5.2.2.5
Process
5.2.2.6
Sub Process
100
100
101
104
106
107
108
110
112
112
x
6.
7.
5.2.2.7
Recommendation
5.2.2.8
Question
5.2.2.9
Answer
5.2.2.10 Rule Base
5.2.2.11 Diagnose
5.2.2.12 Login
5.3 Test Result/System Evaluation
5.3.1 Unit Testing
5.3.2 Black box Testing
5.3.3 White box Testing
5.3.4 Usability Testing
5.4 User Manual
5.5 Summery
112
113
113
113
115
116
116
117
117
124
125
126
126
ORGANIZATIONAL STRATEGY
6.1 Introduction
6.2 Roll Out Strategy
6.2.1 Installation of Infrastructure Process
6.2.2 Trainings
6.2.3 System Implementation Process
6.2.4 The Organizational Support
6.3 Summery
127
128
128
129
130
130
130
DISCUSSION AND CONCLUSION
7.1 Introduction
7.2 Achievements
7.3 Constraints and Challenges
7.4 Aspiration
7.5 Summery
132
132
133
135
135
REFFERENCES
APPENDICES A-K
137
139 - 250
xi
LIST OF TABLES
TABLE NO
TITLE
PAGE
3.1
Activities in Project Development Phase
46
3.3
Software required to developing the system
52
4.2
Hardware and Software Specification
94
5.1
Table Black box Testing Process
118
xii
LIST OF FIGURES
FIGURES NO
2.1
TITLE
Key Component of an Expert System
PAGE
14
2.2
A Depth First Search
18
2.3
A Breadth First Search
19
2.4
Heuristic Search
19
2.5
Knowledge Management Architecture
25
2.6.
The KM/OM/OL Model (Jennex-Olfman, 2002)
27
2.7
29
2.8
Major Interaction between manufacturing and other
industrial activities
Principle of Integrated Quality control
2.9
Architecture of Integrated DES
34
2.10
36
3.1.
Simplified frameworks in influence diagram Knowledge
base
System Framework
3.2
Gantt chart
55
4.1
Production Process Flowchart
60
4.4
Examples of defects-1
67
4.5
Examples of defects-2
68
4.6
Examples of defects-3
68
4.7
Examples of defects-4
69
4.8
The Rule Base Process
70
4.9
The semantic Network
71
4.10
Decisions Three Diagram
75
4.11
The structure of DSMPP system
77
32
39
xiii
4.12
Dialogues route analysis process
78
4.13
Knowledge Management Structure
85
4.14
Use Case Diagram
87
4.62
Class Diagram
89
4.63
State Chart Diagram: User Class
90
4.64
State Chart Diagram: Tile Class
91
4.65
DSMPP System Architecture
92
4.66
The Structure Chart of DSMPP System
96
4.67
Window Navigation Diagram for DSMPP System
98
5.1
Knowledge base relationship
101
5.2
Rule base relationships
102
5.3
The symptom page
107
5.4
The symptom_def page
108
5.5
The Defect Page
109
5.6
The Defect Relation Page
110
5.7
The Cause Page
111
5.8
The Cause Relation Page
111
5.9
The Rule Base Page
115
5.10
Login Page
116
5.34
Diagnose Page
188
5.35
Result Page
189
5.36
Diagnose-causes Page
189
xiv
LIST OF APPENDICES
APPENDIX
TITLE
PAGE
A
Quality Control Plan and Form Report
139
B
Manual ways to detect the defect
144
C
Sequences Diagram
146
D
Tiles Defect and Probable Causes Table
176
E
Organization Chart of White Horse Ceramic
191
F
Usability Testing Form
192
G
Rule Base
194
H
Use Case Description
198
I
Class Diagram
224
J
Database Design
234
K
User Manual
238
CHAPTER 1
PROJECT OVERVIEW
1.1
Introduction
The ability to mechanize or automate has become a major requirement of those
manufacturing fields that are planning to compete well into the future that combines
process expertise, advanced software and mechanical systems in unique and creative
ways. Regardless of the precautions and best efforts of all management in the company,
a time may come when one of the products dies in the field. Usually the cause is either
improper use or the failure of a part.
Safety concerns have become more important in product design and how to
produce well of product quality. But the great pressure is to develop new products fast.
So developers often lack the time to fully test product-quality features. As a result, the
manufacturing companies must be sure that the companies have a program in place to
effectively and efficiently respond to product failures. Otherwise the rush to market may
levy a "pay me now or pay me later" tax on company profits. The manufacturing
companies must be prepared to react quickly to hazards that might arise when one of its
products fails. It must also be ready to give customers enough information to correct the
problem.
Unfortunately, many manufacturing companies lack a program that outlines how
to investigate and diagnose product failures. An incident can have serious consequences.
There must be a prompt investigation to diagnose its root causes.
2
After a failure, it is imperative that a representative of the manufacturer quickly
visit the site of the incident. The representative should ask some basic questions during
the diagnostic, including:
-
Was the product defect?
-
What is the product defect?
-
How does the defect product conditions?
-
What the common causes of the product defect?
-
How to manage or solve the problem occurred?
-
Which parts of the manufacturing process causes the defects?
Actually all the problems especially in production must be communicated to all
employees. Personnel must be aware of their responsibilities if a product fails.
Sometimes the causes are in the machine problems but it is not impossible that the
causes are the human factors, dereliction of duty. Knowing the root causes of the defect
or fail product needs special skill, experiences and knowledge base from the experts in
manufacturing area including machines, manufacturing process, specification standard
production process (ISO) and controlling. To gain the correct and accurate diagnostic,
the entire processing root must be recorded and controlled on every step of
manufacturing process, from the incoming raw material in mixing or grinder process,
preparation/semi manufacturing, building/assembly, curing, inspection until product
claims after sales service.
Lines of knowledge base obtained from the experts should be communicated,
from top management downwards because it will be reflected to the product quality,
management process, company profit and responds to the critical competitor.
In this study, the researcher proposes a prototype of the management system that
aid to diagnose the source of product defect in manufacturing and besides
troubleshooting information for any type failure production by studying the defect
characteristic of knowledge management in manufacturing process, the quality of the
product can be controlled.
3
In order to capture the domain knowledge base concepts in diagnostics system
for manufacturing process problems (DSMPP), case study is doing on manufacturing
company in Johor Bahru, White Horse Ceramics Industries Sdn.Bhd located in Pasir
Gudang.
The remaining part of the study is divided into three chapters. The second
chapter presents the literature review; while the third presents the research methodology.
The initial findings and analysis are discussed in chapter four.
1.2
The Background of the Study
The manufacturing industries are facing a rapidly changing landscape and new
challenges are evolving. The competitive edges and innovation that can be developed
include integrating quality services into well - established manufacturing processes for
efficiency and effectiveness. The manufacturing challenges for the industrial practices in
the twenty first century are concerned with the manufacturing solution, not just products
and machines and build with confidence and guaranteed performance but also industries
are giving tighter specifications and are asking for manufacturers to be responsible for
the manufacturing losses caused by the failure of the manufacturing product.
To ensure the production quality and fulfill the customer satisfaction,
technologies are needed to monitor the performance of manufacturing process and the
flow of activity run based on the procedure. The problems occurred in manufacturing are
caused by the defect in the machines or human factor problems such as the operator of
the machines mechanism. When an unknown error is resident in a production, others
trust and build upon it. The longer it goes undiscovered, the more disruptive the effect to
other processes.
4
The proposed of this project, Diagnostics System for Manufacturing Process
Problems (DSMPP) is to help by diagnose the sources of product defects and gives the
solutions to suggest troubleshooting to manage and maintain the defects.
Diagnostic of cause the defect is obtained from the expertise’s knowledge and managed
in a database system for knowledge management sharing for all components in the
manufacturing firm. The defect knowledge management is hoped to reduce the problems
and increase the effectiveness and efficiency in manufacturing process.
1.3.
Statement of the problem.
The problem statements of this project are:
1. What are the characteristics of the system that can aid diagnostics of product
defects in manufacturing firm?
2. How can a system to diagnose the source of defect product for
manufacturing firm be building knowledge base with forward chaining in
inference engines?
3. How effective is a system to diagnose the product defect help in
manufacturing firm?
This study attempts to provide answers to these questions and other related ones.
The researcher intends to use a relevant system development methodology to develop a
prototype of diagnostic system for manufacturing process problems which use
knowledge base with forward chaining concept method rules in inference engines
procedure. The system which enhances to diagnose source of defect problems and
support problem solving to manages and handles the process problems. The tool will be
tested at White Horse Ceramics Industries Sdn.Bhd.
5
1.4.
Objective of the project
The main objective of the project is to build a prototype of system to diagnose
the source of product defect in manufacturing process problems, another objective are:
·
To study the characteristics of defect knowledge management in manufacturing
process.
·
To build a prototype of management system that can help to diagnose the source
of product defect in manufacturing firm.
·
To test the prototype of the system that can help to diagnose the product defect in
manufacturing firm.
1.5.
Scope of the project
In developing this project, I have determined and set the limit and scope for the
case study of this project. The scopes are:
i. The manufacturing process special for automate machine problems defects
ii. The prototype of the system are limited for use in ceramic industries
manufacturing
iii. The diagnostic of defect problem is to be traced to glazing and firing processes.
iv. The defect or rejected product is only regarding to line production for certain type
of ceramic tiles.
v. The diagnostics for the sources of product defects is through forward chaining
concept method rules in inference engines procedure.
vi. The system that will be developed will only be used by top level management,
quality control department and research & development department.
6
1.6.
Importance of project
By implementing ICT strategies in manufacturing company, specifically DSMPP
system, it is hoped that it can bring some effective and efficient improvement to
company for managing the product defect in manufacturing process problems. The
project will give a new way management concept to face the competitor in business
field.
Some benefits of the system for the organizations, manufacturing firms, engineers,
Quality Assurance personnel are:
1. Organization:
i.
ii.
Propose effective management strategies to the top management.
Give the business a superior competitive advantage, and offer better service
and improve customer satisfaction by producing the good quality of the
product.
iii.
Pervade information technology and communication (ICT) concept in
business organization.
2. Manufacturing Firms:
i.
Help in reducing the defect products in manufacturing production.
ii.
Help in diagnosing the product defects which influencing in manufacturing
process and production.
3. Engineers
i.
Manage the knowledge sharing in the company after the experts left.
ii.
Manage the expertise’s knowledge and skill to diagnose the product defect.
iii.
Sharing the knowledge for the new worker.
4. Quality Assurance , Quality Control, R&D personnel
i.
Give the solving problems for each causes of the defect for maintenance in
effectiveness and efficiency.
ii.
Help in diagnosing the defect problems in production processes.
iii.
Sharing the knowledge representation for reducing the defects.
7
1.7
Summary
In this first chapter a brief introduction about the project and how the project is
going to be implemented has been discussed. The problem background and statement
has also been discussed in this chapter to give an introduction of the project and to
explain why this project has been proposed. The objective, scope and the importance of
this project have also been pointed out. Hopefully, by developing the project
successfully, the objective and aim of the project can be achieved.
CHAPTER 2
LITERATURE REVIEW
2.1
Introduction
In the second chapter, the study centers on the review of the literature on this
subject. The chapter presents a brief introduction and information to the curious readers
and researchers on the area of diagnostic the defect product in manufacturing process.
The aim is to broaden the understanding of the interested readers and also unfold the
uncovered areas in literature which the study attempts to make a useful contribution. To
reach the objectives of this project, the main concepts of knowledge base which uses as
a method to solve the defect problems in diagnostics, the way of knowledge
management system maintains the knowledge worker and sharing information among
the expertise and ceramics production in manufacturing firms must be understood well.
The sub-topics that will be discussed in this chapter are:
i.
Knowledge Base System
ii. Knowledge Management System
iii. Manufacturing Process
iv. Benchmarking Study
9
The aspects will be analyzed and discussed in each sub-topic. All the information
gathered will be used as guide in developing the prototype system necessary in
execution of the whole project. This information was helpful in establishing the
theoretical background of knowledge base for diagnostics of defects.
2.2.
Knowledge Base system
The knowledge base system that has been chosen and implemented as method in
this project has been explained detail in this chapter.
2.2.1. Knowledge Base
Knowledge base systems or expert systems are computer programs embodying
knowledge about a narrow domain for solving problems related to that domain.
2.2.1.1. Definitions of Knowledge Base
Knowledge base is defined as is a special kind of database for knowledge
management which provides the means for the computerized collection, organization,
and retrieval o f knowledge(wikipedia). The Knowledge base can also be defined a s
encoded knowledge for an expert system and in a rule-based expert system, a
knowledge base typically incorporates definitions of attributes and rules along with
10
control information (expert system glossary). Knowledge base format is specific to the
expert system shell or other software for system implementation.
Knowledge understanding is gained through experience or study, it is know-how
or a familiarity with the way to do something that enables a person to perform a task. It
may also be an accumulation of fact, procedural rules, or heuristics (Awad, 1996). These
things are defined as follow:
·
A Fact is a statement of some element of truth about a subject matter or a
domain.
·
A procedural rule is a rule that describes a sequence of relations relative to
domain.
·
A Heuristic is a rule of thumb base on years of experience.
Knowledge is concepts, experiences and insight that provide a framework for
creating, evaluating and using information (Laudon-Laudon, p. 373).
Regarding to (Frost, 1986) knowledge base system is a set of resources hardware, software, and possible human – whose collective responsibility include
storing the knowledge base, maintaining the security and integrity and providing user
with the required input/output routines, including deductive retrieval facilities, so that
the knowledge base can be accessed as required.
Knowledge base contains domain knowledge which may be expressed as any
combination of “If- Then” rules, factual statements, frames, objects, procedures and
cases. The inference mechanism is that part of an expert system that manipulates the
stored knowledge to produce solutions to problem (Cakir-Cavdar ,2005).
11
2.2.1.2. Type of Knowledge base
Knowledge is classified into variety of types, including:
·
Shallow and Deep Knowledge
Swallow, or radically recalled surface, knowledge indicates minimal
understanding of the problem domain. Deep knowledge acquired through years
of experience would be required to decide some decision.
·
Knowledge as Know- how
It accumulate lessons of practical experience, is what is needed for building
expert system. Know-how distinguishes an expert from a novice, especially
when building an expert system base on an expert judgment.
·
Common Sense as knowledge
The type of knowledge that all human beings possessed in varying forms and/or
and varying amounts. It is collection of personal experiences and facts acquired
over time and type of knowledge that human tends to take for granted.
·
From Procedural to Episodic knowledge
Procedural Knowledge: an understanding of how to do task to carry out a
procedure, it usually involves psychomotor skills; Declarative Knowledge: is
information that experts can easily talk about. On the other hand, procedural
knowledge is awareness knowledge or routine knowledge of which the expert is
conscious about semantic knowledge: is deeper kind of knowledge which is
organized into chunked knowledge that reside in long term memory; Episodic
knowledge is a knowledge base on experiential information or episodes, the
longer an expert takes to explain or verbalize his or her knowledge, the more
semantic or episodic it is.
12
From the literature about the knowledge base, the author can conclude that the
knowledge is data and information that collected in databases which could act the
domain facts by using the “if then” rules procedure to provide the solution to problems.
Regarding to this project the knowledge base are the information or any facts of defects
problems founded in glazing and firing process, the network defects, causes of problems,
sysmptom and recommendation given to solve the problems by using the “if then” rules
procedures.
2.2.2.
Expert System
The expert system consept that have been adopted in this project are discussion
in this section.
2.2.2.1. Definitions of Expert System
What's an expert? An expert is one who possesses specialized skill, experience,
and knowledge that most people do not have along with the ability to apply this
knowledge using tricks, shortcuts, and rules-of-thumb to resolve a problem efficiently
( Harmon-King, 1985).
An expert's advice has to be good enough most of the time for the expert to keep
his or her reputation, but is not expected to be perfect or even the globally best available
to be considered useful ( Hayes-Roth, 1983). There are a lot of attributes of effective
consultants and consulting: consulting is goal oriented, a good consultant is efficient, a
consultation is adaptive, consultants are able to work with imperfect information, and
good consultants justify their recommendations by explaining their reasoning.
13
An Expert System is a knowledge-intensive computer program that captures the
expertise of one or more humans in limited domains of knowledge (Laudon & Laudon,
2005)
An Expert system is a system which is capable of carrying out a task generally
regarded as being difficult and requiring some degree of human expertise (Frost, 1986)
Regarding to this project the experts are the person who are the knowledge
absorbed in specific skills for managing and controlling the productions process, such as
engineers, the quality assurance personnel, quality control and R&D personnel.
2.2.2.2. Key component of an Expert System
An Expert system has four components (refers to Figure 2.1):
·
Knowledge base
The knowledge base is the repository of the rules, facts and knowledge acquired
from the human expert. Most expert system use production rules, which is why
expert system are often referred to as a rule base systems.
·
Inference Engine
The inference engine is a cluster of computer programs that coordinate the
reasoning and inference based on the rules of the knowledge base to come up
with solution.
When the system searches for an appropriate rule, it may not arrive at a single
conclusion, but number of possibilities having different degrees of certainty.
Depending on the domain, the inference engine may use backward chaining or
forward chaining.
14
·
Justifier/Scheduler
Part of the inference engine, the scheduler is set up to coordinate and control the
sequencing of the rules. It is important because it coordinates and ensures
efficient use of knowledge base.
·
User Interface
It facilitates all communication between the user and the system. The system
asks for information through questions or multiple choice menus, and the user
answers by typing on the keyboard.
User
Invironment
USER
INTERFACE
INFERENCE
ENGINE
JUSTIFIER
(Explain the “how” and
“why” of an answer in
plain English )
Operational
Invironment
(The Problem Solving
Mechanism )
1. Reasoning
2. Inferencing
3. Searching
4. Conclusion
SCHEDULER
(Coordinates and
controls rule
processing )
KNOWLEDGE BASE
(Domain Knowledge
1. Facts
2. Rules
Facts
Developmental
Invironment
Rules
KNOWLEDGE
ACQUISITION
( Interaction between
knowledge engineer
and domain expert )
1. Rule definition
2. Verification /
validation of knowledge
acquired
Figure 2.1 Key Component of an Expert System
15
2.2.3. Reasoning in ruled base systems
The rule base contains rules called production rules to represent general
knowledge about the problem domain. The condition part of rule is established by
reference to the database and/or by questioning the user of the production system. The
ruled base methodology was originally proposed by Post (1943) and has since been used
in various application, the rule base approach was used in construction of expert systems
such as DENDRAL (Buchanan et al, 1969), and MYCIN (Shortliffe, 1976) and in the
HEARSAY speech recognition system (Lesser et al, 1975).Since then, the rule base
approach has been used extensively in the construction of “intelligent systems”.
The reasoning models or approaches to problem solving which are typically
implemented in rule-based expert systems are:
·
Forward chaining (The Data-Driven Approach).
The Forward chaining methods begins with a set of known facts or attribute
values and applies these values to rules that use them in their premise. Any rules
that are proven true fire and produce additional facts that are again applied to
relevant rules. The process continues until no new facts are produced or a value
for the goal is obtained. This approach works well when it is natural to gather
multiple facts before trying to draw any conclusions and when there are many
possible conclusions to be drawn from the facts.
·
Backward chaining (Goal driven approach). An alternative approach begins
with a rule that could conclude the goal for the consultation tries to obtain values
for the attributes used in the rule's premise, then backtracks through additional
rules if necessary to determine a value of the goal attribute. When there are many
attributes employed in many rules, the backward chaining mechanism produces a
more efficient interview than forward chaining because it will not be necessary
to ask the user to input values of all of the facts.
16
·
Mixed Method
Data driven and Goal driven approach can be combined in various ways.
An expert system's reasoning mechanism might employ either or both chaining
techniques. Knowledge representation and reasoning mechanisms are combined with a
user interface in software that represents the complete expert system.
2.2.4. Comparison of chaining methods
·
Backward chaining which has goal oriented behavior is efficient because it
avoids requests for input hat won’t contribute to determining the value of the
consultation’s goal. As a result, it provides the foundation for most rule based
expert systems. Backward chaining systems are described as hypothesis driven
because they operate by selecting successive that can determine the value of a
goal or sub goal: this value becomes the hypothesis to be proven or disproved.
·
In some interview scenarios it is natural collect data in advance, perhaps using a
paper questionnaire. In other cases input to an expert system is collected
automatically, perhaps using sensors on a machine. For these two situations the
forward chaining approach makes sense. Forward chaining systems are described
as data driven because they deduce everything they can from an asset of data
rather than working backward from a hypothesis.
·
To provide the most flexibility, many expert system shells support both forward
and backward chaining even in the same interview. For example, some initial
data might be requested and forward chained before the backward chaining
operation of the inference engine is started. The inference engine's control
capabilities enable this flexibility. Expert systems are often able to deal with
attributes that are assigned values with some degree of uncertainty
17
Relate to point 2.2.3 and 2.2.4, the rule base approach that has been chosen in
developing the prototype of DSMPP system is forward chaining, because regarding to
the expert’s interviews that has been done by the author, the first way to diagnose the
defect problems in tiles productions is by doing some inspections of the probable causes
and symptom. The author collects the entire facts of defect problems from the experts
before makes some conclusion and recommendation with the product defect.
2.2.5
Search Strategies used in Problem Solving
The data driven and goal driven approaches may both be regarded as attempts to
find a path linking the initial problem state to the goal state. In the data driven approach
(forward chaining) the system searches for a path by identifying sets of applicable rules
whose condition parts are satisfied by database and applying rules from these sets. In the
Goal driven approach (backward chaining) the system searches for a path by identifying
sets of applicable rules from these sets as sub goals which it then attempts to establish.
In forward chaining the search moves from the initial problem state to the goal. In the
backward chaining the search moves in the opposite direction. In both cases the system
has to make decisions which have not yet been considered in any details. This decision
is determined by conflict revolution strategy, which is use by scheduler. Other decision
involve consideration of backtracking, depth first, bread first or heuristics.
The search strategies uses are:
·
The Conflict Resolution Strategy
The way in which a rule is selected from a set of applicable rules. It’s based on
the following principles: select rules arbitrarily; select a different rule in
successive stages; select the first applicable rule identified; rules are given values
which are used to determine which one to select.
18
·
Backtracking
It executes in the forward chaining and uses sub goal in backward chaining, the
rules can be identified and path is established linking the initial problem state to
the goal.
·
Depth First / Breadth First / Heuristics Search
Depth First Search is searching processes start at the root of the problem space
three and work down the leftmost branch to the end node before they embark on
any other branch, the advantages of depth first search is the simplicity with
which it can be implemented, another advantage is that requires less memory
than the alternative, breadth first search. See Figure 2.2.
Figure 2.2 a Depth First Search
Breadth First Search: This method can have huge overhead, at any stage, all
nodes to the left and all nodes above the node being develop must be memorized.
19
All of the nodes in the problem space three at depth leftmost are develop first
then all the nodes at depth right are develop and so on. See Figure 2.3.
1
3
8
2
4
5
depth 1
6
7
9
depth 2
depth 3
Figure 2.3 a Breadth First Search
Heuristic Search: It uses meta rules to determine which node to develop next. A
meta rules might then be used which state that if a choice exist, the rule with the
highest score should be applied. The resulting search will be neither depth first
nor breadth first. For the more explanation see Figure 2.4.
Figure 2.4 Heuristic Search
20
·
The Explicit Representation of Control Knowledge
The search strategy is coded as a procedure or function and control knowledge is
embedded in the code.
Implementing the scheduler process (refers to Figure 2.1), the DSMPP with
forward chaining rules, searches for a path by identifying sets of applicable rules whose
condition parts are satisfied by database. The searching rules apply on these sets start
from the root of the problem then work down the leftmost branch until the end node.
DSMPP’s searching process begins with the common question as the root, responding
the user interface (similar to binary search) then the applicable rules will search another
question that support the conclusion, the root embarks from right to left until the end
node. In the forward chaining for DSMPP this is the best and easier ways to determine
the searching strategy.
2.2.6. Type of Task Carried Out by Expert System
Most Expert tasks fall into one or other of the following categories:
·
Design: which involves the specification of a system or object such that the
system or object satisfies some given set of requirements and can be built using
some given set of resources
·
Diagnosis: which involves fault finding in a system given some sets of
symptoms
·
Interpretation: which involves the analysis of data to determine its meaning
·
Monitoring: which involves the continuous analysis of signal and the invocation
of action and/or alarm as appropriates.
·
Planning: This involves the creation of a plan of actions to achieve a given goal.
21
2.2.7. Advantages of Expert System
Expert systems have a number of attractive features (Giarratano-Riley, 1989):
·
Increased Availability
Expertise is available on any suitable computer hardware. In a very real sense, an
expert system is the mass production of expertise.
·
Reduce Cost
The cost of providing the expertise per user is greatly lowered
·
Reduce Danger
Expert system can be used in environments that might be hazardous for human
·
Permanent
The Expertise is permanent. Unlike human experts who was retired, quit or die,
the expert system knowledge will last indefinitely.
·
Multiple Expertise
The knowledge of multiple experts can be available to work simultaneously and
continuously on a problem at anytime of day or night. The level of expertise
combined from several experts may exceed that of a single human expert
(Harmon, 1985)
·
Increase Reliability
Expert system increase confidence that the correct decision was made by
providing a second opinion to human expert to break a tie in case won’t work if
the expert system was programmed by one of the experts.
22
The expert system should always agree with the expert, unless a mistake was
made by the expert. However this may happen if the human expert is tired or
under stress.
·
Explanation
The expert system can explicitly explain in detail the reasoning that led to a
conclusion. A human may be too tired, unwilling or unable to do this all the time.
This increases the confidence that the correct decision is made.
·
Fast Response
Fast or real time response may be necessary for some applications. Depending on
the software and hardware used, an expert system may response faster and be
more available than the human expert. Some emergency situation may requires
responses faster than a human and so a real time expert system is a good choice
(Hug, 1988) (Ennis, 1986)
·
Steady, unemotional and complete response at all the times
This may be very important in real time and emergency situation when a human
expert may not operate at speak efficiency because of stress or fatigue.
·
Intelligent tutor
The expert system may act as an intelligent tutor by letting the student run
sample program and explaining the system’s reasoning.
·
Intelligent Database
Expert system can be used to access database in an intelligent manner
(Kerschberg, 1986) (Schur, 1988).
23
2.3.
Knowledge Management System
2.3.1. Definition
(Alavi- Leidner, 2001) defined a Knowledge Management System as “IT
(Information Technology)-based systems developed to support and enhance the
organizational processes of knowledge creation, storage/retrieval, transfer, and
application.
(Maier, 2002) expanded on the IT concept for the Knowledge Management
System by calling it an ICT (Information and Communication Technology) system that
supported the functions of knowledge creation, construction, identification, capturing,
acquisition, selection, valuation, organization, linking, structuring, formalization,
visualization, distribution, retention, maintenance, refinement, evolution, accessing,
search, and application.
(Stein-Zwass,1995) define an Organizational Memory Information System
(OMS) as the processes and IT components necessary to capture, store, and apply
knowledge created in the past on decisions currently being made.
Regarding to the literature of knowledge management system, the author
concludes the definition of knowledge management system is DSMPP system as a tools
that is developed to support the manufacturing processes of knowledge creation,
capturing, acquisition, selection, valuation, sharing, accessing and applications to
improve the production effectiveness and efficiency in White Horse Ceramics.
24
2.3.2. The purpose of KMS
Organizations engage in Knowledge Management in order to enhance:
·
The productivity and quality, through innovation;
·
The stock of intellectual capital;
·
The relationships with customers, suppliers, shareholders, regulators;
·
The strategic decision- making;
·
The competitiveness; and/or
·
The profitability
2.3.3. Knowledge Management Architecture
Knowledge Management Architecture (Borghoff-Pareschi ,1998) see Figure 2.4.
·
Knowledge Repositories& Libraries : Documents
·
Knowledge
Cartography
consist
of Knowledge Navigation,
&Simulation
·
Communities of Knowledge Workers: People
Mapping,
25
Knowledge
Repositories
&
Libraries
Flow of
Knowledge
Communities
of
Knowledge
Workers
Documents
Distribution &
Sharing
People
Knowledge
Cartography
Knowledge Navigation,
Mapping, &
Simulation
Figure 2.5. Knowledge Management Architecture
The Knowledge Management System (KMS) consists of processes and
technologies for identifying and capturing knowledge, knowledge repositories, processes
for storing, searching, retrieving, and displaying knowledge, and users. The first
represents manual capture by individuals who identify knowledge to be retained and
then take the necessary steps to place the knowledge in a repository. The second is a
capture process integrated into automated processes. An automated capture process
requires that someone identify knowledge products or artifacts of the process up front so
that system designers can build databases and automated processes into the system to
capture the knowledge (simulation). As knowledge is used its impact should be
monitored and assessed. Knowledge found to improve organizational effectiveness
should be retained and possibly expanded. The knowledge workers as users are
emphasized since they are the ones using knowledge.
26
Implementing the knowledge management architecture in White Horse Ceramics
refers to Figure 2.5, the knowledge repositories and libraries are documented in database
at DSMPP system. The system absorbs the knowledge from the knowledge worker
(engineer, QA, QC and R&D personnel) or experts and builds the knowledge base, rules
and inference machines to diagnose the defect problems. The entire resources are
restored in database and will be distributed or shared among the personnel in the
organizations.
2.3.4. Impacts KMS to Organizational Effectiveness
To better understand KM, it needs to define the concepts of Organizational
Memory (OM) and Organizational Learning (OL). (Jennex-Olfman, 2002) found that
these three concepts are related and have an impact on organizational effectiveness.
Organizational effectiveness is measured by how well the organization performs those
activities critical to producing what the organization sells. OL is the process the
organization uses to learn how to perform these activities better. OL results when users
utilize knowledge. That OL may not always produce positive results can be determined
by monitoring organizational effectiveness. Effectiveness can improve, get worse or
remain the same. How effectiveness changes influences the feedback provided to the
organization using the knowledge. KM and OM are the processes used to identify and
capture critical knowledge. Knowledge workers and their organizations 'do' KM; they
identify key knowledge artifacts for retention and establish processes for capturing the
knowledge. OM is what IT support organizations 'do'; they provide the infrastructure
and support for storing, searching and retrieving knowledge artifacts.
27
Figure 2.6 illustrates these relationships and the following sections expand on
these concepts.
Figure 2.6. The KM/OM/OL Model (Jennex-Olfman, 2002)
2.3.5. Knowledge strategy goals
·
Agility: The early warning system – to spot trends and act quickly.
·
Collaboration: Two minds are better than one which the whole is greater than the
sum of its parts.
·
Decision- making: Real-time support for quality decisions that are widely owned.
·
Coherence: Collective sense- making that engenders continuous, dynamic
alignment.
·
Innovation: Creating value by building on each other’s ideas to offer better
solutions.
28
As the knowledge strategy goals of DSMPP that will be implemented in White
Horse Ceramics (refers to point 2.3.3) are the collaboration of the knowledge worker
developments in the organizations and the innovations to increase profit, effectiveness
and efficiency.
2.4. Manufacturing Process
2.4.1 Definition
Manufacturing is a human activity that pervades all phases of life. It has been
defined as “the making of goods and articles by hand or especially by machinery often
on the large scale and with division of labor (Schey, 2000).
The growth of industry which parallels the growth of manufacturing has since
led to undeniable advances, not only in providing an abundance of material possessions,
but also in creating economic basis for genuine improvements in the quality of life.
Manufacturing process is a series of complex interactions between materials,
machines, energy, and people. It begins with the creation of individual parts that finally
make up a finished product. The finished parts will have to be made into an end product
through assembly operations. In both unit assembly and unit operations, automation and
computer control play increasingly important roles. Technical improvements are
meaningful only if costs are also controlled along with the quality, and this requires
efficient organization of all phase of manufacture. Organizational and management
aspect can be neglected or ignore only at the expense of the competitiveness of the entire
manufacturing activity.
29
The multiple interactions indicated above require that many engineers and
technologist (including manufacturing, material, mechanical, industrial and system
specialist) who make up manufacturing team.(refers to Figure 2.7)
Figure 2.7 Major Interaction between manufacturing and other industrial activities
2.4.2. Manufacturing systems
2.4.2.1. Production Characteristics
Two important factors in the choice of process are:
·
Total number of parts produced
·
The rate of production
The rate of production is the number of unit produce in a time period such as an
hour, month or a year).
When the total production quantity is insufficient to keep production unit
continuously occupied, the lot size will be determined by weighing the cost of setting up
30
(changing over) against the costs of stocking (warehousing) parts between production
runs. All these decisions effect costs, profitability and productivity.
In terms of production characteristics, DSMPP system is limited the process in
one line production (line A), the reasons are because line A has the highest rate of
productions and the biggest total number in this type of tiles regarding to customer
demands.
2.4.2.2 Mechanization and Automation
Mechanization means that something is done or operated by machinery and not
by hand, industrial development has been mostly a story of mechanization. Feedback is
not provided and thus one deals with open loop systems (Schey, 2000)
Automation means a system in which many or all of the processes in the
production movement and inspection of part and materials are automatically performed
or controlled by self operating devices. This implies the essential elements of automation
include in addition to mechanization-sensing and feedback devices and some degree of
decisions making, thus a close loop system is created. Frequently, the meaning of
automation is loosely expanded to encompass mechanization, and then the terms
describe all action that make the life of the worker easier while also increasing
productivity. Automation reduces the number of operators and also demands a different
mix of skills, shifting some of production functions to specialists while removing certain
special skills from the operator level, or replacing the operator with a machine
supervisor.
31
2.4.2.3.
Assembly
The final phases of manufacturing, individual components are assembled into the
end product, the wide range of problems depending on production quantities. In
assembling a complex machine, great process can be made by breaking down the
operation into smaller unit, this also facilitates material handling by assuring that all
parts can be supplied the proper place and sequence.
2.4.2.4.
Quality Control
Quality control is a task must be shared by everyone involved in manufacturing.
Certain functions can and usually are centralized in quality control department. These
functions include examination of incoming materials and parts and of finished products
for conformance to specifications and may require extensive instrumentation and
laboratory facility facilities. The operators have to check all visible or measurable
manifestation of quality and it necessary to arrive at sampling strategy. Sampling
strategy is checking randomly the parts of production. Ideally all parts and assemblies
should be inspected at every important stage and economically accomplished with
automatic inspection techniques.
For details information of quality control plans in White Horse Ceramics refers
Appendix A at Table 2.1, the inspection steps in glazing and firing process.
32
2.4.2.5. Function of Quality Control
Quality is the sum of all attributes and characteristics of a product or an activity
which contributes to the usability of these to perform a specified function (Rembold,
1985). It is clear that quality control is a regulatory process through which actual quality
performance is measured, compared with standards and if necessary corrective action is
taken.
The backbone of quality control system is the company’s internal, national and
international’s standards, which become a contractual object between the manufacturer
and customer. For detail information see Figure 2.8
Reference Input
Specification ,
Development and
Manufacturing
,
Records ,
Standards
Error
+
Controller
Manipulated
Variables
Actuator
manufacturing
Planning and
Scheduling
Plant Control
Manufacturing
-
Quality Control Acquisition
Measuring systems
,
Measured Quality
Quality Control
Figure 2.8 Principle of Integrated Quality control.
Method
Material
Man
Machine
Disturbances
Customer
33
2.5. Benchmarking Study
There are some research and projects that already using the diagnostics system to
solve many problems in manufacturing process. This system’s main purpose is to detect
the defect of the problems occurred in processing and aim to solve the problems. Some
of the researches and projects are:
2.5.1
Integrated Diagnostic system for production and service
In order to properly diagnose a defect in any manufacturing firms, as the starting
point is the raw materials inspection result, going through all manufacturing steps, to the
final inspection results and customer claims. To gain the customer satisfaction and
increase the product quality, the zero defects must be produced in all manufacturing
steps. Therefore, a study of this research is to develop the system that is able to diagnose
the probable cause(s) of the defect by tracing the quality and production information at
the various steps of tire manufacturing processes.
The ways to identify defect root cause(s) is essential in reducing scraps amounts
as well as defining the corrective and preventive actions. This process requires input
from many disciplines. It is always practiced through brainstorming session and
interpreted using cause-and-defect diagrams (Ali-Khamis, 2003). It is really helpful
when the author performs the knowledge base, through a number of interviews, by
collecting and articulating related knowledge.
Building an integrated diagnose expert system (DES) in manufacturing firms,
this research divides it into two distinct sub systems: a DES and a database management
system.
34
The basic components of DES are: knowledge base, an inference engine and
user interface (Wolfgram,Dear&Galbraith, 1987).
Figure. 2.9 Architecture of Integrated DES
This study follows the DES steps when developing a system. Acquiring and
formalizing the related knowledge, design and development of different knowledge
bases using the particular technique provided by the development tool and the last step is
testing the knowledge base for validation.
During the knowledge rule processing, this study uses two methods of inference
engines, the backward and forward chaining. The backward chaining is appropriate for
diagnostics reasoning.
2.5.2
Knowledge-based expert system in manufacturing
The objective of modern manufacturing is to have efficient control over the
organizational facilities in order to produce high quality products at lower prices within
shorter time period. To achieve better quality at a lesser price, every attention has to be
paid in the entire manufacturing division, employing better tools and high precision
35
machines (Cakir-Cavdar, 2005). From this study, there are certain types of wear
mechanisms on different types of problems. The ways to analyze a problem are by
defining precisely and determining well the possible causes. The Information that is
gotten from the experts must be maintained well.
This study explains the advantages of the expert systems and how they come to
their solutions to solve the problems, set the rules, and as the strong benefit of the expert
system is being able to accumulate the knowledge of several widely separated experts in
one place.
This study represents that the sources of knowledge base are from two main
resources, first from human experts working in the field of manufacturing and second
from technical documents, catalogues or handbooks of their tool producing companies.
2.5.3
An On-Line Diagnostic Expert System for Intelligent Manufacturing
The operator’s decision can be very critical in both an economical and preventive
point of view, because the decision comes from the preferences and experiences.
In order to guarantee the decision based on the similar information, this study
seeks a way of using a computer to mimic human reasoning by using the expert system
technology to analyze experts' reasoning under a certain circumstance and implement
this knowledge to the computer in a form of rules, data base, etc.
Sensor validation plays a vital role in the ability of the overall system to correctly
determine the state of a system monitored by imperfect sensors (Kim, 1995). This study
divides the system into two subsystems which are algorithmic (ASV) and heuristic
(HSV) sensor validation. Uncertain information in sensory values is represented through
probability assignments on three discrete states, "high," "normal," and "low," and
additional sensor confidence measures in ASV. HSV exploits deeper knowledge about
parameter interaction within the system to cull sensor faults from the data stream.
36
The importance of sensor validation performs and the reasons under uncertainty
in expert systems evoke this study to purpose a framework in the interrelated functions
diagnostics.
Finally, the framework produce validated data as input to the reasoning
scheme and the influence diagram as the output which represents the backbone of
reasoning under uncertainty in the knowledge base.
This Study shows the relationship between symptoms, sensor behaviors and the
defect causes in a causal direction to root the probable cause(s). The output of the
influence diagram is a diagnostic mapping from the symptoms or sensor readings to a
determination of likely failure modes. Once likely failure modes are identified, a
detailed diagnostic knowledge base suggests corrective actions to improve performance.
This framework provides sensor validation and reasoning under uncertainty applies in a
milling machine process diagnosis. This framework has implications for how to build
the relationship between the symptom and cause of failure in the knowledge base even
thought the input comes from sensor or manual.
Figure 2.10 Simplified frameworks in influence diagram
Knowledge base
This Influence diagrams have proven successful in complex decision making
problems with uncertainty by graphically representing the diagnostic problem domain
37
through simple topological symbols and arcs between them (Pearl, 1988). Knowledge
engineering schemes (Moore, 1985), allow exploiting both first principle knowledge of a
system along with subjective assessments based on experiential knowledge and Bayes'
Theorem is the backbone of the influence diagram inference procedure.
2.6.
Summary
In this chapter, we have identified the literature review that support the DSMPP
system, the brief definitions and understanding of knowledge base, expert system, key
components of expert system, rules base reasoning method, searching approaches to root
the rule base and any information that are related to this project. To achieve objectives
and understand the methods that used for diagnosing the defect problems in
manufacturing process, benchmarking study purpose the principle concept to solve the
problems. Hopefully by getting and studying the literature review will serve as guide to
develop DSMPP system prototype.
CHAPTER 3
METHODOLOGY
3.1.
Introduction
Based on the discussion in chapter 2, the methodology used in developing
Diagnostics system for Manufacturing Process Problem (DSMPP) is knowledge base.
The rule base procedure approaches forward chaining concept. For the searching
strategy in rule procedure, the system uses DSF Search concept. To develop the
prototype of DSMPP system, the Object Oriented approach has been chosen. The
System Development Life Cycle (SDLC) has been used as a guide in developing the
system. The development process must follow the SDLC.
3.2.
System Framework
To develop the prototype of DSMPP system and base on the discussion in
chapter 2 as the concept, in chapter 3 we produce system framework that shows the
flows of activities to build the system.
39
It’s components as depicted in the framework (see Figure 3.1) with regard to:
·
Project Methodology and
·
System Development
Figure 3.1. System Framework
40
3.3.
Project Methodology
The project development life cycle (SDLC) (Dennis et al, 2005) is central
development of an efficient information system. The project development life cycle
consists of four main stages or phase:
Phase 1: Planning Phase
Phase 2: Analysis Phase
Phase 3: Design Phase
Phase 4: Implementation and Operation.
3.2.1 Planning Phase
The first phase of the systems development life cycle is system planning phase.
The planning phase is the fundamental process of understanding the why an information
system should be built and determining how the project will go about it. This phase also
deals with process of identifying, selecting, initiating, and planning of system at which
an organization creates and assesses the original goals and expectation for a new system.
41
There are two primary activities in planning phase;
3.2.1.1 Project Initiation
During this activity, we will identify and select projects, discussion appropriate
projects and literature. As the output of this activity is system request that presents a
brief summary of business needs, and explain how the system that support the need will
create the business value in manufacturing firms, the other output is feasibility analysis
which examines the key aspect of the proposed project.
3.2.1.2 Project Management
Once the project is approved, it enters the project management. During this
activity we create a work plan and techniques to help and control the project through the
entire SDLC. As the output of this activity is Gantt chart (Refers to Figure 3.2)
3.2.2 Analysis Phase
System analysis is the part of the system development life cycle in which will
determine who will use the system, what the system will do, and where and when it will
be used. During this phase the activities involved are investigating the current system,
42
identifies improvement opportunities, and develops a concept for the new system. This
phase has three steps:
3.2.2.1 Analysis Strategy
An Analysis strategy is developed to guide the project, usually includes an
analysis of the current system (called as-is system) and its problem, and the ways to
design the new system (called the to-be system). As the output of this phase are problem
identification and feasibility determination.
3.2.2.2 Requirements Gathering
The analysis of the information in conjunction with input from many other
people, leads the development of the concept for a new system, the output of this steps is
analysis models that describes how the business will operate if the new system will
develop and the models represent the data and process necessary to support the
underlying business process.
3.2.2.3 System Proposal
In this phase combined the analysis, system concept, and models into a
document.
43
3.2.3 Design Phase
The Design Phase decides how the system will operate, in terms of hardware,
software and network infrastructure, the user interface, forms and reports, and the
specific programs, databases, and files that will be needed. This is a phase, where we
will design the new system according to the result from analysis phase and determine
exactly how the system will operate. The system design phase consists of four main
activities;
3.2.3.1 Design Strategy
This clarifies how the system will be developed and resources that support the
system development.
3.2.3.2 Architecture Design
This step describes the hardware, software and network infrastructure that will be
used. In this step including the interface design which specifies how the user will move
through the system, the forms and reports that the system will be used.
44
3.2.3.3 Designing Database and file specification
These define exactly what data will be stored and where they will be stored.
3.2.3.4 The Program Design
During this activity we will define the program that need to be written and
exactly what each program will do.
3.2.4 Implementation Phase
The implementation phase of the system development life cycle is the most
expensive and time-consuming phase because all the work that has to be completed
through the entire life of the system. This phase has three steps:
3.2.4.1.System Construction
The system is built and tested to ensure it perform as designed. Since the cost of
bugs can be immense, testing is one of the most critical steps in implementation.
45
3.2.4.2.Installations
Installations is the process by which the old system is turned off and the new
system one is turned on, it might include a direct cutover approach, a parallel conversion
approach or
phase convention strategy. One of the most importance aspects of
conversion is the development of training plan to teach the users how to use the new
system.
3.2.4.3.Support Plan
This plan usually includes a formal or informal post implementation review, as
well as systematic way for identifying major and minor changes needed for the system.
46
Table 3.1: Activities in Project Development Phase
Project
Activities
Activities that have done in DSMPP
Development
Phase
Planning Phase
1. Project Initiation
1. List down all the project title that
i. Identify and select projects. Make
system request and the system that
support the business value,
ii. Produce
feasibility
analysis
2. Discuss with supervisor and choose
an appropriate project title.
which
examines the key aspect of the proposed
project.
2. Project Management
i. Produce a work plan, techniques and
schedule to control SDLC.
has been suggested.
3. Project DSMPP has been chosen.
4. Identify the background problem of
Company involved (Ceramic Horse)
5. Determine project scope, objective
and importance.
6. Determine
business
the
business
requirements,
need,
business
value, special issue and constraints
documented in system request
7. Produce system request
8. Produce the feasibility analysis
9. Produce the work plan to schedule
the project using Gantt Chart (see
47
Figure 3.2)
Analysis Phase
1. Analysis Strategy
1. Interview the staffs and officers in
different departments (Research and
i.
Analyze the current system (called as- is
system).
ii.
Identify the problem and determine the
feasibility
2. Requirements Gathering
Development Department, Quality
Control
Assurance
Department,
Quality
Department)
in
the
company about the diagnostics and
identify the defect problems.
2. Choose the experts.
i.
ii.
iii.
iv.
Gather information on what the system
3. Identify the inspection procedure
should do from as many sources as
hold by quality control and research
possible. (Users of the current system,
development (Refers to Appendix
reports, forms and procedures.)
table 2.1. Appendix A, The White
Produce the knowledge acquisition
Horse Ceramics Quality Control
from interview, case studies, protocol
Plan) for glazing processing and
analysis, brainstorming, etc.
Firing.
Produce analysis models and process
4. Identify the defects problems on the
necessary to support the underlying
ceramics
business process using UML.
Appendix A, table 3.2 Form Target
Produce the knowledge representation.
and Tiles Problem).
5. Identify
industries
the
(refers t o
classification
of
48
3. System Proposal
production in grade A, B and C.
6. Identify and analyze the symptom of
i.
Document the analysis, system concept,
and models.
the tiles problems regarding to the
information get from the experts.
7. Identify and study the tiles defect
problems and all the specification of
it.
8. Study how the experts conclude and
find the defects. (refers to Appendix
B, Figure. 3.3 )
9. Transform analysis data and models
in Object Oriented method using
UML.
10. Produce Use Case Diagram, Class
Diagram, Sequence Diagram and
Activity Diagram by using UML.
11. Produce knowledge representation
by developing the rules base on the
facts.
12. Build the reasoning in rule base
system using forward chaining.
13. Produce
the
three
or semantic
49
network using DSF Search to find
the search strategies.
14. Identify and study the user interface
design to develop the DSMPP
system.
Design Phase
1. Design Strategy
1.
Identify and learn the hardware and
software
2. Architecture Design
i.
Design the hardware, software.
ii.
Design the interface
iii.
Document the forms and reports that the
system will be use.
needed
to
design
the
DSMPP system
2. Design the questions for the user that
describes the root cause of the defect
product in DSMPP system
3. Design the interface of the DSMPP
system by using macromedia as tools
3. Design Database and file specification
and design base on the Object
Oriented Methodology
Produce the databases.
4. The Program Design
4. Design the database for DSMPP
system by using MySQL Server to
represent the data.
Design the entire program of the system
5. Report or document the design and
dialogue use in the DSMPP system.
50
Implementation
1. System Construction
Phase
1. Implement the analysis and designs
by using PHP programming.
The system is built and tested to ensure it
perform as designed.
Verify and Validate the system with the experts.
2. Discuss and test the result of
diagnostic process with the experts.
3. Discuss
and
test
the
recommendations that have provided
2. Installations
by the system.
4. Discuss and test the entire result of
Do some installation, document and training to
support the system.
3. Support Plan
system to the experts.
5. Discuss the effectiveness test
6. Install the prototype of DSMPP
system.
Formal or informal post implementation
7. Document the implementation steps
review, as well as systematic way for
8. Produce questionnaire to get the
identifying major and minor changes needed
feedback according to the system
for the system and doing maintenance.
prototype.
9. Do some training for the user of
DSMPP system (QA,QC and R&D
Personnel)
51
3.4.
System Development Methodology
Knowledge base was used to develop a prototype of Diagnostic System for
Manufacturing Process Problems (DSMPP). This development tools uses an Object
Oriented knowledge representation formalism as well as production rules. It also
features an inference engine of the order capable of forward chaining. This mode was
used in present the prototype in order to solve the manufacturing process problems in a
deductive way. Furthermore the ‘DSF’ searching option was preferred over the depth
first in order to diagnose numbers of problems before chaining stops.
Prototyping is known as designing and building a scaled-down but workable
version of desired system. Figure 3.1 illustrated prototyping. The first step in
prototyping is to determine the initial or basic requirements of the systems. After the
problem is identified, the system prototype will be build. When the prototype is
complete, the users will work with it and will come out with their opinion on the system.
According to this feedback, the prototype will be improved and the new version will be
given to the users again. This iterative process continues until the users are relatively
satisfied with the system (see Appendix B, Figure.3.4)
Figure 3.4: The Prototyping Method
3.5.
Hardware and Software Requirements
Some hardware and software are required to help us in developing the DSMPP
efficiently, systematically, and effectively. Table 3.3 shows the software that needed to
develop the system and the purpose or function of it.
52
Table 3.3: Software required to developing the system.
Software
Purpose
1. Microsoft Project 2002
Microsoft project used to generate Gantt and
Pert chart that used as a tool to schedule the
project development.
2. Rational Rose 2002
Rational Rose is software for UML modeling.
This software will be used as a helping tool to
model the system.
Example: Use case diagram, Class Diagram,
Sequence Diagram and etc.
3. Microsoft Office Visio 2003
Microsoft Visio is used to draw diagrams.
4. Macromedia Dream weaver MX This software is used to design, build and
2004
develop the system. This software provides a
combination of visual layout tools, application
development features, and code editing support.
5. Macromedia Flash MX 2004
This software is used to create some animation
features that will be included in system.
6. Adobe Photoshop 7.0
This software used to create and edit images.
7. PHP Programming
This is a programming language that will be
used to program and develop the system.
8. MySQL Server
SQL server will be used to develop the
database for the system.
9. Internet Explorer
Used to view the WebPages.
53
Hardware that are needed to develop the system:
1. Computer
Processor
: Intel Pentium 3 or Intel Pentium 4 MAinboard,
Clock Speed 1.80GHz, 256MB DDR 400 RAM, 80GB 7200rpm Harddisk, 52x
CDROM Drive and 15” Monitor
2. Printer
3. Scanner
3.6.
Project Schedule
The project schedule, as shown in Figure 3.2 Gantt chart, scheduled all the
activities that will be done to develop the project according to the development phase.
The project schedule covers the period of 32 weeks. The studies started in the first week
of July, 2006 with preliminary search and identify the project that will be proposed. The
development process hoped to be end on second week of February 2006 in
Implementation and Testing phase.
3.7.
Summary
In this chapter we have identified the project development methodology and
methods or approaches to develop the system. The flow of development activities
through the SDLC has been highlighted. The project schedule is also has been elucidated
in this chapter. The chapter laid down the methodology on how this project will be
conducted and how to systematically pursue and achieve the research objectives.
54
We hope that the identified methodology will serve as a guide to develop intangible
information asset valuation system prototype.
CHAPTER 4
ANALYSIS AND DESIGN
4.1.
Introduction
This chapter discusses initial findings, organization analysis, as- is process, to-be
process and data model, process design for the proposed system. The initial findings and
analysis that have been developed, regard to the discussion in the previous chapter. The
discussion was used to come out with the solutions to the problems in White Horse
Ceramics.
4.2. Organization Analysis
In 1992, three parties from Taiwan, Malaysia and Singapore joined forces to
create a new company (White Horse Ceramic Industries Sdn Bhd) that would set a
standard for the region's ceramic tile manufacturing industry. The company was
established in Pasir Gudang, Malaysia. The vision of the company is
“To be a world-class manufacturer producing premium-grade building materials that
will uplift the construction industry”.
58
Propelled by the goal, the company consciously put in the best and latest
technology in every area of operation, from management to staff training and
development. They emphasize on employee involvement and teamwork in the
organization. The White Horse Ceramic offers exquisite tile layout design services to
further
enhance
customer
satisfaction,
adhering
to
the
service
motto:
”Providing the Best to Our Customer".
To achieve the company’s vision, and provide the best to their customer, the
White Ceramics Horse increases the product quality and does some innovations which
involve the combination of technical expertise and creative excellence. The product
quality with zero defects as measures has brought the company into being the world
market leaders in tiles industries. The products quality is being responsible to overall
employee and department. Therefore the knowledge about defect products must be
shared and maintained well. Sometimes, the defect problems were caused by the
machine problems, fault operations, the component standards problems (not accurate) or
even the human factor errors. That’s why by rooting the probable causes will reduce the
defect problems and maintain the operational production. It influences and contributes to
the management system in the company.
The White Horse Ceramic produces several types of products:
·
Ceramic Floor Tiles
·
Ceramic Wall Tiles
·
Homogenous Tiles
They market their products to Australia, Austria, Brunei, Cambodia, Fiji, Hong
Kong, Indonesia, Italy, Japan, Maldives, Myanmar, New Zealand, Philippines,
Singapore, Sri Lanka, Taiwan, Thailand, U.S.A. and Vietnam.
59
4.3.
As Is Process
White Horse Ceramic divides ceramic tiles into two categories regarding to the
processing:
·
Dust Press Ceramic Floor Tiles
Dust Press Ceramic floor tiles is a thin ceramics made from clay which changes
continuously, coloring and comes from natured. It is used to cover the Floor,
wall or front of building, it is provided through grinder, filter, mixing, muddy
and so on which is formed by pressing, bumping, printing and other processing.
This process is conducted at room temperature. It is followed by drying and
firing process in the oven at high temperature. Ceramic tiles are maybe had
glaze, homogenously and not homogenously. It is not burnt or fired and did not
give any defects through lighting.
·
Glazed Ceramic Tiles for Internal Wall
Glazed Ceramic Tiles for Internal Wall is a unit used to form rectangles or oval.
It is provided by combination of clay, silicate or other processes, at least one
firing process at high temperature and one glazing layer.
4.3.1. Production Process Operation
4.3.1.1.Flowchart Operation Process
To identify the defects problems in process production, the flow of activities in
operation process must be understood well. It helps to find and root the probable causes
60
of each defect. Starting from the raw material, mixing process, spray dryer, press
machine, glazing and firing until packaging process (refers to Figure 4.1).
Raw material
Crusher and Grinder
mixer
/
Wastewater
Distributor
Spray Dryer
Wastewater
Silo for Storage of
Process Powder
Press mechine
Accessory Section
Homogeneous /ceramic
tiles for glazing
Unglazed homogenous
ceramic tiles
Rotor Color glazing
machine
Glazing
/
Mobile Storage
container
Wastewater
Mobile Storage
containers
Roller Kiln
Trolley Storage
Container
Roller Kiln
Storage and packing
Roller Kiln
Storage and packing
Printing Colors
Wastewater
Storage and packing
Figure.4.1 Production Process Flowchart
White Horse Ceramic uses raw material which exports clay from Hong Kong,
white clay, kaolin clay, feldspar, profilit glazed porcelain stone, feldspar diopsid, talcum
stone and balling clay. The overall raw materials will be mixed to produce mud in ball
61
milling section. In muddy processing, they use crusher grinder machines which has
processing time of around 8 hours and speeds around 11000-15000 Rpm.
The muddy clay that has been produced will be examined by Quality Assistance
Department for inspecting the quality of mud regarding to the specification standard
(ISO), if mud quality does not fulfill the standard, mud must be remixed in ball milling.
In the next step, mud will be saved in silos by using hydraulics pump and will be
continued by the Spray Dryer section. The Spray Dryer section produces 8 tons
dusts/hour. The dusts that have been produced will be examined according to the
specification standard.
The Dusts which is saved in silos container, will be prepared for Mould Press
section. Mould Press Section uses conveyer which is operated by computer
programmed. Mould Press provides guidance for tiles bases on the size and models
requirements.
The tiles that have been produced are still fragile, it will be sent to the Vertical
Dryer machine for pre drying process. The drying process takes time for 35 until 45
second bases on the models requirements.
For the next step, tiles will be sent to the glazing section for the tiles glazing
process and will be given some decoration then the tiles will be saved in mobile storage.
By using the box it will be sent to the Kiln roller for firing processes, it takes time 40
until 45 seconds.
At least the tiles will be divided by grade specification: A, B or C which will be
destroyed or throw away.
62
4.3.2. Detection the Tiles Defects
The Defect of tiles is defined as any strange or abnormal substances existing to
tiles compared to ideal or standard one.
·
The surface defects
The abnormal substances could be seen by naked eyes. The examples are the surface of
tiles is rough to the touch, the body’s surface is bent, the surface appears no glaze when
the light comes , appears some black spot on the surface, the body’s performance
failures, the surface has chipped, appears some small holes and so on.
·
The structures defects
The Inspection will be examined by naked eyes or tested using the laboratory
instruments.
The examples are the structure is bent, the problems found in curve in/out process, the
structure of tiles are bending to divergence and convergence, the structure’s size is not
balance, cutting process is not manageable or controlled well.
4.3.3. Production Classification
By setting the rules and procedures for guiding the workers to analyze and
classify the products defects, the knowledge of grade classification is divided into:
63
4.3.3.1.Grade A
The production in grade A, sets as a perfect product based on the rules and
conditions (dimension, surface, chemical resistance, and physical). The grade A has
passed the whole inspection by QA and QC regarding the specification, such as water
absorption test, modulus of rupture, sizing test, thickness test, curvature test, straightness
of sides, polishing gloss test and rectangularity test. The information was found, must
fulfill the specification standard.
Inspection of the surface state in grade A, must be in a perfect manner and no
defects are found from the cracks, the short glazing, the surface’s pin hole, the surface
has no sunny glaze , the surface’s black spot , the nipped edges, nipped corner, screen
printing and powder. All the products regard standard sample.
·
The Dimension State
The Dimension of tiles in grade A has to be in a perfect manner and no defect is
found in dimensions, thickness, the straightness of side (curve in/out), rectangularity,
center curvature, edge curvature and warp age.
·
Physical and Chemical Requirement
The Production in grade A has passed physical test and the chemical test. Examples
of physical test are water absorption, modulus of rupture, scratch hardness of surface and
abrasion resistance. The resistance of chemical test is resistance of dirtiness, chemical
substances used at home, a cleaner of swimming pool and the resistance from acid and
alkali.
64
4.3.3.2. Grade B
The Production in grade B classification is a product with little defect problems,
the overall product has fulfilled the specification standard but little defect found.
To pass for grade B, a tile has to follow water absorption test, modulus of rupture
test, sizing test, thickness test, curvature test, straightness of sides, polishing glaze test,
and rectangular test.
Inspection to the surface state in grade B must be in a perfect manner and just
little defect found from small holes, small chipped, the surface is bent, on the surface
appears little dusty or powder which doesn’t much effect on it, little black spot and
cracker glazing.
·
The Dimension State
The dimension of tiles in grade B has to be in a perfect manner with little defect
found in dimensions, size, thickness, the straightness of side (curve in/out),
rectangularity, center curvature, edge curvature and warp age.
4.3.3.3.Grade C
The Production in grade C classification is a product which has not fulfilled the
specification standard and a lot of defects found.
The production in grade C does not pass the surface inspection particularly. The
examples of defects are found in the colorings, screen printing, glaze and powder
process.
65
The Dimension state of tiles in grade C has not be in a perfect state, many defects
found in dimensions, size, thickness, the straightness of side (curve in/out),
rectangularity, center curvature, edge curvature and warp age.
4.3.4. The Tiles Inspection
·
Visual Inspection (refers to Appendix B Figure 3.3 for visual inspection)
-
The Inspection is starting from the surface, front and back of tiles, example:
edge crack, crack, dusty tiles and the basic tiles.
-
The visual inspection is used in order to obtain the dimension test, example:
the curve’s inspection.
-
Compare the tiles condition (coloring, glaze, the surface’s state) with the
standard one, if the sample has almost the same condition with the standard,
it will be grouped in the same quality control number, and the other hand if it
has a lot of differences, the QA or QC needs to be categorized in other
quality number.
·
Laboratory inspection (refers to Appendix B Figure 4.2 and Figure 4.3 for
examples of Laboratory inspection).
-
Distance inspection
-
Lighting inspection
-
Put the tiles into the table at 1 meter
-
Look into the tiles by naked eyes
-
Controlling the worker inspection which have not carried out by similar
person (check the human error or the neglect worker)
-
The intentional action on the surface has not to be classified as defect.
66
4.3.5. Identify the knowledge base
Controlling the whole production process in White Horse Ceramics is under the
responsibility of Quality Control Department. As shown in Appendix B Table.2.1 for
quality control plan, the inspection of this project is limited by glazing and firing
process. The reasons to decide the glazing and firing process are because on those steps
have been found the most defects problems, especially in firing process. The knowledge
of defects were obtained from the knowledge worker (experts) QA,QC and R&D. Many
interviews and case studies regarding to diagnostic defect problems, have been obtained
from the experts.
4.3.5.1.
The Tiles Defect and Probable causes
As mention in the constraint of this project, the research is limited on line A
production, and the type of tiles is ceramic tiles 30 x 30. Information of the knowledge
defects has been brainstormed and analyzed as sources of knowledge base. The first
stage of the knowledge base was shown in Table 4.1 for the tiles defect and probable
causes, symptom, recommendation that can be taken to solve the problems and also the
process and sub process that shows the defects or the problems.
In the proposed system (DSMPP), the output does not just constraint on the
defect problems and probable causes, the recommendation also proposed to solve the
problems regarded to the diagnostic result. As the example is the defect in firing process
“water absorption”, the recommendation to solve the problems is “the product must be
adjusted, if the size fulfills the standard, re- fired the tiles, but if the size does not fulfill
the standard, the tiles must be mould adjusted”.
67
For the detail information about the tiles defect and probable causes can be
referred from the Table 4.1 at appendix d.
4.3.5.2.The Tiles Defect and Classification
Besides the recommendation proposed the solving problems, it also classifies the
tiles production when it can not be adjusted any more. From the classification, we know
the quality of the tiles, and how to maintain it. The quantity and quality in grade A will
influence the profits and market trading in this company. The tiles defect and
classification in White Horse Ceramics can be referred to Table 4.1. at appendix d.
The condition and symptom of defects found can be shown in Figure 4.4, Figure
4.5, Figure 4.6 and Figure 4.7 as examples of defects.
Figure 4.4 Examples of defects-1
68
Figure 4.5 Examples of defects-2
Figure 4.6 Examples of defects-3
69
Figure 4.7 Examples of defects-4
These figures describe the types, physical condition and characteristics defect
found in production process, for examples are crack, mould, pin hole, dip sink, glazing
problems, tiles dust, dirty dust etc.
4.3.6. Knowledge Representation
The knowledge representation has been build to represent the rules. There are
two main parts of the rule base, “the premise” and “the conclusion”. Premise consists of
the conditional “IF” and logical expression with one or more objects attribute value
connected by logical operators “AND” and “OR”. The conclusion is composed of the
adverb “THEN” and logical expression.
70
4.3.6.1. Rule base
The rule base step in DSMPP system is implemented by the forward chaining
procedure for searching strategies in the problem solving process and can be described
from Figure 4.8
Figure 4.8 The rule base process
In DSMPP system, the rule base has produced by starting to diagnose the defects
of the tiles. The tiles defects can be concluded from the symptom of the tiles defects.
After diagnosing the defects, the rule base will investigate related the sub process and
process which contains the probable causes of defects. The final result for the
diagnostics is the output that will make some conclusion of the defects, source problems,
grade classification and solution or recommendation as the corrective action for the
defects found.
71
The semantic network for that process can be described from the diagram in
Figure 4.9
Figure 4.9. The semantic Network
This semantic network represents the searching network processing for the
diagnostic system. The searching processes are starting by examining the symptoms and
causes of defects where many symptoms and causes have many defects and many
defects have many symptoms and causes. Then, the defect found will be examined to
check the sub process and process where the defect can be found. The sub process may
have many defects. And the process also have many sub process and many defects.
Developing the rule base in DSMPP system, can be starting with this rule base
pattern as described below:
Rule base pattern :
R1
: IF (The symptom-1)=” A ” AND (symptom-2) = “ B “ THEN causes1
R2
: IF causes1 = “a”
AND (symptom-3)=”C” OR (symptom-4)=”D” THEN
causes2
R3……….
RN
RN
: IF causes2 =”b” AND (symptom-5)=”E” THEN defects=”I” AND sub process
= “I” AND process= “I” AND rec =”1” AND grade=”1”
72
This rule base pattern is used to build the overall rules base in the diagnostic
system. It is helping in searching process development (implementing DSF) to build the
engines machines. The rule base consists of several components in the knowledge base,
such as symptom, causes and for the result consists of defect, process, sub process,
recommendation and grade.
For the example of the rule base from the DSMPP system are can be listed bellows:
The defects: Pin Hole (I)
R1
: IF The defects are on the surface of tiles (37) AND (There is small hole on the
tiles surface, commonly on the edge of tile (5) OR the small hole in the graphics
design of tiles (it will appear clearly before firing process) (6)) THEN Check
temperature (b).
R2
: IF Check temperature (b) AND Temperature
80 – 100 degree Celsius (3)
AND Check machine speed (not base on the specification) (4) THEN check the
water spraying (a)
R3
: IF The defects are on the surface of tiles (37) AND the small hole in the
graphics design of tiles (it will appear clearly before firing process) (6) THEN
check the firing inspection report (d)
R4
: IF check the firing inspection report (7) AND any defects were found in firing
inspection report (136) THEN Check the water spraying (a)
R5
: IF Check the water spraying (a) AND Water spraying weight less than spec (1)
AND Nozzle is clogged up 99) AND Picture ( 8) THEN Pin Hole [ I ] AND Sub
Process = Water Spraying (1) in Process =Glazing (1)
The defects: Dipsink (II)
R6
: IF The defects are on the surface of tiles (37) AND defects were found in the
glaze of tiles (83) THEN Check Glaze level (e) OR Check viscosity of glaze
(g) OR Check weight of glaze (f) OR Check Density of glaze (h)
73
R7
: IF Check Glaze level (f) AND Glaze level less than spec (15) AND there are
some problems on the curvature reports THEN Dip sink [II] AND Sub Process is
Water Spraying (1) AND Process glazing (1)
R8
: IF Check viscosity of glaze (g) AND (Viscosity less than specification (9)
OR Viscosity more than spec (10)) ) AND There are some problems on the
curvature reports THEN Dip sink [II] AND Sub Process is Water Spraying (1)
AND Process= glazing (1)
R9
: IF check weight of glaze (f) AND Glaze weight less than spec (12) OR Glaze
weight more than spec (14) ) AND There are some problems on the curvature
reports THEN Dip sink [II] AND Sub Process is Water Spraying (1) AND
Process=glazing (1)
R10
: IF Check Density of glaze (h) AND Density less than spec (11) OR Density
more than spec (13) ) AND There are some problems on the curvature reports
THEN Dip sinks [II] AND Sub Process is Water Spraying (1) AND Process=
glazing (1)
R11
: IF the defects were found behind of the tiles (42) AND there are number of
windy holes. (40) AND it influences the problems of windy holes at the back of
tiles. (41) ) THEN check aluminums slip condition (m)
R12
: IF check aluminums slip condition (m) AND No Aluminums slip.(39)
THEN Dip sink [II] AND Sub Process is Aluminum Slip (3) AND Process=
Glazing (1)
The defects: Dirty Dust (IV)
R16
: IF The defects are on the surface of tiles (37) AND There is some dust on the
tiles surface (19) AND the dust didn’t impact much on the tiles surface (48)
THEN check the sieve condition (i)
R17
: IF Check the sieve condition (i) AND The sieve has been broken (16) AND
Mesh is worn out (17) THEN Dirty Dust [ IV ] AND Sub Process is Sieve(2)
AND Process= glazing (1)
For the detail rule base and normalization, see appendix g.
74
4.3.6.2. Decision Three
To support the rule base process that has been produced, DSF in searching
process procedure for the rule base is implemented. It can be explained by this decision
three diagram in the Figure 4.10
75
37
19
42
5
6
20
4
16
17
e
136
3
i
40
d
b
21
83
g
15
h
9
1
3
99
1
23
24
8
25
4
III
I
II
2
1
1
1
Symbol for symptoms
1
1
Symbol for defect
Symbol for Sub Process
Symbol for Process
Figure 4.10 Decisions Three Diagram
11
41
13
12
14
m
39
II
1
33
10
200
a
j
f
76
The decision tree diagram shows the DFS searching process in the engines
machines which implements the rule base. It was starting from the root of the three (set
root=0) than it was following by the common symptom (37), than checks the entire
symptom on the left sides, if it fulfills the rule base requirements, continues to the right
sides. It follows the decision tree diagram until finds the defect. As the result, the defect,
the sub process and the process will be found.
4.3.6.3. The User Interface
The user interface has been built as the explanation facilities form for the user to
make sure or to test that the rule base runs properly. The user interface receives the
inference engine process that will produce the conclusion and gives the recommended
action for the defects found. For that the user interface has been built in terms of
questions and symptom entering that the user must answer and choose correctly.
77
The detail user interface process can be explained from Figure 4.11:
Figure 4.11. The structure of DSMPP system
This structure explains how the DSMPP system works. The DSMPP system can
be divided by two environments. One is consulting environment. The other is knowledge
development environment. Starting from the knowledge development environment, the
knowledge bases were built and analyzed in this part. The knowledge was absorbed
from knowledge engineers in the contexts of symptom, causes, defect, sub process,
recommendation, grade and any facts related to the tiles defect. Therefore, the
knowledge acquisition and the knowledge managements are taking parts on this
environment. Admin acts as operator to entry the knowledge base. All the activities in
knowledge development will support the inference engines developments. To test the
inference engines in diagnostic process, the consultation environment has been built in
terms of symptom entering and question forms.
78
The types of questions and optional answers produced by the system are based
on the expert’s information about the rules or route to diagnose the product defect such
as describes in Figure 4.10 (Decision Tree Diagram). Accordingly, the analysis step
process will be described from this diagram below:
Figure 4.12. Dialogues route analysis process
The route analysis process in the consultation environments were starting from
the location of defect, such as in the tile surface, tile sides, tile back and all location.
After finishing the location, check the condition, in terms of color and machines
condition (speed, temperature and component/equipment). Finally, the system will check
the report, such as the pressing report, glazing report, printing report and curvature
report etc.
79
The examples of dialogues can be listed below. The user must declare and
choose their answer based on the physical defects found to gain the correct conclusion
for the diagnostics process.
1.
Where is the location of the defects find?
0 The defects are on the surface of tiles
0 When it is looked in each sides
0 The defects are found on the back of tiles
0 $OO
0 Not on the optional
2.
How does the defects on the tiles surface look like ?
0 There is small hole on the tiles surface, commonly on the edge of tile
0 There is small hole in the graphics design of tiles (it will appear clearly
before firing process)
0 The defects were found in the glaze of tiles
0 It seems the different color in one kinds of tiles
0 There is some dust on the tiles surface
0 There are some drop glazing on the tiles
0 There are some accumulation glazing on the tiles surface
0 There are number of scratches on the tiles surface
0 Glaze problems
0 The color of tiles not base on the spec
0 The surface of tiles are little bit fluctuate
0 There are some hollow on the tiles surface
0 The tiles surface seems billow up
0 It seems many chipped on the edge of tiles
0 There are some crack on the tiles surface
0 Size not base on the spec
0 The tiles seem too shinning
0 It is not base on the spec when it compare with the standard tiles
0 The tile’s design not on the spec
80
0 When it looked by naked eye for the distance around 1 meter, there is
some hump or outgrowth on the tiles surface.
0 When handy touched, the surface of tiles are covered with bumps or
bruised.
0 After the tiles washing process, there are the water shadow on the surface
of tiles in 1 m distance
3.
How is the defects condition when it looks at from each sides?
0 The tiles are not symmetric
0 The tiles are swelling up shape or wrinkled shape in the firing process
0 The curve in and out process not base on the spec
0 Size not base on the spec
0 Tiles condition is dome shape
0 Curvature has problem
0 The glaze surface fluctuates or it seems not flat.
0 The surface of tiles are little bit fluctuate
0 The tiles surface is not flat
0 There are number of windy holes
0 All
0 Not on the optional
4.
How is the machine condition?
0 Machines speeds normal
0 Machine speed not base on the specification
0 Temperature 80 – 100 degree Celsius or more than spec
0 Temperature is less than spec
0 The water spraying problems
0 The sieve problems
0 The brush problems
0 Motor belting speed not base on the spec
0 Support bracket condition has a problem
81
0 Machine belting glaze has problems
0 Machine printing problem
0 Roller problem
0 Ball milling processing has problem
0 The skin has problems
0 The skin position not base on the spec
0 The gas pressure is not base on the spec
0 The air pressure not base on the spec
0 The Aluminums slip problems
0 All
0 Not on the optional
5.
How is the sieve condition?
0 Normal
0 The sieve has been broken
0 Mesh is worn out
0 Mesh is coarse or rough to touch
0 Picture (23
0 Picture (24)
0 Picture (25)
0 All
0 Not on the optional
6.
How is the water spraying conditions?
0 Water spraying weight less than spec
0 Water spraying weight more than spec
0 The water spraying is not full enough to spray into the tiles
0 Nozzle is clogged up
0 All
0 Not on the optional
82
7.
How is the brush condition?
0 Brush is not flat
0 Condition of brush is not on the spec standard
0 The weight of brush is not base on the spec
0 Some problems are found on the brushing loss
0 All
0 Not on the optional
8.
Which kinds of investigation report that finds the problems?
0 There are some problems in press investigation report
0 There are some problems on the firing inspection report
0 There are some problems on the printing process report.
0 There are some problems on the glazing report
0 There are some problems on the roller problems report
0 There are some problems on the curvature reports
0 All
0 Not on the optional
9.
How is the condition of tiles in terms of its color?
0 The color of tiles not base on the spec
0 It seems the different color in one kinds of tiles
0 The color is not balance
0 shading problems
0 It seems blur (darken)
0 The tiles seem too shinning
0 The tiles is mate tiles
0 All
0 Not on the optional
83
10.
Any problems on the press investigation report?
0 there is some defect report on press investigation report
11.
Any problems on the firing inspection report?
0 Some defects were found in firing inspection report
0 The tiles are swelling up shape or wrinkled shape in the firing process.
0 The small holes in the graphics design of tiles
0
All
0 Not on the optional
12.
Any problems on the printing process report?
0 Machine printing problems
0 There is some verification of shade/shading problems
0
All
0 Not on the optional
13.
Any problems on the glazing report?
0 The glazing surface looks not flat
0 Glaze weight less than spec
0 Glaze weight more than spec
0 Glaze level less than spec
0 Glaze Problems
0 There are some accumulation glazing on the tiles surface
0 The accumulations of glazing are influencing the tiles performance
0 The surface of glaze is not flat and it influences the structure of tiles
0 The un- flattened glazing will influence the tile structure and performance
0 It seems the different texture of glaze
0 Machine belting glaze has problems
0 Screen problem
0 All
0 Not on the optional
84
14.
Any problems on the roller problems report?
0 There are some tiles that get caught on the roller kiln
0 Roller problem
0 The roller has broken
0 there is some problems in roller kiln
0 All
0 Not on the optional
15.
Any problems on the curvature report?
0 Viscosity less than specification
0 Viscosity more than spec
0 Density less than spec
0 Density more than spec
0 The tiles are not symmetric
0 The curve in and out process not base on the spec
0 Size not base on the spec
4.3.7. Knowledge Management
The knowledge management in the DSMPP system is used to manage the know
ledges from many experts or many knowledge engineers in that company. Every
knowledge engineer in that company has the same access to edit and manipulate the
knowledge base in the system. For that reason, every knowledge engineer can share their
knowledge and produce the best and complete knowledge base.
85
Detail information about the knowledge management in DSMPP system can be shown
in the Figure 4.13
Recommended Action
Solution
/
Knowledge Base
Facts : the symptom
, the
defects , the probable causes
and the relations
Rules : Logical inferences
Grade Classification
Knowledge Management
Knowledge Acquisition
Knolwedge
Enginner
Knolwedge
Enginner
Knolwedge
Enginner
Experts
Figure 4.13 Knowledge Management Structure
By using this system, the concept of knowledge management can be built where
every knowledge engineer can share and increase their current knowledge by modified
the knowledge base and the rule base procedures. As the result, every knowledge
engineer is responsible with their knowledge and expertise.
86
4.4.
To-Be Process
All the activities and process that are involved in diagnostic system has been
modeled using Use Case Diagram and Sequence Diagram. Data involved in each process
or activity is illustrated using Class Diagram and Sequence Diagram.
1.
Use case Diagram
Figure 4.14 shows use case diagram for Diagnostics System for Manufacturing
Process Problem (DSMPP). The admin will record all the information needed by the
system such as information of tiles, the defects in the ceramics industries, and the
symptoms that probably cause the defects and the recommendations for solving the
defect problems. The knowledge engineer will manage the rule base system, edit and
validate the rule base and do some acquisition. The User who could be the Quality
Control or Quality Assistance or even the Research and Development will do some
diagnoses to find the source of defect. The User Interface in diagnostics system uses
dialogue format to obtain the problems. It will provide the answer to defect problems
and recommendations to solve the problems or classification information of tiles (tiles in
grade A, B or C).
87
ViewCompanyInfo
<<extend>>
Diagnose
User
EditAnswer
<<include>>
ViewTheResult
ViewProduct
ViewAll
EditRulebase
ViewProcess
ViewCauses
ViewRulebase
ListAll
ViewSymptom
<<include>>
<<include>>
<<include>>
ViewSubProcess
<<include>>
Admin
Searching
<<include>>
ViewRec
LoadDefect
Logout
Login
EditRec
EditSubProcess
EditProcess
Knowledge
Engineer
EditSymptom
ViewDefects
EditProduct
ViewQuestion <<include>>
LoadQ_A
<<include>>
EditDefect
EditCauses
ViewAnswer
EditQuestion
Figure 4.14: Use Case Diagram
88
2.
Use Case Description
Use Case description is created based on use cases in use case diagram. There are
thirty use cases involve in DSMPP Figure, 4.15, 4.16, 4.17, 4.18, 4.19, 4.20, 4.21, 4.22,
4.23, 4.24, 4.25, 4.26, 4.27, 4.28, 4.29, 4.30, 4.31, 4.32, 4.33, 4.34, 4.35, 4.36, 4.37,
4.38, 4.39, 4.40, 4.41, 4.42, 4.43 and 4.44 shows the use case description for each use
case. See appendix h for the use cases description.
3.
Class Responsibility Collaboration (CRC cards)
Class-Responsibility-Collaboration (CRC) cards are created to point out the
responsibilities and collaborations of a class. Figure 4.45, 4.46, 4.47, 4.48, 4.49, 4.50,
4.51, 4.52, 4.53, 4.54, 4.55, 4.56, 4.57, 4.58, 4.59, 4.60 and 4.61 at appendix i shows
CRC cards for each class involves in class diagram.
4.
Class Diagram
Figure 4.62 shows class diagram of Diagnostic System for Manufacturing
Process Problems, it categorizes the users in three types, user, knowledge engineer and
admin. The systems also record the information of the tiles, grade, symptom or probable
causes, defects and recommendation. The main process in managing the knowledge
worker (experts) is restoring in rule base database. The rule base shows the relationship
of the defects, symptom, grade, tiles and recommendation. The Result provides
information of defect, grade classification and recommendation purposed (see Figure
4.62)
89
Admin
id
password
+1
User
EditProduct()
EditSymptom()
EditCauses()
EditProcess()
EditSubProcess()
Edit Grade()
Edit Defect()
Edit Rec()
EditAnswer()
EditQuestion()
EditRulebase()
Searching()
ListAll()
ViewProduct()
ViewSymptom()
ViewDefect()
VeiwCauses()
ViewProcess()
ViewSubProcess()
ViiewAnswer()
ViewQuestion()
ViewRec()
id
password
+1
+1 +1
+1
Users
Load diagnose()
Load Result()
ViewAll()
ViewProcess()
ViewSymptom()
ViewCauses()
ViewSubProcess()
ViewRec()
ViewCompanyInfo()
Load Defect()
+1
uses
uses
mantain
+1
+1
maintain
mantain
mantain
+0..*
+0..*
Symptom
+1
+1
id_sym
Name_sym
veryfiedby
date
maintain
maintain
+0..*
Grade
+0..*
has
id_Def
Name_Def
veriyfiedby
date
+0..*
Tiles
id_Tiles
Name_Tiles
Information
Price
Image
veryfiedby
date
display
+0..*
+1..*
+0..*
id_question
name_question
veryfiedby
date
id_subproc
name_subproc
veryfiedby
date
has
+0..*
Question
mantain
SubProcess
has
+0..*
+0..*
Login()
Logout()
View()
has
+0..*
+0..*
+0..*
Websites
+1..*
has
Defect
+0..*
Proces
id_process
name_process
veryfiedby
+0..*
date
has
id_Grade
name
veryfiedby
date
+0..*
Answer
id_answer
name_answer
+1..*
veryfiedby
date
mantain
+0..*
maintain
+1
Result
caused by
mantain
Rec
id_Rec
+0..*
Name_Rec
veryfiedby
date
Causes
id_causes
name_causes
veryfiedby
date
mantain
mantain
Give Result()
Load diagnose()
+0..*
+1
mantain
+0..*
mantain
+0..*
proposed
mantain
mantain
source of
+1..*
+1..*
KE
id
password
mantain
+1..*
+1..*
+1..*
mantain
+1..*
+1..*
Edit Rule base() +1..*
EditSymptom()
+1..*
EditCauses()
+1..*
EditDefect()
EditProcess()
+1..*
EditSubProcess()
EditRec()
+1..*
EditGrade()
EditAnswer()
EditQuestion()
Searching()
+1..*
ListAll()
ViewProduct()
ViewRulebase()
ViewSymptom()
ViewCauses()
ViewDefect()
ViewSubProcess()
ViewRec()
ViewGrade()
Rule base
+0..*
id_rulebase
id
rules
generate
Manage Rule base()
inference for
Figure 4.62: Class Diagram
uses
90
5.
Sequence Diagram
Sequence diagram is developed to show the interaction that involve in DSMPP.
Appendix C illustrates how the users interact with the DSMPP to diagnose and manage
the defects problems. The Sequence Diagrams are: Login, Logout, Searching,
EditProduct,
EditSymptom,
EditSubProcess,
EditDefect,
EditRuleBase,
EditCauses,
EditQuestion,
EditRec,
EditAnswer,
EditProcess,
ViewSymptom,
ViewProduct, ViewCauses, ViewProcess, ViewSubProcess, ViewRec, ViewAnswer,
ViewQuestion,
ViewDefect,
ViewAll,
ViewTheResult,
ViewCompanyInfo,
ViewRuleBase, ViewListAll, LoadDefect, Diagnose, LoadQ_A.
6.
State-chart Diagram
The state chart diagram represents a dynamic model that shows the different
states that a single object passes through during its life in response to events, along with
its responses and actions. For the DSMPP system, the state chart diagram for class user
and class tiles have been produced. See Figure 4.63 and Figure 4.64.
enters the system
Entering
userInput
Symptom
SymptomSelected
Question and Answer
Dialogues
AnswerSelected
Diagnose
check in
Release
Conclusion
RuleBase
displayResult
Figure 4.63 State Chart Diagram: User Class
91
Figure 4.63 shows the state chart diagram for class user as the person who do
some diagnostics. When the users (QA/QC/R&D, it could be experts, knowledge
engineer, admin or common users) use this system to diagnose the defect of the
products. The system provides three types of diagnostic form, first by inserting the
physically symptom found, the second is doing some dialogues by answering the
questions that relevance to the symptom found and the third step is combination of two
types. The diagnostic processes will be conducted by checking the rule base and the
knowledge base. Finally it will release the conclusion as the result that contains the
whole information, such as the defect, the process, the sub process, the grade and the
recommendation.
symptom chosen/entered
user submits the symptom list
In Process
check and route the inference engines
diagnostics
Rule Base
check the knowledge base
The rest symptom are released
Answer
Result doesn't reach 100% checked
Question
Form
Process
display result 100% checked
display result for=100% or not
Relase
Conclusion
Give Report
Figure 4.64 State Chart Diagram: Tile Class
Figure 4.64 shows the activities or diagnostic process to detect the tile defect in
the DSMPP system. The user will insert the symptom or physical defect found in the
tile.
92
Then, the inference engines will check and process the entered symptom until it releases
the conclusion. If the result 100% bases on the rule base, the conclusion will be given,
unless the user must answer the question form. In that situation, if the question form did
not fill 100% rule base, the result and conclusion must be displayed and reported as the
final conclusion.
4.5.
System Architecture
To describe the entire system in DSMPP, the system architecture has been built
to explain as shown in this Figure 4.65.
Knowledge Engineer
Admin
1
Entry
knowledge
Knowledge
Knowledge Engineer
knowledge
Knowledge
Meta data
Server
SQL Database ,
PHP ,
Internet Technology for connection
DSMPP System
User 1/
Worker
(Consultation Environment
User 2/
Worker
)
User 3/
Worker
Figure 4.65 DSMPP System Architecture
2
93
DSMPP system is used by three categories user, the knowledge engineer as the
expert, the admin as operator and workers as common users. The knowledge engineers
entry their knowledge in the system, they can entry the knowledge base by themselves or
even by admin. By implementing the technology and certain query processing for the
knowledge base, the diagnostics system will be produced. It helps the workers to
diagnose the tile defects.
Some technologies and platform that support the DSMPP system are listed:
·
The technological architecture of the DSMPP system – tiers, communication
links, etc.
·
Hardware platforms used - server machines, disk subsystems, load balancers
and other hardware resources, etc.
·
Software platforms used (e.g. operating systems, web servers, application
servers, database servers and servers)
·
Communication equipment used – networks, routers, firewalls, switches, etc.
·
Communication protocols used- (HTTP)
94
Hardware and Software Specification to support the DSMPP system:
Hardware
Software
·
Web Browser:
Intel Pentium 3 or Intel
Pentium 4 Mainboard
·
Microsoft Internet Explorer v5.0+
·
256MB DDR 400 RAM
·
Netscape v6.2+
·
80GB 7200rpm Harddisk
·
Mozilla v1.0+
·
52x CDROM Drive
·
15” Monitor
Network Security
·
Network Packet Filter
·
Firewall
Operating System
·
Windows XP
Programming Language
·
PHP Programming
Database
·
MySQL (PhpMyAdmin)
Text Editor
·
Macromedia Dreamweaver
Table 4.2. Hardware and Software Specification
Table 4.2 is used to explain the hardware and software specification to develop the
DSMPP system.
4.6.
Physical Design
The physical design for DSMPP system contains the database design, program
structure chart, interface chart and test plan.
95
4.6.1. Database Design
The DSMPP system consists of the tables as described in Table 4.3, Table 4.4,
Table 4.5, Table 4.6, Table 4.7, Table 4.8, Table 4.9, Table 4.10, Table 4.11, Table 4.12,
Table 4.13, Table 4.14, Table 4.15, Table 4.16, Table 4.17, Table 4.18, Table 4.19,
Table 4.20, Table 4.21, Table 4.22, Table 4.23, Table 4.24 and Table 4.25. The entire
tables are saved in diagnose database and for the database design refers to appendix j.
4.6.2. Program (Structure) Chart
DSMPP system is divided by the three categories of user, first is admin as the
operator that can add, edit and manipulate the knowledge base and the user information,
the second is knowledge engineers that can add, edit and manipulate the knowledge base
and the rule base and the last user is user as a common user that can use the system to do
some diagnoses to know the source defects of the tiles.
96
The entire system structure in DSMPP can be shown from this structure chart:
DSMPP System
Login
Admin
Company Info
Diagnose
Knowledge Engineer (KE )
Product
Product
Symptom
Symptom
Reccomendation
Reccomendation
Defects
Defects
Causes
Causes
Rule base
Rule base
Process
Process
SubProcess
SubProcess
Question
Question
Answer
Answer
ListAll
ListAll
Logout
Logout
The Result :
The Defects :
Grade :
Recommendation :
The Process
The SubProcess
Figure 4.66 The Structure Chart of DSMPP System
97
The structure chart has been created by implementing the overall module in the
analysis step. As explained in the previous chapter that the DSMPP system contains two
type of environment, first is knowledge development environment and the second is the
consultation environment. In part of knowledge development environment, just the
authenticate user can access this modules whereby they can manipulate the knowledge
base or even the rule base. For the consultation environment, there is no specification
user can use or access this function. At this part, the user can diagnose the tile defect and
get the result as the recommendation action to solve the problems.
4.6.3. User Interface Design
The goal of the user interface design is to present information to users so they
can accurately understand it with the least effort. The fundamental principles for the
interface design reflect how the interfaces are used and ways to make it simpler for user
to understand them. There are many tools to describe the interface design. Here the
author uses window navigation diagrams to simulate the whole system. The window
navigation diagram of DSMPP system can be seen in Figure 4.67
98
Figure 4.67 Window Navigation Diagram for DSMPP System
99
4.7.
Summary
This chapter discusses about some introduction or company profiles of Ceramic
Horse, “as is process” as explanations of the current processes and “to be process” as
explanations of DSMPP flows step system developments. Therefore, system analysis
and designs have been developed by implementing the whole method and theories that
have been explained in chapter 2 and chapter 3. Besides, this analysis and design have
been used and implemented many tools to develop the system, for examples are UML to
object oriented system developments and PhPMyAdmin for database designed. For this
reasons, the chapter 4 can be used as a guide to implement and test the system.
CHAPTER 5
IMPLEMENTING AND TESTING
5.1.
Introduction
Basic coverage and boundary testing has been performed all the way through the
creation of code. After each class was completed, the boundaries of the loops were
checked and print statements were used to make sure that every element of the class is
exercised by at least one use case.
The main documentation has been generated during the testing of the completed
system. The test cases in the next section represent the results of black box (functional)
testing, where each use case is executed and the actual results are compared to the real
results. Each test that has been conducted is given a pass or fails status. This chapter not
really explains about the testing phase but also system implementation phase.
5.2.
System Implementation
The analysis and design that has been produced in chapter 4 was created as a
guide to develop the system in implementation section. As explained in the previous
chapter, DSMPP system has been developed by using PHP (AppServ) and Macromedia
Dreamweaver version 7. For databases development MySQL PhpMyAdmin was used.
101
5.2.1. Database Development
Databases development in DSMPP system focuses on the knowledge base
structure development and inference engines process development. Both processes
developments will form the diagnostics process to solve the defect problems. For that
reasons, relationship must be built among tables, such as: defect, symptom, causes,
process, subprocess, recommendation, grade, question and answer.
The tables
relationships can be seen in the Figure 5.1
Figure 5.1. Knowledge base relationship
As it is explained above, besides the knowledge base developments, the
inference engines machines must be produced. It involves the whole databases in the
knowledge base and includes the table rule base.
102
This table is used to record the rules base in DSMPP system. For detail
information see Figure 5.2
Figure 5.2 Rule base relationships
Every rule base that has been produced will give the information about the
defect, the process, the sub process, the grade and the recommendation as the final
conclusion. The rule bases it selves, are developed by symptom and causes components
base on the pattern that has been explained in chapter 4.
103
Rule base pattern :
R1
: IF (The symptom-1)=” A ” AND (symptom-2) = “ B “ THEN causes1
R2
: IF causes1 = “a”
AND (symptom-3)=”C” OR (symptom-4)=”D” THEN
causes2
R3……….
RN
RN
: IF causes2 =”b” AND (symptom-5)=”E” THEN defects=”I” AND sub process
= “I” AND process= “I” AND rec =”1” AND grade=”1”
The defects: Pin Hole (I)
R1
: IF The defects are on the surface of tiles (37) AND (There is small hole on the
tiles surface, commonly on the edge of tile (5) OR the small hole in the graphics
design of tiles (it will appear clearly before firing process) (6)) THEN Check
temperature (b).
R2
: IF Check temperature (b) AND Temperature
80 – 100 degree Celsius (3)
AND Check machine speed (not base on the specification) (4) THEN check the
water spraying (a)
R3
: IF The defects are on the surface of tiles (37) AND the small hole in the
graphics design of tiles (it will appear clearly before firing process) (6) THEN
check the firing inspection report (d)
R4
: IF check the firing inspection report (7) AND any defects were found in firing
inspection report (136) THEN Check the water spraying (a)
R5
: IF Check the water spraying (a) AND Water spraying weight less than spec (1)
AND Nozzle is clogged up 99) AND Picture ( 8) THEN Pin Hole [ I ] AND Sub
Process = Water Spraying (1) in Process =Glazing (1)
104
Therefore, to be easily in system developments by implementing DSF searching
process, the rule base can be normalized as described below:
Rule base for Defect: Pin Hole.
v The defects are on the surface of tiles (37) AND There is small hole on the
tiles surface, commonly on the edge of tile (5) AND Temperature
80 –
100 degree Celsius (3) AND machine speed (not base on the specification)
(4) AND Water spraying weight less than spec (1) AND Nozzle is clogged
up 99) AND Picture ( 8) THEN Pin Hole [ I ] AND Sub Process = Water
Spraying (1) in Process =Glazing (1) AND Recommendation AND Grade
v The defects are on the surface of tiles (37) AND the small hole in the
graphics design of tiles (it will appear clearly before firing process) (6)
AND any defects were found in firing inspection report (136) AND Water
spraying weight less than spec (1) AND Nozzle is clogged up 99) AND
Picture ( 8) THEN Pin Hole [ I ] AND Sub Process = Water Spraying (1)
in Process =Glazing (1) AND Recommendation AND Grade
5.2.2. System Development
Regarding to the DSMPP system structures that have been developed in chapter
4, the entire DSMPP system development can be divided into 4 main components.
(i)
Login as security function
(ii)
Knowledge Base Development
(iii)
Inference Engine Development
(iv)
User Interface Development
105
For the knowledge base development steps, there are some entities that are used
as the knowledge base basics. Therefore, the whole knowledge bases that have been
produced will be a basis for the rule base development and inference engines machines
of DSMPP system.
The knowledge base part i n system development can be used by two user
categories, first is admin as system operator and knowledge engineer that acts as the
experts. The users of this system apply the login system as security platform to
manipulate the knowledge bases, in terms of insert process, update or even delete the
knowledge base records. However, the manipulation will be recorded by using the
verified and date system.
Some of the knowledge base categories are listed below:
(i)
Product
(vi)
SubProcess
(ii)
Symptom
(vii)
Recommendation
(iii)
Defect
(viii)
Question
(iv)
Causes
(ix)
Answer
(v)
Process
(x)
Rule Base
Some functions that are involved in knowledge base operation, besides insert,
update and delete records are:
(i)
Searching
: This function is used to search any information in the
knowledge base by entering the keyword.
(ii)
ListAll
: This function is used to display the list of information on
the knowledge base, based on the knowledge base categories, such as
defect, symptom, causes, process, sub process and recommendation.
For the Inference Engines Development steps, there are some entities that are
involved in, such as the rule base system and the knowledge base. This helps the
machine inferences to produce the solution base on the main objectives of the system.
106
The main goal of this system is to diagnose the tiles defect and give some corrective
actions to solve the problems.
The User Interface development part in consultation environment has been
developed as a testing process on the diagnostic or improvement of the inference engines
process. Therefore, the user of this part or this function is free and no specialize category
user and no need the security platform. Therefore, the admin, the KE (Knowledge
Engineer/ the experts) and the common user (staff in QA, R&D and QC) can use this
function.
This functional system produces some features or facilities, such as:
(i)
Display information about t h e c a u s e s , symptoms, processes, s u b
processes, recommendations and combinations in knowledge base. This
system also provides the relationship between the defects to the entire
part of knowledge base.
(ii)
Dialogues
: On this step, the system provides two types of dialogues
form, input symptom process and question-answer form. First, the input
system process, the user inserts the symptom by choosing any kinds of
symptom found on the tiles defects. The second, the question-answer
form will be used to root the symptom and tiles defect for the extended
diagnostic to gain the 100% result.
5.2.2.1 Product
Product is part of the knowledge base category. It has been developed to provide
the information about the tiles production in the Ceramic Horse Company. Admin and
knowledge engineer have accesses to manipulate the data in this function.
107
The product page provides some information about the tiles, such as: name of
tiles, prices, type of tiles and picture.
5.2.2.2 Symptom
Symptom is part of the knowledge base category. It has been developed to
provide the information about any symptom found on the tiles defect, especially in
physical form, such as brush is not flat, the aluminums slips and the sieve has been
broken, etc. the symptom bases on the situation of tiles defect and roots of
manufacturing processes. Every symptom will be related to the defect. The query for
those system symptom-defect developments is:
SELECT *
FROM def_sym, defect
WHERE id_s = 'colname' AND def_sym.id_d = defect.id_def
The output page for symptom processes can be seen on the Figure 5.3 and Figure 5.4:
Figure 5.3 The symptom page
108
Figure 5.4 The symptom_def page
5.2.2.3 Defect
The Defect is part of the knowledge base category. It has been developed to
provide the information about any defect found on the tiles production, especially in
physical form, such as pin hole, crack, glazing peel and deep sink, etc. Every defect will
be related to the symptom, the grade, the causes, the recommendation, the process and
the sub process. The queries for those system developments are:
For defect-causes:
SELECT *
SELECT *
FROM def_sym, symptom
FROM cause_defect, causes
WHERE
id_d='$id'
AND
def_sym.id_s = symptom.id_sym
WHERE id_d = 'colname' AND
cause_defect.id_c =
causes.id_cause
For defect-sub process:
SELECT *
For defect-symptom:
109
FROM def_subproc, subprocess
SELECT *
WHERE id_d= '$id' AND
def_subproc.id_sp =
subprocess.id_subproc
FROM def_grade, grade
WHERE id_d = 'colname' AND
def_grade.id_g = grade.id_grade
For defect-process:
For defect-rec:
SELECT *
SELECT *
FROM def_proc, process
FROM rec_def, rec
WHERE
id_d='$id'
AND
def_proc.id_p AND
process.id_proces
WHERE id_d = 'colname' AND r
ec.id_rec = rec_def.id_d
For defect- grade:
The defect page can be seen on the Figure 5.5 and Figure 5.6:
Figure 5.5 The Defect Page
110
Figure 5.6 The Defect Relation Page
5.2.2.4 Causes
The Causes is part of the knowledge base category. It has been developed to
provide the information about any causes found on the tiles defect or probably causes of
defect by rooting the root in manufacturing process. The probable causes of defects are
check machine speed, check the temperature, check the sieve maintenance and etc.
Every cause will be related to the defect. The query for those system causes-defect
developments is:
SELECT *
FROM cause_defect, defect
WHERE id_c = 'colname' AND defect.id_def = cause_defect.id_d
111
The cause page can be seen on the Figure 5.7 and Figure 5.8:
Figure 5.7 The Cause Page
Figure 5.8. The Cause Relation Page
112
5.2.2.5 Process
The Process is part of the knowledge base category. It has been developed to
provide the information about any process in manufacturing process that probably
causes the defect. The processes in this DSMPP system are limited by Glazing and firing
process. Every process will be related to the defect.
5.2.2.6 SubProcess
The Sub Process is part of the knowledge base category. It has been developed to
provide the information about any sub process in manufacturing process that probably
causes the defect. The sub processes in this DSMPP system are water spraying,
aluminums slip, sieve, row bodies and brushing loss. Every process will be related to the
defect.
5.2.2.7 Recommendation
The Recommendation is part of the knowledge base category. It has been
developed to provide the information about any recommendation or corrective action to
solve the defect found. Every recommendation will be related to the defect.
113
5.2.2.8 Question
The question is part of the user interface development. It is useful to develop the
dialogues that will provide the interactive diagnostics. All the questions are following
the symptom and causes of defect found in knowledge base area. This dialogue is used
when the diagnostics process by inputting the symptom can not give 100% solution.
5.2.2.9 Answer
The answer is part of the user interface development. It is useful to develop the
dialogues that will provide the interactive diagnostics. All the answers are following the
symptom and causes of defect found in knowledge base area. This dialogue is used when
the diagnostics process by inputting the symptom, can not give 100% solution. For that
reasons the user must choose the correct answers from the whole questions given.
Admin and knowledge engineer can manipulate the answers and the question.
5.2.2.10 Rulebase
Rule base is part of inference engines development. As explained above, the rule
base was built regarding to rules of this system. That rules come from the symptom, the
causes and the defect as the solution of the rule base process. The rule bases are
implementing the DSF searching process to gain the objectives of this diagnostics.
114
The query or the snap shot code for this module is listed below:
<? $query_rule = "SELECT * FROM `rule base` AS rb, defect, subprocess, process, rec,
grade WHERE rb.id_def=defect.id_def AND rb.id_subproc=subprocess.id_subproc
AND rb.id_proces=process.id_proces
AND
rb.id_rec=rec.id_rec
AND
rb.id_grade=grade.id_grade ";
$rule = mysql_query($query_rule, $lisa) or die(mysql_error()); $row_rule =
mysql_fetch_assoc($rule); $totalRows_rule = mysql_num_rows($rule); ?>
<?do { if ($row_rule['id_rb']!="") {?>
<tr>
<td valign="top"><?php echo $row_rule['id_rb']; ?></td><td valign="top">
<? $sym = explode(",", $row_rule['id_listsym']); for ($i = 0; $i < count($sym); $i++) { if
($sym[$i]!="") { $query_symptom = "SELECT * FROM symptom WHERE
id_sym=".$sym[$i];
die(mysql_error());
$symptom
=
mysql_query($query_symptom,
$row_symptom = mysql_fetch_assoc($symptom);
(".$row_symptom["id_sym"].")
".ucfirst(strtolower($row_symptom["name_sym"]))."<br>";}} ?>
$lisa)
or
echo
"-
115
The detail processes in the rule base module is described in the Figure 5.9 below:
Figure 5.9. The Rule Base Page
5.2.2.11 Diagnose
The diagnose function are used to test the diagnostics process in DSMPP system.
By doing the diagnostic, the solution will be produced as the result of the system, such
as: the defect of tiles, the process and the sub process that probable cause of defect, the
grade classification that showing the quality of tiles and finally the recommendation as
the corrective action to solve the problems.
116
5.2.2.12 Login
Login is used as security mode for accessing this DSMPP system. Therefore, the
login users are divided into several categories user, for examples are admin and
knowledge engineer that could be KE 1 and KE 2. By using this authority, the users can
share and update the data or record in the database system. The login form can be seen
in the Figure 5.10 below.
Figure 5.10 .Login Page
5.3.
Test Result / System Evaluation
System evaluation is used by implementing many testing approaches in system
development, such as unit testing, black box testing, white box testing and usability
testing. The overall chosen testing approaches are used to test the system correctness,
and to make sure that the system is running properly regards to system objectives and
the user requirements.
117
5.3.1. Unit Testing
Unit is the smallest modules in system development. This unit testing is focusing
on the coding testing to make sure that the system is running properly. The testing
processes in implementing phase are running by inserting the data values on the
functional system until it find no error in the testing implementation. Thus, helps the
DSMPP system development. The unit testing is applying two types of testing processes,
the black box testing and white box testing.
5.3.2. Black box Testing
The black box testing is used to test the system output. This approach is more
focusing on the system output then the processes structure of the system. Whether the
system are running base on use cases analysis process and producing the correct output
regards to the user requirements.
118
The black box testing processes can be explained in the Table 5.1 below:
Table 5.1 .Black Box Testing Process
Use Case
Normal Input Values
Expected Results
Login
Admin and KE are able
(Admin , KE)
to login and manipulate
Actual Results
Pass (As expected)
the DSMPP homepage
Logout
Admin and KE are able
(Admin , KE )
to logout and back to
Pass (As expected)
main menu
View CompanyInfo
User can view the
(User)
company info, such as
Pass (As expected)
the organization info, the
founder, the factories, the
branches and the
resident.
View The Result
User can view the
(User)
diagnostic result. The
conclusion for the overall
system, about the defect,
the process, the sub
process, the grade and
the recommendation for
the defect found.
Pass (As expected)
119
Use Case
Normal Input Values
Expected Results
View All
All users can view all
(User ,Admin, KE )
information about the
Actual Results
Pass (As expected)
process, the sub process,
the symptom, the causes
and the recommendation
regards to the defect
found.
View Process
All users can view the
(User, Admin , KE)
information about the
Pass (As expected)
process in manufacturing
firm regards to the defect
found.
View Sub Process
All users can view the
(User , Admin, KE)
information about the sub
Pass (As expected)
process in manufacturing
firm regards to the defect
found.
View The Defect
Admin and KE can view
(Admin, KE)
the information about the
defect and relationship
with the symptom,
causes, grade, process,
sub process and
recommendation.
Pass (As expected)
120
Use Case
Normal Input Values
Expected Results
View Causes
All users can view the
(User,Admin, KE)
information about the
Actual Results
Pass (As expected)
causes of defect regards
to the defect found.
View Symptom
All users can view the
(User,Admin, KE)
information about the
Pass (As expected)
symptom of defect
regards to the defect
found.
View Rec
All users can view the
(User,Admin, KE)
information about the
Pass (As expected)
recommendation of tiles
defect regards to the
defect found.
View Product
All users can view the
(Admin, KE)
information about the
Pass (As expected)
tiles product.
View Answer
Admin and KE can view
(Admin, KE)
the answers for dialogues
process regards to the
symptom and questions.
Pass (As expected)
121
Use Case
Normal Input Values
Expected Results
View Question
Admin and KE can view
(Admin, KE)
the questions for
Actual Results
Pass (As expected)
dialogues process regards
to the symptom and
answers.
View Rule base
Admin and KE can view
(Admin, KE)
the rule base DSMPP
Pass (As expected)
system.
Load Defect
User, Admin and KE can
(User,Admin, KE)
load the defect regards to
Pass (As expected)
the symptom, the causes,
the process, the sub
process and the
recommendation of tiles
defect.
Diagnose
User can diagnose the
(User)
tiles defect by inputting
Pass (As expected)
the symptom and
answers some questions.
List All
Admin and KE can list
(Admin, KE)
all the data in knowledge
base system.
Pass (As expected)
122
Use Case
Normal Input Values
Expected Results
Searching
Admin and KE can
(Admin, KE)
search the tiles
Actual Results
Pass (As expected)
information by typing the
keyword.
Load Q_A
Admin and KE can load
(Admin, KE)
the question and answers.
Edit Rule Base
Admin and KE can
(Admin, KE)
manipulate the rule base
Pass (As expected)
Pass (As expected)
by doing some
processing, such as
insert, delete and edit.
Edit Answer
Admin and KE can
(Admin, KE)
manipulate the answers
Pass (As expected)
by doing some
processing, such as
insert, delete and edit.
Edit causes
Admin and KE can
(Admin, KE)
manipulate the causes by
doing some processing,
such as insert, delete and
edit.
Pass (As expected)
123
Use Case
Normal Input Values
Expected Results
Edit symptom
Admin and KE can
(Admin, KE)
manipulate the symptom
Actual Results
Pass (As expected)
by doing some
processing, such as
insert, delete and edit.
Edit process
Admin and KE can
(Admin, KE)
manipulate the process
Pass (As expected)
by doing some
processing, such as
insert, delete and edit.
Edit sub process
Admin and KE can
(Admin, KE)
manipulate the sub
Pass (As expected)
process by doing some
processing, such as
insert, delete and edit.
Edit rec
Admin and KE can
(Admin, KE)
manipulate the
recommendation by
doing some processing,
such as insert, delete and
edit.
Pass (As expected)
124
Use Case
Normal Input Values
Expected Results
Edit Product
Admin and KE can
(Admin, KE)
manipulate the product
Actual Results
Pass (As expected)
by doing some
processing, such as
insert, delete and edit.
Edit Question
Admin and KE can
(Admin, KE)
manipulate the question
Pass (As expected)
by doing some
processing, such as
insert, delete and edit.
5.3.3. White box Testing
White box testing is used base on the method specifications associated with the
class. However, white box testing has had limited impact in object oriented
development. This is due to the rather small size of the individual methods in class. As
such, most approaches to testing classes use black box testing to assure the correctness.
As the result, the white box testing process in DSMPP are focusing on the coding,
functional and method for the overall class in DSMPP system and to make sure that
every line of code execute at least once. Fortunately this DSMPP system can identify the
starting state and stopping state base on the requirements. For examples are:
(i). Starting-state: Authentication credentials (username and password) in login
case, Stopping-state: session-token (cookie) used as proof of authentication.
(ii). Starting-state: Logout signal, Stopping-state: system logout.
(iii). Starting-state: Product entry, Stopping-state: the product values are recorded
in database.
125
(iv). Starting-state: diagnose the defect found, Stopping-state: give the final
result as conclusions.
5.3.4. Usability Testing
Usability testing is an important part of the process, in light of the user interface
problems the Ceramic Horse Company as the user of this system, has experienced with
this DSMPP system. The usability of the interface was assessed by issuing a taskorientated questionnaire to a sample of six prospective users of varying computer and
Internet literacy.
Each user was asked to work and act as the categories users, such as admin,
knowledge engineer and the common users.
After each scenario, the user was asked to answer the following questions about
their experience of the system interface, system validation, system security and
Management Company. The overall questioners can be seen in appendix f.
(i).
Does the DSMPP system produce the valid information for each process?
(ii).
Does the DSMPP system respond at each stage of error processing?
(iii).
Does this system help and implement the information sharing or
knowledge sharing in the company?
(iv).
Does this system implicitly help to increase the company profit?
As the result, the feedback from the usability sessions was positive. The users
felt that the new system was a considerable improvement on their existing methods and
increased their knowledge related to the tiles defect in the manufacturing firm. More
specifically, the questionnaires revealed the following strengths and some areas where
improvements could be made before the system is released.
126
5.4.
User Manual
To access the DSMPP system, these user manuals will help the user or
administrator or KE to operate this system, the flow activities for this system manual are
explained in appendix k (see appendix k).
5.5.
Summary
After conducting the initial finding, analysis-design of knowledge base concept
and implementing the expert system as an interactive computer based system. Finally,
many concept approaches have been implemented in DSMPP system development for
instances are black box testing, white box testing and usability testing. The testing and
implementing phase are used to examine the implementation of basic concept in
knowledge base and knowledge management area. Thus, relate to the forward chaining
method to root the probable causes of defect and implement the depth first search
concepts in rule base searching process. Besides, the overall testing helps the system to
reach the system objectives and fulfill the user requirements.
CHAPTER 6
ORGANIZATIONAL STRATEGY
6.1.
Introduction
Knowledge management in manufacturing process, not just involve in one
department in the company. To produce and increase the quality of process production
has been responsibility for the overall components, such as quality control department,
quality assurance, research and development department, etc. Therefore, it needs good
management in form of managing and maintenance the knowledge worker in that
company. The information that are related to the diagnostics process of tiles defects and
the probable causes root of defects found in manufacturing process, must be shared to
the entire workers in that company, specially for the new workers, knowledge workers
and experts in each departments.
This chapter proposed the method or strategy that can be undertaken by
organization in order to implement the prototype of DSMPP system that can help the
worker or knowledge worker to diagnose the tiles defects and root the sources of product
defect in manufacturing firm. It is starting from the equipping the system infrastructure,
system implementation, training or workshop for the workers, knowledge workers (the
users of the system) and organizational support to implement this system.
128
6.2.
Roll Out Strategy
For effective implementation, there are some steps processes that can be
undertaken by the organization.
6.2.1. Installation of Infrastructure Process
In this step process, the company must be completed by the system requirements,
hardware and software specification. For the first step, implementation of DSMPP
system needs 1 (one) computer and 1 (one) printer, which has some standard computer
specification as listed below:
Table 4.2 Hardware and Software Specification
Hardware
Software
·
Intel Pentium 3 or Intel
Web Browser:
Pentium 4 Mainboard
Microsoft Internet Explorer v5.0+/ Mozilla
·
256MB DDR 400 RAM
v1.0+
·
80GB 7200rpm Harddisk
Network Security
·
52x CDROM Drive
·
15” Monitor
·
Network Packet Filter/Firewall
Operating System
·
Windows XP
Programming Language
·
PHP Programming
Database
·
MySQL (PhpMyAdmin)
Text Editor
·
Macromedia Dreamweaver
129
This system was designed under web base design programming. The
implementation can be connected to the internet or intranet. Therefore for the next step
equipping the infrastructure process can be completed by server and internet connection
devices. By using the internet or intranet, the DSMPP system can be accessed by the
entire workers and knowledge workers where ever they are, anyway, anytime and any
places.
6.2.2. Trainings
After finishing the i nfrastructure installation, the next step is starting with the
workshop. Fortunately, t h e White Horse Ceramics always does some trainings and
workshops for the new workers and all the staff. The workshop more explains about the
ways to increase the product quality and how to classify the tiles base on the quality and
defect found, but all the information given uses manual tools and standard equipments
such as slides, documents, etc. By using the DSMPP system, the workshop can be
modified and more computerize wherever DSMPP system will help the trainee how to
recognize the tiles defects, probable causes of defects and defects maintenance.
For the trainee who has been classified as workers, the information given is
limited just for helping to diagnose the tiles defects and understanding well the way to
solve the problems. Meanwhile, the trainee who has been classified as knowledge
workers or even the experts in each department, will be trained to operate the whole
function in the DSMPP system, therefore they will understand how this system helps
their work to detect the defects, increase the tiles production and decrease the defects
problems in manufacturing process firm. For the user who had been categorized as
admin or operator, the workshop given more concentrates to operate the system that is
under commanded the knowledge worker or the experts.
130
6.2.3. System Implementation Process
System Implementation process starts by implementing the system in their daily
activities in manufacturing firm. It is more to cultivate the staff and workers with
DSMPP system. The knowledge workers start to do their duty by creating the new
knowledge base and admin starts to insert the new data in this system, therefore this
system is always upgraded and up-to-dated. The new information of knowledge bases
can be shared to the whole knowledge workers in that company. It means there is some
knowledge management implementation in terms of knowledge transferring and sharing
process in manufacturing firm.
6.2.4. The Organizational Support
The transitional system in the organization, needs big support from the whole
management form in that organizations, especially the people who are directly involved
as the user of the system, such as the experts or knowledge workers, workers and admin
as operator. The entire categorized user will be directed by the organizations. To support
that system, the organization must be made this DSMPP system as obligatory tools to
help them work in the manufacturing firm.
6.3.
Summary
Many ways, many strategies and many approaches had been done to reach the
main objectives of DSMPP system. Of course the successful of this system depends on
the entire parts in the organizations, starting from the researcher, the staff and all
131
elements in that organization to apply this system in their work activities. All the
obstruction and challenges can be well settled if it gets full support from the whole part
in that organizations especially support from the high level management.
CHAPTER 7
DISCUSSION AND CONCLUSION
7.1.
Introduction
After passing by the overall system development in this project, starting from the
project 1 that explained the project review, literature review, methodology, initial
findings and ending by project 2 that discussed about the advanced analysis and design,
system implementation on method and tools application, system testing and finally the
organizational strategy which produces the strategic planning to apply the system
environment in that company. Therefore in this chapter, t h e discussions are more
focusing on project achievement, constraints, challenges and new aspiration for the
future system development.
7.2.
Achievements
As discussed in the previous chapter, a basic concept and theory of this research
had been recognized and identified.
133
The main findings that had been highlighted in this project are listed below:
1.
The basic concepts of knowledge base and expert system as an interactive
computer based system that helps to diagnose the defect problems in
manufacturing process had been applied in this project.
2.
The idea of forward chaining method and depth first searching to root the
probable causes of defects, starting by recognizing the knowledge base,
data, information and physically facts of tiles defect and ending by the
diagnostic result as a conclusion of the system had been implemented
well which related to the method explained in the previous chapter.
3.
The concept of knowledge management in this project are focusing on the
knowledge sharing of the data, the fact, the knowledge base and the rule
bases among the knowledge worker or experts. Hence, by using this
system KE and admin can manipulate the knowledge.
4.
How’s big the defects will influence the production quality in the
manufacturing firm and How’s the system can reduce the defect
problems.
7.3.
Constrains and Challenges
There were some constrains and challenges that had been faced while conducting
the research process for the analysis phase and implementation phase. Some of the
constraints were:
1.
The case study organization is unable to provide all the real information
about the specifications standard that needed for the research because
some of the information was private and confidential.
134
2.
This system involves many knowledge workers in the difference
background of knowledge and department. Therefore, it is quite difficult
to absorb and brainstorm the knowledge workers on the same views and
perspectives in the diagnostic processes.
3.
In the diagnostic system did not use the lingual languages dialogues.
Therefore, the knowledge base was entered by chosen symptom or causes
in database.
4.
DSMPP System is limited by 100% possibility for the diagnostics
solution of course it is related to the knowledge base and rule bases in
this system. The conclusion was produced with 1 or 0 defect found.
5.
Time constrains, as it took a long time to identify and analyze the process
that involved in White Horse Ceramics.
6.
Taking too much time in absorbing the knowledge.
The challenges are as follow:
i.
Object Oriented Methodology has been chosen as the accurate system
development model as a guide in developing this project. Choosing the wrong
model, can lead to improper project development and this will cause a longer
period to develop the project.
ii.
Modeling the to-be process for DSMPP. The to-be processes should be designed
carefully because it will be the guidelines to develop the system later.
iii.
The Good co operations of the experts help the researcher in absorbing and
transferring the knowledge.
iv.
Implementing the knowledge management sharing towards the implementation
of IS knowledge management concept in that company.
135
7.4.
Aspiration
Following the procedures of system development, finally it is hoped that:
i.
All the objectives of the system that have been outlined in the beginning of the
project are achieved.
ii.
The rule base has been implemented successfully in the system.
iii.
The DSMPP has been developed and implemented successfully. The user can
accept the system.
iv.
The DSMPP system can be used to help the new workers.
v.
To develop future system, the diagnostics processes in DSMPP system can be
applied by implementing the artificial intelligent concept which is more focusing
on the lingual language dialogues.
vi.
The future DSMPP system can be applied a logical fuzzy concept. That concept
has possibility of defect found between 1 and 0.
vii.
The DSMPP system can be developed by implementing others concept of
knowledge base, such as backward chaining, combination backward-forward
chaining and matching search concept.
7.5.
Summary
As a conclusion, all the activities that should be completed in Project 1 and
Project 2 had been successfully completed, starting from initial phase and followed by
analysis phase. The designing phase of t he project also had been implemented well,
where the to-be processes had been modeled in chapter 4, implementing and testing was
completed, organizational strategy to implement this system had been discussed and
finally the conclusion for the overall system development had been produced.
136
Besides, the strength area and system achievements, this chapter also discuss the
weakness of DSMPP system mainly the system implementation that not fulfilled the
user requirement. Many aspirations for the future system development have been
produced, in a sense to increase the system quality in terms of method application.
137
REFFERENCES
Elias M.Awad (1996), Building Expert Systems, Principles, Procedures and
Applications. America: West Publishing Company
Richard Frost (1986), Introduction to Knowledge Base System, America: Macmillan
Publishing
M. Cemal Cakir , Kadir Cavdar (2005), Development of a knowledge-based expert
system for solving metal cutting problems, Uludag University, Mechanical
Engineering Department, Gorukle 16059, Bursa, Turkey
Giarratano/Riley (1989), Expert System, Principles and Programming, Boston, PWSKENT Publishing Company.
John A Schey(2000), Inttroduction to Manufacturing Process, London:Mc.Graw Hill
M.G. Abou-Alia, M. Khamisb (2003), TIREDDX: an integrated intelligent defects
diagnostic system for tire production and service, Faculty of Engineering,
Alexandria University, Alexandria and University of Akron, Akron, OH, USA
Young-Jin Kim(1995) , A Framework for an On-Line Diagnostic Expert System for
Intelligent manufacturing, Kyunghee University, Kyung-Gi-Do, Korea
Dennis (2005), System Analysis and Design with UML Version 2.0, USA: John Wiley
and Son, Inc.
Et Keravnou and L.Johson (1996), Competent Expert Systems, A case Study in Fault
diagnosis, Mc.Graw Hill.
Alavi, Maryam, Leidner, Dorothy E. (2001), Review: Knowledge Management and
Knowledge Management Systems: Conceptual Foundations and Research Issues,
MIS Quarterly, 25 (1) Mar., pp. 107-136.
Luke Welling, Laura Thomson, Second Edition (2003), PHP and MySQL Web
Development, Sam Publishings.
Jason T.Roff (2003), UML: A beginner’s Guide, Mc.Graw-Hill/Osborne
138
Jeffery L.Whitten, Loonie D.Benthley, Kevin C.Dittman ,Six Edition (2004), Metode
Desain and Analisis sistem, Mc.GrawHill- Education, translated by Penerbit
ANDI-Indonesia.
Jay Liebowitz, LylenC.Wilcox (1997), Knowledge Management and Its Integrative
Elements, CRC Press, Boca Raton-New York.
James W.Cortada, John A.Woods (2000), The Knowledge Management Yearbook 20002001, Butterworth Heinemann.
C.S. Krishnamoorthy; S. Rajeev (1996), Artificial Intelligence and Expert Systems for
Engineers, CRC Press, CRC Press LLC
Stuart J. Russell and Peter Norvig (1995), Artificial Intelligence A Modern Approach,
Prentice Hall, Englewood Cliffs, New Jersey
139
Table 2.1 Quality Control Plan
White Horse Quality Control Plan
Production Stage: Glazing Operation
No
Type/Category
1
Inspection
Item
Weight
Condition
Water Spraying
Height
2
Glaze Slip
Density
Viscosity
Glazing Weight
Responsible
Unit
Glazing QA
Inspection
Frequency
Sample
Size
Form
Used
Reference
Standard
At least
1/hour
At least
2/shift at
every line
At least
1/hour
At least
2/shift at
every line
At least
2/shift
At least
2/shift at
every line
At least
1/hour
At least
2/shift at
every line
At least
1/hour
At least
2/shift at
every line
At least
1 steel
plate
Glazing
In process
section
glaze
Test Report specification
Water weight
of glazing
Spry Gun
Height
specification
of glazing
1
capacity
of a
density
cup
1
capacity
of a
viscosity
cup
1 steel
Referen
ce
(Proced
ure
W.I)
Glazing
Operatio
n
Procedur
e
QA in
process
procedur
e
In
Process
Glazing
Inspectio
n
140
Glaze Level
3
4
Sieve
Aluminum Slip
Condition
Density
Weight
Conditions
5
Brushing Loss
Weight
Thickness
1/hour
At least
2/shift at
every line
At least
1/hour
At least
2/shift at
every line
At least
1/hour
At least
2/shift at
every line
At least
1/hour
At least
2/shift at
every line
At least
2/shift
At least
2/shift at
every line
At least
1/hour
At least
2/shift at
every line
At least
1/hour
At least
2/shift at
every line
At least
1/hour
plate
-
-
-
-
1
capacity
of a
density
cup
1 pc
Aluminum
slip density
and weight of
glazing
-
-
1 pc
1 pc
Glazing
section
Test Report
-
-
141
6
7
Tile Edges
Raw Bodies
No
Type/Category
1
Finished Product
At least
2/shift at
every line
Trimming
At least
1 pc
1/hour
At least
2/shift at
every line
Appearance
At least
1 pc
1/hour
At least
2/shift at
every line
Body
At least
³ 1 pc
Temperature
1/hour
At least
2/shift
White Horse Quality Control Plan
Production Stage: Finished Product Inspection for Firing
Inspection
Responsible
Inspection Sample
Form
Item
Unit
Frequency Size
Used
Firing
Q
A
Curved In
Sample
taken form
each line
on every
shift, if the
models The
QC
numbers or
sizes has
been
changed
during a
5 pc
Finished
good sizing
curve
in/out,
rectangular
ity and
gloss
inspection
record.
The printing
body
temperature
specification
Reference
Standard
Firing finished
goods for wall
tiles,
homogenous,
ceramic tiles.
Refer
ence
(Proc
edure
W.I)
142
Firing Section
size
Visual
Sampling
Shading
Comparison
Chemical
Resistance(hou
sehold,
swimming pool
salts, acid,
alkalis)
Glaze tiles
crazing
Stain
Resistance
shift,
samples
will be
taken again
³ 30 x 30
cm once
per shift
base on the
size.
Every
pallet
1 cartoon
Finished
good
Inspection
record
1 pc
Ceramic
5 pc
tiles, wall
tiles,
homogenou
s tiles and
homogenou
s special
sample
taken once
a week
For whole
5 pc
tiles
sample
taken in 1
day
Ceramic
5 pc
tiles, wall
tiles,
homogenou
Laporan
Uji Kaji
Ketahanan
Jubin
terhadap
kotoran
dan bahan
kimia
Glaze Tiles
crazing
Resistance
Test Report
Laporan
Uji Kaji
Ketahanan
Jubin
Finished tiles
Visual
sampling test
specification
Model
Standard
sample
143
Perform
abrasions
2
Pneumatic Press
Air Pressure
3
Packaging
Packaging
verification
s tiles and
homogenou
s special
sample
taken once
a week
Once three 11 pc
month for
all type of
tiles
Once per
shift
Every
pallet
1 cartoon
Table 3.2 Form Target and Tiles Problem
terhadap
kotoran
dan bahan
kimia
Glaze tiles
surface
abrasions
Resistance
Test Report
Finished
Good
inspection
record
146
1:mainMenu
: Admin
: Knowledge
Engineer
2:loginPage
3:adminPage
4:kePage
5:userPage
: User
browse
browse
browse
login
validateLogin
successLogin
sucsessLogin
successLogin
unsuccesfullLogin
Sequence Diagram: Login
147
1:adminPage
: Admin
: Knowledge
Engineer
2:kePage
3:userPage
4:logoutPage
5:mainMenu
: User
browse
browse
browse
logout
logout
logout
validate logout
back to
Sequence Diagram: Logout
148
KEPage
: Knowledge
Engineer
AdminPage
SearchingPage
: Admin
browse
browse
insertKeyWord
insertKeyWord
SearchingResult
Sequence Diagram: Searching
149
1:displayScreen
: Knowledge
Engineer
2:deleteProductPage
3:
insertProductPage
4:updateTilesPage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewProduct
KEselectProduct
AdminviewProduct
AdminselectProduct
deleteProduct
InsertProduct
updateProduct
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditProduct
7:ProductDB
150
1:displayScreen
: Knowledge
Engineer
: Knowledge
Engineer
2:
deleteSymptomPage
3:
insertSymptomPage
4:
updateSymptomPage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewSymptom
KEselectSymptom
AdminviewSymptom
AdminselectSymptom
deleteSymptom
InsertSymptom
updateSymptom
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditSymptom
7:SymptomDB
151
1:displayScreen
: Knowledge
Engineer
2:deleteDefectPage
3:insertDefectPage
4:updateDefectPage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewDefect
KEselectDefect
AdminviewDefect
AdminselectDefect
deleteDefect
InsertDefect
updateDefect
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditDefect
7:DefectDB
152
1:displayScreen
: Knowledge
Engineer
2:deleteRecPage
3:insertRecPage
4:updateRecPage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewRec
KEselectRec
AdminviewRec
AdminselectRec
deleteRec
InsertRec
updateRec
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditRec
7:RecDB
153
1:displayScreen
: Knowledge
Engineer
2:deleteCausesPage
3:
insertCausesPage
4:updateCausesPage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewCauses
KEselectCauses
AdminviewCauses
AdminselectCauses
deleteCauses
InsertCauses
updateCauses
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditCauses
7:CausesDB
154
1:displayScreen
2:deleteProcessPage
3:
insertProcessPage
4:
updateProcessPage
5:ConfirmPage
6:messageDisplay
: Admin
: Knowledge
Engineer
KEviewProcess
KEselectProcess
AdminviewProcess
AdminselectProcess
deleteProcess
InsertProcess
updateProcess
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditProcess
7:ProcessDB
155
1:displayScreen
: Knowledge
Engineer
2:
deleteSubProcessPage
3:
4:
insertSubProcessPage updateSubProcessPage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewSubProcess
KEselectSubProcess
AdminviewSubProcess
AdminselectSubProcess
deleteSubProcess
InsertSubProcess
updateSubProcess
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditSubProcess
7:
SubProcessDB
156
1:displayScreen
: Knowledge
Engineer
2:deleteRulebasePage
3:insertRuleBasePage 4:updateRuleBasePage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewRuleBase
KEselectRuleBase
AdminviewRuleBase
AdminselectRuleBase
deleteRuleBase
InsertRuleBase
updateRuleBase
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: Edit Rule Base
7:RuleBaseDB
157
1:displayScreen
: Knowledge
Engineer
2:deleteQuestionPage
3:insertQuestionPage 4:updateQuestionPage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewQuestion
KEselectQuestion
AdminviewQuestion
AdminselectQuestion
deleteQuestion
InsertQuestion
updateQuestion
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditQuestion
7:QuestionDB
158
1:displayScreen
: Knowledge
Engineer
2:deleteAnswerPage
3:insertAnswerPage
4:updateAnswerPage
5:ConfirmPage
6:messageDisplay
: Admin
KEviewAnswer
KEselectAnswer
AdminviewAnswer
AdminselectAnswer
deleteAnswer
InsertAnswer
updateAnswer
confirmMessage
validateEdit
successEdit
save
Sequence Diagram: EditAnswer
7:AnswerDB
159
KEPage
: Knowledge
Engineer
: Admin
AdminPage
UserPage
SymptomtPage
: User
browse
browse
browse
KESubscribeSymptom
AdminSubscribeSymptom
SymptomShow
Sequence Diagram: ViewSymptom
160
KEPage
: Knowledge
Engineer
AdminPage
ProductPage
: Admin
browse
browse
KESubscribeProduct
AdminSubscribeProduct
ProductShow
Sequence Diagram: ViewProduct
161
KEPage
: Knowledge
Engineer
: Admin
AdminPage
UserPage
CausestPage
: User
browse
browse
browse
KESubscribeCauses
AdminSubscribeCauses
CausesShow
Sequence Diagram: ViewCauses
162
KEPage
: Knowledge
Engineer
: Admin
AdminPage
UserPage
ProcessPage
: User
browse
browse
browse
KESubscribeProcess
AdminSubscribeProcess
ProcessShow
Sequence Diagram: ViewProcess
163
KEPage
: Knowledge
Engineer
: Admin
AdminPage
UserPage
SubProcessPa
ge
: User
browse
browse
browse
KESubscribeSubProcess
AdminSubscribeSubProcess
SubProcessShow
Sequence Diagram: ViewSubProcess
164
KEPage
: Knowledge
Engineer
: Admin
AdminPage
UserPage
: User
browse
browse
browse
KESubscribeRec
AdminSubscribeRec
RecShow
Sequence Diagram: ViewRec
RectPage
165
KEPage
: Knowledge
Engineer
AdminPage
AnswerPage
: Admin
browse
browse
KESubscribeAnswer
AdminSubscribeAnswer
AnswerShow
Sequence Diagram: ViewAnswer
166
KEPage
: Knowledge
Engineer
AdminPage
QuestionPage
: Admin
browse
browse
KESubscribeQuestion
AdminSubscribeQuestion
QuestionShow
Sequence Diagram: ViewQuestion
167
KEPage
: Knowledge
Engineer
AdminPage
DefectPage
: Admin
browse
browse
KESubscribeDefects
AdminSubscribeDefects
DefectsShow
Sequence Diagram: ViewDefects
168
UserPage
AllPage
: User
browse
UserSubscribeAll
AllShow
Sequence Diagram: ViewAll
169
diagnosePage
resultPage
: User
diagnose
view result
resultInfo
print
Sequence Diagram: ViewTheResult
printPage
170
UserPage
CompanyInfoPa
ge
: User
browse
UserSubscribeCompanyInfo
CompanyInfoShow
Sequence Diagram: ViewCompanyInfo
171
KEPage
: Knowledge
Engineer
AdminPage
ListAllPage
: Admin
browse
browse
KESubscribeAll
AdminSubscribeAll
AllInfoShow
Sequence Diagram: ViewListAll
172
KEPage
: Knowledge
Engineer
AdminPage
RuleBasePage
: Admin
browse
browse
KESubscribeRuleBase
AdminSubscribeRuleBase
RuleBaseShow
Sequence Diagram: ViewRuleBase
173
KEPage
: Knowledge
Engineer
: Admin
AdminPage
UserPage
SymptomPage
Sym_def
ProcessPage
Process_def
CausesPage
Causes_def
: User
SubProcessPa
ge
SubProcess_d
ef
RecPage
Rec_def
browse
browse
browse
chooseSymptom
chooseSymptom
chooseSymptom
select_defect
sym_defInfo
chooseProcess
chooseProcess
chooseProcess
select_defect
process_defInfo
chooseCauses
chooseCauses
chooseCauses
select_defect
causes_defInfo
chooseSubProcess
chooseSubProcess
chooseSubProcess
select_defect
subProcess_defInfo
chooseRec
chooseRec
chooseRec
select_defect
Rec_defectInfo
Sequence Diagram: LoadDefect
174
UserPage
DiagnosePage
: User
ConfirmMessag
e
ResultPage
browse
inputSymptom
ConfirmSymptom
validateSymptom
question
answer
ConfirmAnswer
validateAnswer
getTheResult
ResultInfo
Sequence Diagram: Diagnose
175
KEPage
: Knowledge
Engineer
AdminPage
QuestionPage
AnswerPage
Question_Answ
erPage
: Admin
browse
browse
chooseQuestion
select_Answer
Q_AInfo
chooseAnswer
select_question
Q_AInfo
chooseQuestion
select_Answer
Q_AInfo
chooseAnswer
select_question
Q_AInfo
Sequence Diagram: LoadQ_A
176
Process
Glazing
(1)
Sub
Process
Water
Spraying
(1)
Defect
Pin Hole
(I)
Table 4.1 Tiles Defect and Probable causes
Causes
Symptom
-Check the water
spraying condition(a)
-Check temperature
(b)
-Check machine speed
(normal or not or base
on the specification?)
(c)
- Check the firing
inspection report (d)
Grade
- The defects are on the
B
surface of tiles (37)
-There is small hole on
the tiles surface,
commonly on the edge of
tile (5)
-The small holes in the
graphics design of tiles
(it will appear clearly
before firing process) (6)
-Nozzle is clogged up
(99)
-Water spraying weight
less than (1) spec
- the water spraying is
not full enough to spray
into the tiles (2)
-Temperature 80 – 100
degree Celsius (3)
-machine speed (normal
or not or base on the
specification?) (4)
- Any defects were
found in firing inspection
report (136)
- Picture ( 8)
Solution/Corrective
Action
-If the hole is small
then categorized in B
- If the water spraying
weight less than spec
so adjust “ball valve”
by closing it until get
the specification
needed.
-if the water spraying
condition is not full
enough to spray into
the tiles so :
1. Open the water
nozzle, clean it and
plug in again
2. Change the nozzle
with the new one
-if body temperature
more than spec so
1. throw away tiles
from the line
2. Open the fan blower,
suction fan A ,axis fan
3. Inform the press
department to adjust
temperature of tiles.
If body temperature
less than spec: do the
same things.
177
Dip sink
(glaze
slip) (II)
-Check Glaze level (e)
-Check weight of
glaze (f)
-Check viscosity of
glaze(g)
-Check Density of
glaze (h)
- Check the curvature
report
-Viscosity less than
specification (9)
-Viscosity more than
spec (10)
-Density less than
spec(11)
-Glaze weight less than
spec (12)
-Density more than
spec(13)
-Glaze weight more than
spec(14)
-Glaze level less than
spec(15)
-Glaze Problems (83)
- The defects are on the
surface of tiles (37)
B/C
-If density more than
spec : add the water
and mix up base on the
spec needed
-if density less than
spec: add the similar
glaze slip and mix up
until get the spec
needed.
-If glaze slip viscosity
more than spec: add
some water and mix up
until get the spec
needed.
-If glaze slip viscosity
less than spec: add
some similar glaze slip
until get the spec
needed
-if glaze weight more
than spec: close the
“ball valve” until get
the weight base on the
spec
-if the glaze weight less
than spec: open the
“ball valve” until get
the weight needed.
-if glaze level less than
spec add the glaze
more than the
spec.(until get the spec
needed)
178
Sieve(2)
Tiles
Dust
(III)
-Check the sieve
condition (i)
-Check sieve
maintenance (j)
Dirty
Dust
(IV)
-Check the sieve
condition (i)
-Check sieve
maintenance (j)
-loss control in sieve
maintenance (l)
-The sieve has been
B/C
broken (16)
-Mesh is worn out (17)
- The defects are on the
surface of tiles (37)
-There is some dust on
the tiles surface (can be
categorized in B) (19)
- When handy touched,
the surface of tiles are
covered with bumps or
bruised. (20)
-When it looked by
naked eye for the
distance around 1 meter,
there is some hump or
outgrowth on the tiles
surface. (21)
- If the hump more than 2
it is categorized grade C
(22)
-loss control in sieve
maintenance (33)
-Picture1 (23)
-Picture 2 (24)
-Picture3 (25)
- The sieve has been
B/C
broken (16)
- Mesh is worn out (17)
- There is some dust on
the tiles surface (can be
categorized in B) (19)
-The defects are on the
-check the sieve
condition
-Change the sieve with
the new one
-check the sieve
condition
-Change the sieve with
the new one
179
Alumuniu
m Slip(3)
Drop
Glazing(
V)
-Check the sieve
condition (i)
-Check sieve
maintenance (j)
-human factor error
(k)
- loss control in sieve
maintenance (l)
Crack
Glazing
(VI)
-check aluminum slip
condition (m)
-Check the weight of
tiles (n)
surface of tiles (37)
- the dirty dust didn’t
impact on the tiles
surface, it can be
categorize grade B, if not
grade C (27)
-Picture (28)
-There are some drop
glazing on the tiles (29)
- The sieve has been
broken (16)
- Mesh is worn out (17)
-Mesh is coarse or rough
to touch (30)
-there are some
accumulation glazing on
the tiles surface (31)
- The defects are on the
surface of tiles (37)
- the accumulations of
glazing are influencing
the tiles performance
(32)
-the surface of glaze is
not flat and it influences
the structure of tiles. (64)
- glaze problems (83)
-Aluminums slips on the
tiles (138)
-Weight of tiles is more
than spec (35)
C
B/C
-check the sieve
condition
-change the sieve with
the new one
-
Check the
aluminums
components in
the aluminums
180
-Weight and density of
aluminum is not base on
the spec (density 111113g and weight 1.0g)
(36)
- The defects are on the
surface of tiles (37)
-there are number of
scratches on the tiles
surface.(38)
- glaze problems (83)
Dip sink
(II)
Brushing
Loss (4)
-check aluminum slip
condition (m)
-Check behind of the
tiles (o)
Changing - the brush problems
color(VII -human factor error
)
(k)
-Check brush
condition (p)
- Check Brush setting
(q)
-Check weight and
thickness of brushing
loss.(r)
-No Aluminums
slip .(39)
-there are number of
windy holes. (40)
-the defects influences
the problems of windy
holes at the back of
tiles.(41)
-
C
-Brush is not flat (43)
B/C
-Condition of brush is not
on the spec standard (44)
-The weight of brush is
not base on the spec(132)
-The color of tiles not
base on the spec (48)
-it seems the different
color in one kinds of tiles
(49)
-The defects are on the
surface of tiles (37)
machine, full
enough or not?
Adjust
aluminum
machines until
the back of tiles
are touched by
aluminum.
-Check Aluminum slip
condition
-if the brushing loss is
less than spec: adjust
brushing machine and
falling down the brush
to get the weight base
on the standard.
-if the thickness
brushing loss is more
than spec: adjust
machine brushing by
increasing the brushing
loss position to get the
181
The tiles
surface is
not flat
(VIII)
Glaze is
not flat
(IX)
-human factor error
(k)
-Check brush
condition (p)
- Check Brush setting
(q)
-Check weight and
thickness of brushing
loss.(r)
-check the sides of the
tiles
-
pin hole (I)
commonly
happened in
glazing
process(s)
-screen problem (50)
-picture (132)
-picture (133)
-Brush is not flat (43)
- Condition of brush is
not on the spec standard
(44)
- The defects are on the
surface of tiles (37)
- The color of tiles not
base on the spec (48)
- the surface of tiles are
little bit fluctuate (53)
-the tiles are not
symmetric, when it is
looked in each sides (54)
- the glaze surface
fluctuates or it seems not
flat. (55)
-there are some hollow
on the tiles surface (56)
-the tiles surface seems
billow up (57)
-when it is touched, the
surface is not flat(58)
-Picture(59)
- Glaze
problems(83)
- screen
problem(50)
- shading
problems(62)
- The color is not
weight base on the
standard or the spec.
B/C
C
-if the brushing loss is
less than spec: adjust
brushing machine and
falling down the brush
to get the weight base
on the standard.
-if the thickness
brushing loss is more
than spec: adjust
machine brushing by
increasing the brushing
loss position to get the
weight base on the
standard or the spec.
-
if the common
problems is
shading
problems, so
refers to
printing process
and report.
182
Tiles
Edges (5)
The
overlappi
-the chipped are
caused by overlapping
balance (60)
- The color of tiles
not base on the
spec (48)
- The design is not
base on the
spec/sample (61)
- The defects are on the
surface of tiles (37)
the un-flattened
glazing will influence
the tile structure and
performance (64)
- the glazing
surface looks not flat
(65)
-it seems the different
color in one kinds of tiles
(49)
- it seems the
different texture
of glaze (66)
- the un-flattened
glazing will not
influence the tile
structure and
performance (67)
- Picture(68)
- Picture (69)
- Picture(70)
- Picture (71)
-there are some tiles that B/C
get caught on the roller
-
-
For the glazing
problems refers
to glazing
report.
-If there is
shading
problems, the
light color tiles
can not be
mixed with the
dark color tiles
(if it is still
categorized
grade B)
-tends to grade C
- Adjust trimming
183
ng
tiles/chip
ped Tiles
(X)
tiles (t)
-Tiles are overlapping
(u)
-check the roller
problems report(v)
-the blocking
condition (w)
-it can cause the pin
hole (I)
-Check Trimming(x)
kiln (74)
-it seems many chipped
on the edge of tiles (77)
- The defects are on the
surface of tiles (37)
- the chipped are on the
edge of tiles(78)
- support bracket
condition has a problems
(79)
- motor belting speed not
base on the spec(80)
-Picture(81)
-Picture (82
Glazing
peel (XI)
-Check belting in
machine glaze (y)
-Check Glaze level (e)
-Check weight of
glaze (f)
-Check viscosity of
glaze(g)
-Check Density of
glaze (h)
- Check the curvature
report
-Tiles are overlapping
(84)
-glaze problems (83)
-Viscosity less than
specification (9)
-Viscosity more than
spec (10)
-Density less than
spec(11)
-Glaze weight less than
spec (12)
-Density more than
spec(13)
-Glaze weight more than
spec(14)
-Glaze level less than
spec(15)
C
machine until it cleans
the glaze slip.
-Check the edge of
support bracket, does it
adjust correctly or not?
-the speed of motor
belting must be on the
spec.
- -Check stepping
motor belt in
the printing
machines, does
it too tight or
not, ensure the
skin is on its
rolling way.
1. Check density,
weight,
viscosity glaze
slip
2. Check disco
machine, it is
clean or not
3. if there are
some line or
scratches in the
glazing
fountain, it must
be cleaned by
sponge. If it
doesn’t work,
cleans the
machine.
184
Row
bodies (6)
Crack
(XII)
-bad quality of raw
material (z)
-Tiles are overlapping
(u)
-Check temperature
(b)
-check press
investigation report
(aa)
-check motor belting
speed (bb)
- belting in machine
glaze has problems(86)
- there are number of
scratches on the tiles
surface.(38)
-there are some glazing
chipped on it (87)
-the position of glazing
chipped are just on the
edge of tiles.(88)
-Temperature less than
B
spec (89)
-there are some crack on
the tiles surface (91)
-the crack influences the
tiles structure (92)
- The defects are on the
surface of tiles (37)
-there are number of
scratches on the tiles
surface.(38)
-it is still in a good
condition when knocks
on (102)
- motor belting speed not
base on the spec(80)
-machine printing
problems (94)
-Picture (95)
-Picture (96)
-Picture (97)
-
Clean the back
of tiles where
there are some
glazes on it, and
clean the
machine glaze
belting
-if body temperature
more than spec so
1. throw away tiles
from the line
2. Open the fan blower,
suction fan A ,axis fan
3. Inform the press
department to adjust
temperature of tiles.
If body temperature
less than spec: do the
same things.
-if the problems comes
from section press,
inform to press
department
-if it is happened when
the glazing is
overlapping, adjust
motor belting speed
-If the problems are in
machine printing,
adjust the basement of
185
Bend
(XIII)
-human factor error
(k)
-Check temperature
(b)
-check the raw
material (z)
-check roller kiln (cc)
-check the roller
problems report
- check the sides of the
tiles
Easy
broken
(XIV)
-check the raw
material (z)
-Check temperature
(b)
-human factor error
(k)
-Check ball milling
process (dd)
-Check machine
problems (ee)
-Less time control (ff)
Temperature less than
B
spec (89)
-Temperature more than
spec (3)
-bad quality of raw
material (90)
- roller problem(52)
-the roller has broken
(133)
-the tiles are not
symmetric, when it is
looked in each sides (54)
- The defect are found
When it is looked in each
sides.
-Temperature less than
C
spec (89)
- the crack influences the
tiles structure (92)
- The defects are on the
surface of tiles (37)
-there are number of
scratches on the tiles
surface.(38)
-it will broken when
knocks on (101)
-there is some defect
report on press
investigation report (93)
-ball milling processing
printing machine until
it touches the back of
tiles.
-if body temperature
more than spec so
1. throw away tiles
from the line
2. Open the fan blower,
suction fan A ,axis fan
3. Inform the press
department to adjust
temperature of tiles.
If body temperature
less than spec: do the
same things.
-if body temperature
more than spec so
1. throw away tiles
from the line
2. Open the fan blower,
suction fan A ,axis fan
3. Inform the press
department to adjust
temperature of tiles.
If body temperature
less than spec: do the
same things.
186
Printing
Problems
(XV)
Firing(2
)
Finished
Size
Product (1) (XVI)
-
Check machine
printing
condition? (gg)
- Use grease
printing that
base on the QA
requirement
(hh)
- Check the skin
position (ii)
-Check the skin
condition (jj)
-loss Control in
pressing process (kk)
- time Control when
pressing process (ll)
has problem (104)
- machine has problems
(105)
-raw material not base on
the spec (98)
-Temperature more than
spec (3)
-easy broken or fragile
(107)
-The color of tiles not
B/C
base on the spec (48)
-it seems the different
color in one kinds of tiles
(49)
- The defects are on the
surface of tiles (37)
- The tile’s design
not on the spec
(108)
- the skin position
not base on the
spec (110)
- the skin has problems
(111)
-the gas pressure is not
base on the spec (112)
-Raw material are not
base on the spec (98)
B/C
1. Check printing
machine, does
the machine
work? And
adjust in the
right ways,
before using the
grease printing
that base on the
QA
requirements.
2. Check the skin
installing
direction or use
the right skin.
3. Adjust printing
machine until
the skin are full
enough with the
tiles surface.
-The overall sizing
differences in one batch
can not be greater than
2.00 mm
187
Bend
(XIII)
-it can cause the
chipped tiles (X)
-Check temperature
(b)
-check gas pressure
(mm)
-check curvature (nn)
-check verification of
shade/shading
problems (oo)
-Check Size (qq)
-machines problems
(105)
- curvature has problem
(48)
-there is some
verification of
shade/shading problems
(62)
-Loss control in
glazing process (pp)
-Check temperature
(b)
-Check curve in and
out process (nn)
-Check Size (qq)
-Check the weight of
the tiles (f)
-Temperature is less than
spec (89)
-Temperature more than
spec(3)
-Tiles condition is dome
shape (115)
- the surface of tiles are
little bit fluctuate (53)
-the tiles are not
symmetric, when it is
looked in each sides (54)
-the curve in and out
process not base on the
spec(48)
-Size not base on the
spec (XVI)
- the weight of the tiles
not base on the spec
(114)
-The defect are found
When it is looked in each
-Sizing differences in
one piece of tile can not
be greater than 1.2 mm
-Diagonal sizing
differences in one
pieces of tile can not be
greater than 1 mm.
B
-if body temperature
more than spec so
1. throw away tiles
from the line
2. Open the fan blower,
suction fan A ,axis fan
3. Inform the press
department to adjust
temperature of tiles.
If body temperature
less than spec: do the
same things.
-if the problems are in
glazing process
Check density,
viscosity weight and
level of glaze.
188
The
maturity
tiles/the
overripe
tiles
(XVII)
-Loss Control in firing
process (rr)
- Check the firing
inspection report (d)
-Check temperature
(b)
- check the raw
material
Water
absorptio
n
(XVIII)
-the deep of molded
(tt)
-Check Temperature
(b)
The
shinning
tiles
(XIX)
-Check or compare
with the standard tiles
(ss)
-Check temperature
sides.
- there is problems on the
firing inspection report
(7)
-Temperature less than
spec(89)
-Raw material ( the
components are not
regarding to the spec)
(90)
-Bad Quality of powder
(103)
- it will broken when
knocks on (101)
C
-the raw material
B/C
composition is not on the
spec (90)
-the molded size is not
on the spec (118)
- The defects are on the
surface of tiles (37)
-After the tiles washing
process, there are the
water shadow on the
surface of tiles in 1 m
distance (119)
-Picture (120)
-the tiles seem too
B/C
shinning (121)
- Temperature more than
spec(3)
-if body temperature
more than spec so
1. throw away tiles
from the line
2. Open the fan blower,
suction fan A ,axis fan
3. Inform the press
department to adjust
temperature of tiles.
If body temperature
less than spec: do the
same things.
-Inform to QA to check
the quality of powder
or raw materials
-Adjust temperature
-Check water
absorption
-check the chemical
resistance
-Adjust temperature
-Check water
absorption
-check the chemical
189
(b)
Pneumatic
Press (2)
The
darken
tiles
(XX)
-Check temperature
(b)
The
overlappi
ng
tiles/chip
ped Tiles
(X)
-Tiles are overlapping
(u)
-Machine
problems(machine
unloading) (ee)
-Human factor (k)
-Check roller kiln
The
windy
layer
(XXI)
-loss Control in
pressing process (kk)
-check press
investigation report
(aa)
- Check the firing
inspection report (d)
-Check the sides of the
tiles (uu)
-Check air pressure
(vv)
-it is not base on the spec
when it compare with the
standard tiles (122)
-the tiles is mate tiles
(123)
-Temperature is less than
spec (89)
-Temperature is high, it
seems blur (darken)
(124)
-the chipped are on the
center of tiles surface
(125)
-the chipped is small
(126)
-it detects in row body
checking process) (37)
-the defect or problems
found in the roller
report(52)
-there is some problems
in roller kiln (134)
-there are some problems
in press investigation
report (93)
- there are some
problems on the firing
inspection report (7)
-the tiles are swelling up
shape or wrinkled shape
in the firing process.
(128)
-it influences the
resistance
B
-adjust temperature
B
-Check machine
-Check roller kiln
C
-Inform press section to
solve those problems.
190
-loss control in firing
process
Packaging
(3)
The
overlappi
ng
tiles/chip
ped Tiles
(X)
-Machine problem (c)
-human factor error
(k)
-Check when mobile
into the box (ww)
problems of windy holes
at the back of tiles.(41)
-it detects in row body
checking process) (37)
-the tiles are not
symmetric, when it is
looked in each sides (54)
- the chipped is small
(126)
-the tiles surface is not
flat (58)
-there are number of
windy holes. (40)
-the air pressure not base
on the spec (129)
-Picture (130)
- Picture
- the tiles was
overlapping when mobile
into the box (131)
- the chipped is small
(126)
B
-Packaging verification
-Check machine
191
Ceramic Horse Organization Chart
192
Usability Testing Form
Application Name
:
Diagnostic System for Manufacturing Process Problems
(DSMPP)
Company Name
:
White Horse Ceramics
Category User
:
1.
Admin
2.
Knowledge Engineer
3.
Common User (Staff)
System Performance Rating :
5 – Very Good
4 – Good
3 – Fair
2 – Poor
1 – Very Poor
System Interface:
1.
Is the DSMPP system interface interesting?
2.
Is the DSMPP system easy to use?
3.
Does the DSMPP system produce the valid information for each
process?
4.
Does the DSMPP system respond at each stage of error
processing?
5.
Does the error message that is produced by the system can help
the user?
6.
Does the interface of DSMPP system fulfill the user
requirement?
193
System Validation
7.
Does the system produce the correct answer for each diagnostic
tile process?
8.
Does the database system up-to-date?
9.
Is the system root processing valid?
10.
Is the dialogues in term of questioning and answering form that
displays in the diagnose module correct?
11.
Is the overall system procedure valid?
Security
12.
Is the system security performance good?
13.
Has the overall database system manipulated by the authority
user?
14.
Has the database system been built regards to the right
knowledge base system development?
Management Company
15.
Does this system help the overall management in the company?
16.
Does this system help and implement the information sharing or
knowledge sharing in the company?
17.
Can this DSMPP system reduce the tiles defect in manufacturing
firm?
18.
Does this system implicitly help to increase the company profit?
19.
Does the system help the employee to detect and diagnose the
tiles defect?
20.
Can this system help the company to manage their knowledge?
21.
Can this system help the expert to manage and share t h e ir
knowledge?
194
The defects: Pin Hole (I)
R1
: IF The defects are on the surface of tiles (37) AND (There is small hole on the
tiles surface, commonly on the edge of tile (5) OR the small hole in the graphics
design of tiles (it will appear clearly before firing process) (6)) THEN Check
temperature (b).
R2
: IF Check temperature (b) AND Temperature
80 – 100 degree Celsius (3)
AND Check machine speed (not base on the specification) (4) THEN check the
water spraying (a)
R3
: IF The defects are on the surface of tiles (37) AND the small hole in the
graphics design of tiles (it will appear clearly before firing process) (6) THEN
check the firing inspection report (d)
R4
: IF check the firing inspection report (7) AND any defects were found in firing
inspection report (136) THEN Check the water spraying (a)
R5
: IF Check the water spraying (a) AND Water spraying weight less than spec (1)
AND Nozzle is clogged up 99) AND Picture ( 8) THEN Pin Hole [ I ] AND Sub
Process = Water Spraying (1) in Process =Glazing (1)
The normalization for the Pin Hole:
R1
: IF The defects are on the surface of tiles (37) AND (There is small hole on the
tiles surface, commonly on the edge of tile (5) AND Temperature
80 – 100
degree Celsius (3) AND Check machine speed (not base on the specification) (4)
AND Water spraying weight less than spec (1) AND Nozzle is clogged up 99)
AND Picture ( 8) THEN Pin Hole [ I ] AND Sub Process = Water Spraying (1)
in Process =Glazing (1)
R2
: IF The defects are on the surface of tiles (37) AND the small hole in the
graphics design of tiles (it will appear clearly before firing process) (6)) AND
any defects were found in firing inspection report (136) AND Water spraying
195
weight less than spec (1) AND Nozzle is clogged up 99) AND Picture ( 8) THEN
Pin Hole [ I ] AND Sub Process = Water Spraying (1) in Process =Glazing (1)
The defects: Dipsink (II)
R6
: IF The defects are on the surface of tiles (37) AND defects were found in the
glaze of tiles (83) THEN Check Glaze level (e) OR Check viscosity of glaze
(g) OR Check weight of glaze (f) OR Check Density of glaze (h)
R7
: IF Check Glaze level (f) AND Glaze level less than spec (15) AND there are
some problems on the curvature reports (200) THEN Dip sink [II] AND Sub
Process is Water Spraying (1) AND Process glazing (1)
R8
: IF Check viscosity of glaze (g) AND (Viscosity less than specification (9)
OR Viscosity more than spec (10)) ) AND There are some problems on the
curvature reports (200) THEN Dip sink [II] AND Sub Process is Water Spraying
(1) AND Process= glazing (1)
R9
: IF check weight of glaze (f) AND Glaze weight less than spec (12) OR Glaze
weight more than spec (14) ) AND There are some problems on the curvature
reports (200) THEN Dip sink [II] AND Sub Process is Water Spraying (1) AND
Process=glazing (1)
R10
: IF Check Density of glaze (h) AND Density less than spec (11) OR Density
more than spec (13) ) AND There are some problems on the curvature reports
(200)THEN Dip sinks [II] AND Sub Process is Water Spraying (1) AND
Process= glazing (1)
R11
: IF the defects were found behind of the tiles (42) AND there are number of
windy holes. (40) AND it influences the problems of windy holes at the back of
tiles. (41) ) THEN check aluminums slip condition (m)
R12
: IF check aluminums slip condition (m) AND No Aluminums slip.(39)
THEN Dip sink [II] AND Sub Process is Aluminum Slip (3) AND Process=
Glazing (1)
196
The normalization for the Dipsink:
R1
: IF The defects are on the surface of tiles (37) AND defects were found in the
glaze of tiles (83) AND Glaze level less than spec (15) THEN Dip sink [II] AND
Sub Process is Water Spraying (1) AND Process glazing (1)
R2
: IF The defects are on the surface of tiles (37) AND defects were found in the
glaze of tiles (83) AND (Viscosity less than specification (9) THEN Dip sink [II]
AND Sub Process is Water Spraying (1) AND Process= glazing (1)
R3
: IF The defects are on the surface of tiles (37) AND defects were found in the
glaze of tiles (83) AND Glaze weight less than spec (12) THEN Dip sink [II]
AND Sub Process is Water Spraying (1) AND Process=glazing (1)
R4
: IF The defects are on the surface of tiles (37) AND defects were found in the
glaze of tiles (83) AND Density less than spec (11) THEN Dip sinks [II] AND
Sub Process is Water Spraying (1) AND Process= glazing (1)
R5
: IF the defects were found behind of the tiles (42) AND there are number of
windy holes. (40) AND it influences the problems of windy holes at the back of
tiles. AND No Aluminums slip.(39) THEN Dip sink [II] AND Sub Process is
Aluminum Slip (3) AND Process= Glazing (1)
The defects: Tiles Dust (III)
R13
: IF The defects are on the surface of tiles (37) AND There is some dust on the
tiles surface (19) AND When handy touched, the surface of tiles are covered
with bumps or bruised.. (20) AND when it looked by naked eye for the distance
around 1 meter, there is some hump or outgrowth on the tiles surface. (21)
THEN Check the sieve condition (i)
R14
: IF Check the sieve condition (i) AND The sieve has been broken (16)
OR Mesh is worn out (17) THEN Check the sieve maintenance (j)
197
R15
: IF Check the sieve maintenance (j) AND loss control in sieve maintenance (33)
AND (Picture (23) OR Picture (24) OR Picture (25)) THEN Tiles Dust [ III ]
AND Sub Process is Sieve(2) AND Process = glazing (1)
The normalization for the Tiles Dust:
R1
: IF The defects are on the surface of tiles (37) AND There is some dust on the
tiles surface (19) AND When handy touched, the surface of tiles are covered
with bumps or bruised.. (20) AND when it looked by naked eye for the distance
around 1 meter, there is some hump or outgrowth on the tiles surface. (21) AND
The sieve has been broken (16) AND loss control in sieve maintenance (33)
AND (Picture (23) THEN Tiles Dust [ III ] AND Sub Process is Sieve(2) AND
Process = glazing (1)
R2
: IF The defects are on the surface of tiles (37) AND There is some dust on the
tiles surface (19) AND When handy touched, the surface of tiles are covered
with bumps or bruised.. (20) AND when it looked by naked eye for the distance
around 1 meter, there is some hump or outgrowth on the tiles surface. (21) Mesh
is worn out (17) AND loss control in sieve maintenance (33) AND (Picture (23)
THEN Tiles Dust [ III ] AND Sub Process is Sieve(2) AND Process = glazing
(1)
198
Use case name : Login
Primary Actor : User,
Knowledge Engineer
Stakeholders and Interest :
ID : 1 Importance Level : High
Admin, Use case type : Detail, essential
User- Login the system by using their own id
Admin- Login before entry the information needed by the system
Knowledge Engineer-Login the system by using their own id for security
Brief Description - This use case describes the login process for all the users.
Trigger : The users entry their id and password to using the system
Type : Internal
Relationships :
Association : User, Knowledge Engineer, Admin
Include : Extend : Generalization : Normal Flow of Events :
1. The users insert their id and password.
2. if their id and password has been recorded in database, the user can access the
system, else the user can not use the system,
3. The login looping process was limited with three times execution.
Sub flows: Alternative/Exceptional Flows : Figure 4.15. Use Case Description for Login
199
Use case name : Logout
Primary Actor : User,
Knowledge Engineer
Stakeholders and Interest :
ID : 2 Importance Level : High
Admin, Use case type : Detail, essential
User- Logout the system.
Admin- Logout after access the system
Knowledge Engineer-Login the system.
Brief Description - This use case describes the logout process for all the users.
Trigger : The users exit the system
Type : Internal
Relationships :
Association : User, Knowledge Engineer, Admin
Include : Extend : Generalization : Normal Flow of Events :
1. After finishing access the system, the users will logout the system. It is
more for the security.
2. After Log out, the user back to main menu.
Sub flows: Alternative/Exceptional Flows : Figure 4.16: Use Case Description for Logout
200
Use case name : Searching
ID : 3
Importance Level : High
Primary Actor : Knowledge Engineer, Use case type : Detail, essential
Admin
Stakeholders and Interest :
Knowledge engineer and admin are using this page to get or search some data in
database by inserting the keyword
Brief Description - This use case provides the searching process services where the
Knowledge engineer and admin can search any information in database system
Trigger: The Knowledge Engineer and admin type some keyword on this page
Type : Internal
Relationships :
Association : Knowledge engineer, Admin
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
The Knowledge engineer browse his page
The Admin browse his page
The searching page will be displayed
They type the keyword, example is “glaze”
The system will display all information in database that relates to the word
“glaze”
Sub flows: Alternative/Exceptional Flows :
Figure 4.17: Use Case Description for Searching
Use case name : EditProduct
ID : 4
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the tiles information
Brief Description - This use case describes how the admin and KE can edit, delete and
add the tile product regarding to the information of tiles, such as:
201
id+name+image+information+prices (as shown in Class Diagram, tiles table)
Trigger : insert, edit , delete the tiles information
Type : Internal
Relationships :
Association : Admin, KE
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the product information
They edit, insert, delete the product information
The system validates all the action
The system will display the updated data
The tiles database will be changed
Sub flows: Alternative/Exceptional Flows : -
Figure 4.18: Use Case Description for EditProduct
Use case name : EditSymptom
ID : 5
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the symptom
Brief Description - This use case describes how the admin and KE can edit, delete and
add the tile product regarding to the symptom
Trigger : insert, edit , delete the tiles information
Type : Internal
Relationships :
Association : Admin, KE
202
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the symptom
They edit, insert, delete the symptom
The system validates all the action
The system will display the updated data
The symptom database will be changed
Sub flows: Alternative/Exceptional Flows : Figure 4.19: Use Case Description for EditSymptom
Use case name : EditDefect
ID : 6
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the defect
Brief Description - This use case describes how the admin and KE can edit, delete and
add the defect
Trigger : insert, edit , delete the defect
Type : Internal
Relationships :
Association : Admin, KE
Include : Extend : Generalization : Normal Flow of Events :
1. Admin and Knowledge Engineer browse their own page
203
2.
3.
4.
5.
6.
They manage the defect information
They edit, insert, delete the defect
The system validates all the action
The system will display the updated data
The defect database will be changed
Sub flows: Alternative/Exceptional Flows : Figure 4.20: Use Case Description for EditDefect
Use case name : EditRec
ID : 7
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest : Admin and Knowledge engineer– edit, delete, insert the
recommendation
Brief Description - This use case describes how the admin and KE can edit, delete and
add the recommendation for each defect found
Trigger : insert, edit , delete the recommendation
Type : Internal
Relationships :
Association : Admin, KE
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the recommendation
They edit, insert, delete the recommendation
The system validates all the action
The system will display the updated data
The recommendation database will be changed
Sub flows: Alternative/Exceptional Flows : Figure 4.21: Use Case Description for EditRec
204
Use case name : EditCauses
ID : 8
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the causes
Brief Description - This use case describes how the admin and KE can edit, delete and
add the causes of defects
Trigger : insert, edit , delete the causes
Type : Internal
Relationships :
Association : Admin, KE
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the causes of defects
They edit, insert, delete the causes
The system validates all the action
The system will display the updated data
The causes database will be changed
Sub flows: Alternative/Exceptional Flows : Figure 4.22: Use Case Description for EditCauses
Use case name : EditProcess
ID : 9
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the process
Brief Description - This use case describes how the admin and KE can edit, delete and
add the process in manufacturing that probable causes the defect
Trigger : insert, edit , delete the causes
205
Type : Internal
Relationships :
Association : Admin, KE
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the process of defects
They edit, insert, delete the process
The system validates all the action
The system will display the updated data
The process database will be changed
Sub flows: Alternative/Exceptional Flows : Figure 4.23: Use Case Description for EditProcess
Use case name : EditSubProcess
ID : 10
Importance Level : High
Primary Actor : Admin, K n o wledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the SubProcess
Brief Description - This use case describes how the admin and KE can edit, delete and
add the SubProcess in manufacturing form that causes the defects
Trigger : insert, edit , delete the SubProcess
Type : Internal
Relationships :
Association : Admin, KE
Include : Extend : -
206
Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the subprocess of defects
They edit, insert, delete the sub process
The system validates all the action
The system will display the updated data
The sub process database will be changed
Sub flows: Alternative/Exceptional Flows : Figure 4.24: Use Case Description for EditSubProcess
Use case name : EditRuleBase
ID : 11
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the rulebase
Brief Description - This use case describes how the admin and KE can edit, delete and
add the rule base
Trigger : insert, edit , delete the rulebase
Type : Internal
Relationships :
Association : Admin, KE
Include : Diagnose
Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the rule base
They edit, insert, delete the causes
The system validates all the action
The system will display the updated data
The causes database will be changed
207
Sub flows: 2.a. Inserts amount of rules that they want to add or insert
b. Types or inserts their rules base on the rulebase pattern
c. Looping until all the rule bases have been added or inserted.
Alternative/Exceptional Flows : Figure 4.25: Use Case Description for EditRuleBase
Use case name : EditQuestion
ID : 12
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the question
Brief Description - This use case describes how the admin and KE can edit, delete and
add the questions that will be used to test the diagnostics system.
Trigger : insert, edit , delete the question
Type : Internal
Relationships :
Association : Admin, KE
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the question
They edit, insert, delete the question
The system validates all the action
The system will display the updated data
The question database will be changed
Sub flows: Alternative/Exceptional Flows : Figure 4.26: Use Case Description for EditQuestion
Use case name : EditAnswer
ID : 13
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
208
Engineer
Stakeholders and Interest :
Admin and Knowledge engineer– edit, delete, insert the answer
Brief Description - This use case describes how the admin and KE can edit, delete and
add the answer from the dialogues page.
Trigger : insert, edit , delete the answer
Type : Internal
Relationships :
Association : Admin, KE
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
5.
6.
Admin and Knowledge Engineer browse their own page
They manage the answer
They edit, insert, delete the answer
The system validates all the action
The system will display the updated data
The answer database will be changed
Sub flows: Alternative/Exceptional Flows : Figure 4.27: Use Case Description for EditAnswer
Use case name : ViewSymptom
ID : 14
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer, User
Stakeholders and Interest :
Admin, User and Knowledge Engineer– browse the symptom
Brief Description - This use case describes how the admin, user and knowledge engineer
can browse or view the symptom
Trigger : admin, user and knowledge engineer are viewing and searching the symptom
Type : Internal
209
Relationships :
Association : Admin, Knowledge Engineer,User
Include : LoadDefect
Extend : Generalization : Normal Flow of Events :
1. Admin, User and Knowledge Engineer lookup the symptom on their own
page
2. The web site will display the symptom list.
3. Admin, User and Knowledge retrieves the symptom
4. The system will display the symptom base on the defect found (defect
chosen)
Sub flows: Alternative/Exceptional Flows : Figure 4.28: Use Case Description for ViewSymptom
Use case name : ViewProduct
ID : 15
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer, User
Stakeholders and Interest :
Admin and Knowledge Engineer– browse the product
Brief Description - This use case describes how the admin and knowledge engineer can
browse or view the product
Trigger : admin and knowledge engineer are viewing and searching the product
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer
Include : Extend : Generalization : Normal Flow of Events :
210
1. Admin and Knowledge Engineer lookup the product on their own page
2. The web site will display the product list.
3. Admin, User and Knowledge retrieves the product
Sub flows: Alternative/Exceptional Flows : Figure 4.29: Use Case Description for ViewProduct
Use case name : ViewCauses
ID : 16
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer, User
Stakeholders and Interest :
Admin, User and Knowledge Engineer– browse the causes
Brief Description - This use case describes how the admin, user and knowledge engineer
can browse or view the causes of defects
Trigger : admin, user and knowledge engineer are viewing and searching the causes
informtion
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer, User
Include : LoadDefect
Extend : Generalization : Normal Flow of Events :
1
2
3
4
Admin, User and Knowledge Engineer lookup the causes on their own
page
The web site will display the causes list.
Admin, User and Knowledge retrieves the causes
The system will display the causes base on the defect found (defect
chosen)
Sub flows: Alternative/Exceptional Flows : Figure 4.30 Use Case Descriptions for ViewCauses
211
Use case name : ViewProcess
ID : 17
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer, User
Stakeholders and Interest :
Admin, User and Knowledge Engineer– browse the process
Brief Description - This use case describes how the admin, user and knowledge engineer
can browse or view the process
Trigger : admin, user and knowledge engineer are viewing and searching the process
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer, User
Include : LoadDefect
Extend : Generalization : Normal Flow of Events :
1
2
3
4
Admin, User and Knowledge Engineer lookup the causes on their own
page
The web site will display the causes list.
Admin, User and Knowledge retrieves the causes
The system will display the causes base on the defect found (defect
chosen)
Sub flows: Alternative/Exceptional Flows : Figure 4.31 Use Case Descriptions for ViewProcess
Use case name : ViewSubProcess
ID : 18
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer, User
Stakeholders and Interest :
Admin, User and Knowledge Engineer– browse the SubProcess
Brief Description - This use case describes how the admin, user and knowledge engineer
can browse or view the sub process
212
Trigger : admin, user and knowledge engineer are viewing and searching the sub process
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer,User
Include : LoadDefect
Extend : Generalization : Normal Flow of Events :
1. Admin, User and Knowledge Engineer lookup the sub process on their
own page
2. The web site will display the sub process list.
3. Admin, User and Knowledge retrieves the sub process
4. The system will display the sub process base on the defect found (defect
chosen)
Sub flows: Alternative/Exceptional Flows : Figure 4.32 Use Case Descriptions for ViewSubProcess
Use case name : ViewRec
ID : 19
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer, User
Stakeholders and Interest :
Admin, User and Knowledge Engineer– browse the recommandation
Brief Description - This use case describes how the admin, user and knowledge engineer
can browse or view the recommendation list and the recommendation base on the defect
chosen.
Trigger : admin, user and knowledge engineer are v i e w ing and searching the
recommendation
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer,User
213
Include : LoadDefect
Extend : Generalization : Normal Flow of Events :
1. Admin, User and Knowledge Engineer lookup the recommendation on
their own page
2. The web site will display the recommendation list.
3. Admin, User and Knowledge retrieves the recommendation
4. The system will display the recommendation base on the defect found
(defect chosen)
Sub flows: Alternative/Exceptional Flows : Figure 4.33 Use Case Descriptions for ViewRec
Use case name : ViewAnswer
ID : 20
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge Engineer– browse the Answer
Brief Description - This use case describes how the admin and knowledge engineer can
browse or view the Answer
Trigger : admin and knowledge engineer are viewing and searching the Answer
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer
Include : LoadQ_A
Extend : Generalization : Normal Flow of Events :
1. Admin and Knowledge Engineer lookup the answer on their own page
2. The web site will display the answer list.
3. Admin and Knowledge retrieves the answer
214
4. The system will display the answer base on question chosen.
Sub flows: Alternative/Exceptional Flows : Figure 4.34 Use Case Descriptions for ViewAnswer
Use case name : ViewQuestion
ID : 21
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge Engineer– browse the Question
Brief Description - This use case describes how the admin and knowledge engineer can
browse or view the Question
Trigger : admin and knowledge engineer are viewing and searching the Question
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer
Include : LoadQ_A
Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
Admin and Knowledge Engineer lookup the question on their own page
The web site will display the question list.
Admin and Knowledge retrieves the question
The system will display the question base on answer chosen.
Sub flows: Alternative/Exceptional Flows : Figure 4.35 Use Case Descriptions for ViewQuestion
Use case name : ViewDefects
ID : 22
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
215
Admin and Knowledge Engineer– browse the Defects
Brief Description - This use case describes how the admin and knowledge engineer can
browse or view the Defects of tiles
Trigger : admin and knowledge engineer are viewing and searching the Defect
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer
Include : Extend : Generalization : Normal Flow of Events :
1.
2.
3.
4.
Admin and Knowledge Engineer lookup the defects on their own page
The web site will display the defects list.
Admin and Knowledge retrieves the defects
The system will display the defect base on the grade, the symptom, the
process, the sub process, the causes and the recommendation.
Sub flows: Alternative/Exceptional Flows : Figure 4.36 Use Case Descriptions for ViewDefects
Use case name : ViewAll
Primary Actor : User
ID : 23
Importance Level : High
Use case type : Detail, essential
Stakeholders and Interest :
User– browse the whole information
Brief Description - This use case describes how the use can browse or view the entire
information, such as the defects, the symptom, the causes, the process, the sub process
and the recommendation.
Trigger : users are viewing and searching the entire information
Type : Internal
216
Relationships :
Association : User
Include : Extend : Generalization : Normal Flow of Events :
1. User looks up the All information on their own page
2. The web site will display the information list, such as the defect, the
symptom, the causes, the process, the sub process and the
recommendation.
3. User retrieves the All information
4. The system will display the all information.
Sub flows: Alternative/Exceptional Flows : Figure 4.37 Use Case Descriptions for ViewAll
Use case name : ViewtheResult
ID : 24
Importance Level : High
Primary Actor : User
Use case type : Detail, essential
Stakeholders and Interest : Quality control/Quality Assurance/R&D – After Using the
diagnose page the system will provide the result to solve the problems
Brief Description – This use case describes how the system gives some solving problems
regarding to the defects and probable cause found on the dialogue page.(diagnose use
case)
Trigger : subscribe the result information
Type : Internal
Relationships :
Association : user
Include : Extend : diagnose
Generalization : Normal Flow of Events :
217
1. The user look up the result on the web sites
2. The system provides the result regarding to the diagnostic conclusion
from the diagnoses use case
While the dialogue page has been finished (the result button has been
checked)
If the all rule base procedure has been tracing then
The defect is proposed (ex. The defect=”crack”)
Probable cause is proposed, the recommendation, the process, the
sub process, the grade are proposed
3. The user retrieve the result
4. If the user want to print the result, just choose print button to print out
Sub flows:
Alternative/Exceptional Flows : Figure 4.38 Use Case Descriptions for ViewTheResult
Use case name : View CompanyInfo
ID : 25
Importance Level : High
Primary Actor : User
Use case type : Detail, essential
Stakeholders and Interest : Quality control/Quality Assurance/R&D (User)– can get all
information about the company
Brief Description – This use case describes how the user can browse the company
information
Trigger : view the company information
Type : Internal
Relationships :
Association : user
Include : Extend : Generalization : Normal Flow of Events :
1. The user look up the company information on the web sites
218
2. The web site will display the company information, such as the founder,
their company profiles, the location and map, etc.
3. User retrieves the Company information
4. The system will display company information
Sub flows:
Alternative/Exceptional Flows : Figure 4.39 Use Case Descriptions for ViewCompanyInfo
Use case name : ViewListAll
ID : 26
Importance Level : High
Primary Actor : Admin, Knowledge Use case type : Detail, essential
Engineer
Stakeholders and Interest :
Admin and Knowledge Engineer– browse the All information
Brief Description - This use case describes how the admin and knowledge engineer can
browse or view the All Information
Trigger : admin and knowledge engineer are viewing and searching the All information
Type : Internal
Relationships :
Association : Admin, Knowledge Engineer
Include : Extend : Generalization : Normal Flow of Events :
1. Admin and Knowledge Engineer lookup the all information on their own
page
2. The web site will display the information list, such as the defect, the
causes, the symptom, the process, the sub process and the
recommendation.
3. Admin and Knowledge retrieves the All information
4. The system will display the entire information.
Sub flows: Alternative/Exceptional Flows : Figure 4.40 Use Case Descriptions for ViewListAll
219
Use case name : ViewRuleBase
Primary Actor : Knowledge engineer
Stakeholders and Interest :
ID : 27
Importance Level : High
Use case type : Detail, essential
Admin, Knowledge engineer – view the rule base
Brief Description – This use case describes how the system gives the rule base
information to the knowledge engineer and admin, it helps the KE and admin to manage
the rule base
Trigger : subscribe the rule base information
Type : Internal
Relationships :
Association : knowledge engineer, admin
Include : Extend : Generalization : Normal Flow of Events :
1. The KE lookup the rule base on the web sites
2. The web site displays the rule base list.
3. The KE retrieves the rule base.
Sub flows: Alternative/Exceptional Flows : Figure 4.41 Use Case Descriptions for ViewRuleBase
Use case name : LoadDefect
ID : 28
Importance Level : High
Primary Actor : Knowledge engineer, Use case type : Detail, essential
Admin,User
Stakeholders and Interest :
Knowledge engineer, Admin, User – load the defect base on the input
Brief Description – This use case describes how the system provides the defect loaded
regarding the symptom, the causes, the process, the sub process and the recommendation
chosen
Trigger : load the defect
Type : Internal
Relationships :
Association : knowledge engineer,admin,user
Include : ViewSymptom,ViewCauses, ViewProcess, ViewSubProcess, ViewRec
Extend : -
220
Generalization : Normal Flow of Events :
1. The Admin, KE and User browse their own site
If they browse the symptom page then
Choose the symptom and select the defect base on the symptom
If they browse the process page then
Choose the process and select the defect base on the process
If they browse the Sub Process page then
Choose the Sub Process and select the defect base on the Sub Process
If they browse the causes page then
Choose the causes and select the defect base on the causes
If they browse the recommendation page then
Choose the recommendation and select the defect base on the
recommendation
2. The web site displays
subprocess_def, rec_def
the sym_def, causes_def, process_def,
4. The KE, Admin and user retrieves the defect
Sub flows: Alternative/Exceptional Flows : Figure 4.42 Use Case Descriptions for LoadDefect
221
Use case name : Diagnose
Primary Actor : User
Stakeholders and Interest :
ID : 29
Importance Level : High
Use case type : Detail, essential
Quality control/Quality Assurance/R&D (User) – Using the web sites in dialogue pages
to diagnose the defect problems.
Brief Description - This use case describes how the system solves the defect problems
found by the user (the users could be QC, QA or R&D personnel). The system provides,
the page where the user can input the symptom they found, the dialogue format in term
of question and answer and combination of it. The system will produce the result and
recommendation to solve the defect problems.
Trigger: The users input the symptom and answer the questions that provided by the
systems.
Type : Internal
Relationships :
Association : User
Include : EditRulebase
Extend : ViewResult
Generalization : Normal Flow of Events :
1. The user browses the diagnose page in the web sites.
2. The system will provide the pages with all information that relates to the defects,
such as from causes view, process view, sub process view, symptom view and
recommendation view.
3. The User can input any symptom that they found from the tiles defect
If the system can conclude the result base on the user input Then
The result will be displayed
Else
The User must answer the questions
4. The User answers the questions.
5. After finishing the answer and question dialogue, the system will conclude and
222
refer to knowledge base rules.
while the users answer the question (by checking the button) do
if the answer1=”user_answers” then check the probable_causes_for _answer1
and proposed the question2
if the answer2=”user_answer” then check the probable_causes_for answer2
and proposed question3…(etc)
Tracing base on the rule base, continue until the result found.
6. The system will provide the result that comes from the knowledge rule base and
inference engines.
7. The system also provides the recommendations regards to the defect found as the
solving problems.
While the defect_found do
If defect=”Crack” then
The process that causes the defect will be proposed
The sub process that causes the defect will be proposed
The recommendation for that defect was proposed.
The classification was proposed
8. All the information obtained by the user can be printed.
Sub flows: -
Alternative/Exceptional Flows :
Figure 4.43 Use Case Descriptions for Diagnose
223
Use case name : LoadQ_A
ID : 30
Importance Level : High
Primary Actor : Knowledge engineer, Use case type : Detail, essential
Admin
Stakeholders and Interest :
Knowledge engineer, Admin– load the question base on the answer, and load the answer
base on the question
Brief Description – This use case describes how the system provides the Q_A loaded
regarding the user requirement
Trigger : load the question and answer
Type : Internal
Relationships :
Association : knowledge engineer,admin
Include : ViewQuestion, ViewAnswer
Extend : Generalization : Normal Flow of Events :
1. The Admin, KE browse their own site
If they browse the question page then
Choose the question and select the answer base on the question
If they browse the answer page then
Choose the answer and select the question base on the answer
3. The web site displays the q_a information
5. The KE, Admin and user retrieves the question and answer
Sub flows: Alternative/Exceptional Flows : Figure 4.44 Use Case Descriptions for LoadQ_A
224
Front:
Class Name: Admin
ID: 1
Type: Concrete, Domain
Description: The person who can entry the Associates Use Cases:
information needed to diagnose
the defects problems
Responsibilities
Collaborators
EditProduct
Tiles
EditSymptom
Symptom
EditDefect
Defect
EditRec
Rec
EditGrade
Grade
EditProcess
Web sites
EditCauses
Causes
EditSubProcess
Process
EditAnswer
SubProcess
EditQuestion
Question
EditRuleBase
Answer
Searching
ListAll
ViewProduct
ViewSymptom
ViewDefect
ViewCauses
ViewProcess
ViewSubProcess
ViewAnswer
ViewQuestion
ViewRec
Back:
Attributes
id(text)
password (text)
Relationships:
Generalization (a-kind-of): users
Aggregation (has-parts): Other associations: -
Figure 4.45: CRC cards for Admin
225
Front:
Class Name: User
ID: 2
Type: Concrete, Domain
Description: The person who can access the Associates Use Cases:
system and diagnose the defects
problems (could be QA,QC, R&D,
Admin or Knowledge Engineer)
Responsibilities
Collaborators
Load Diagnose
Web Sites
Load Result
Load Defect
ViewAll
ViewProcess
ViewSubProcess
ViewRec
ViewSymptom
ViewCauses
ViewCompanyInfo
Back:
Attributes
id(text)
password (text)
Relationships:
Generalization (a-kind-of): users
Aggregation (has-parts): Other associations:
Figure 4.46: CRC cards for User
Front:
Class Name: Knowledge engineer ID: 3
Type: Concrete, Domain
Description: The person who can manage the Associates Use Cases:
rule base engines and edit the
knowledge base
Responsibilities
Collaborators
EditSymptom
Tiles
EditDefect
Symptom
EditRec
Defect
EditGrade
Rec
EditProcess
Grade
EditCauses
Web sites
EditSubProcess
Causes
EditAnswer
Process
226
EditQuestion
EditRuleBase
Searching
ListAll
ViewProduct
ViewSymptom
ViewDefect
ViewCauses
ViewProcess
ViewSubProcess
ViewAnswer
ViewQuestion
ViewRec
SubProcess
Question
Answer
Back:
Attributes
id(text)
password (text)
Relationships:
Generalization (a-kind-of): users
Aggregation (has-parts): Other associations:
Figure 4.47: CRC cards for Knowledge Engineer
Front:
Class Name: Rule Base
ID: 4
Type: Concrete, Domain
Description: The inference engines to trigger Associates Use Cases:
the rules for solving the defects
problems
Responsibilities
Collaborators
Manage Rule Base
Defect
Result
RuleBase
Knowledge Engineer
Admin
Back:
Attributes
Id_rulebase(int)
Id(int)
Rules(varchar)
227
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.48: CRC cards for Rule Base
Front:
Class Name: Tiles
ID: 5
Type: Concrete, Domain
Description: The Information of the product.
Associates Use Cases:
Responsibilities
Collaborators
Record all the product information
Admin
Knowledge Engineer
Back:
Attributes
Id_Tiles(int)
Name_Tiles(vachar)
information(varchar)
price(varchar)
image(jpeg)
veryfiedby(varchar)
date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations: Figure 4.49: CRC cards for Tiles
Front:
Class Name: Symptom
ID: 6
Type: Concrete, Domain
Description: The Information of the symptom Associates Use Cases:
and probable causes of the defects
problem
Responsibilities
Collaborators
Record the symptom
Admin
Knowledge Engineer
Defect
Back:
Attributes
228
Id_Sym(int)
Name_Sym (varchar)
Veryfiedby(varchar)
Date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.50: CRC cards for Symptom
Front:
Class Name: Defect
ID: 7
Type: Concrete, Domain
Description: The Information of the defects Associates Use Cases:
problem
Responsibilities
Collaborators
Record the defects
Admin
Knowledge Engineer
Rule base
Defect
Symptom
Causes
Process
SubProcess
Rec
Back:
Attributes
Id_Def(int)
Name_Def (varchar)
Veryfiedby(varchar)
Date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.51: CRC cards for Defect
Front:
Class Name: Causes
ID: 8
Type: Concrete, Domain
Description: The Information of and probable Associates Use Cases:
causes of the defects problem
229
Responsibilities
Record the causes
Collaborators
Admin
Knowledge Engineer
Defect
Back:
Attributes
Id_causes(int)
Name_causes (varchar)
Veryfiedby(varchar)
Date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.52: CRC cards for Causes
Front:
Class Name: Process
ID: 9
Type: Concrete, Domain
Description: The Information of the process in Associates Use Cases:
manufacturing firm that causes the
defect
Responsibilities
Collaborators
Record the process
Admin
Knowledge Engineer
Defect
Back:
Attributes
Id_process(int)
Name_process (varchar)
Veryfiedby(varchar)
Date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.53: CRC cards for Process
230
Front:
Class Name: SubProcess
ID: 10
Type: Concrete, Domain
Description: The Information of the SubProcess Associates Use Cases:
in manufacturing firm that causes
the defects or problems
Responsibilities
Collaborators
Record the subprocess
Admin
Knowledge Engineer
Defect
Back:
Attributes
Id_subproc(int)
Name_subproc(varchar)
Veryfiedby(varchar)
Date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.54: CRC cards for SubProcess
Front:
Class Name: Question
ID: 11
Type: Concrete, Domain
Description: The Information of the question Associates Use Cases:
that will be used for diagnostics
dialogue
Responsibilities
Collaborators
Record the question
Admin
Knowledge Engineer
Answer
Back:
Attributes
Id_question(int)
Name_question (varchar)
Veryfiedby(varchar)
Date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.55: CRC cards for Question
231
Front:
Class Name: Answer
ID: 12
Type: Concrete, Domain
Description: The Information of the answer that Associates Use Cases:
will be used for diagnostics
dialogue
Responsibilities
Collaborators
Record the answer
Admin
Knowledge Engineer
Question
Back:
Attributes
Id_answer(int)
Name_answer (varchar)
Veryfiedby(varchar)
Date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.56: CRC cards for Answer
Front:
Class Name: Rec
ID: 13
Type: Concrete, Domain
Description: The
Information
of
the Associates Use Cases:
Recommendation to solve the
defects
Responsibilities
Collaborators
Record the recommendations
Admin
Knowledge Engineer
Defect
Back:
Attributes
Id_Rec(int)
Name_Rec (varchar)
Veryfiedby(varchar)
Date(datestamp)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.57: CRC cards for Rec
232
Front:
Class Name: Web Sites
ID: 14
Type: Concrete, Domain
Description: System Interface (web page menu) Associates Use Cases:
Responsibilities
Collaborators
Login
User
Logout
Admin
View
Knowledge Engineer
Result
Back:
Attributes
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.58: CRC cards for Websites
Front:
Class Name: Result
ID: 15
Type: Concrete, Domain
Description: The Result of diagnostic system
Associates Use Cases:
Responsibilities
Collaborators
Give Result
Rule base
Load Diagnose
Web sites
Question
Back:
Attributes
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.59: CRC cards for Result
Front:
Class Name: Users
ID: 16
Type: Concrete, Domain
Description: The person who can access the Associates Use Cases:
system and diagnose the defects
problems (could be QA,QC, R&D,
Admin or Knowledge Engineer)
Responsibilities
Collaborators
Back:
Attributes
-
233
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Admin, User and Knowledge Engineer
Other associations:
Figure 4.60: CRC cards for Users
Front:
Class Name: Grade
ID: 17
Type: Concrete, Domain
Description: The Information of Grade and Associates Use Cases:
qualification of product defect
Responsibilities
Collaborators
Record the grade
Admin
Knowledge Engineer
Defect
Back:
Attributes
Id_grade(int)
Name_grade (varchar)
Relationships:
Generalization (a-kind-of): Aggregation (has-parts): Other associations:
Figure 4.61: CRC cards for Grade
234
Table 4.3 admin table
Table 4.4 All table
Table 4.5 Answer table
Table 4.6 Cause_defect table
Table 4.7 Causes table
Table 4.8 def_grade table
235
Table 4.9 def_proc table
Table 4.10 def_subproc table
Table 4.11 def_sym table
Table 4.12 defect table
Table 4.13 grade table
Table 4.14 id_rulebase table
236
Table 4.15 process table
Table 4.16 q_a table
Table 4.17 question table
Table 4.18 rec table
Table 4.19 rec_def table
Table 4.20 rulebase table
237
Table 4.21 subprocess table
Table 4.22 symptom table
Table 4.23 tiles table
Table 4.24 tiles_type table
Table 4.25 rule_summery table
238
1.
Browse Internet Explorer, type: http://localhost/diagnose3/index.php. The system,
will display the main menu page as describes in this figure.
Figure 5.11 DSMPP main page
On this page the user can operate some function such as login process, diagnose
and company information.
2.
By using the login page, the admin and KE can do some manipulation on the
knowledge base, such as for manipulating the tiles production, symptom, defects,
rule base, causes, recommendation and all categories, as described in the figure
below:
Figure 5.12 .Login Page
239
Figure 5.13 .Admin/KE page after login
3.
By using this page, admin can add, update or even delete the information or data
about the tiles production. The tiles information also is listed from this page
Figure 5.14 .Admin/KE page for product
4.
By using this page, admin can add, update or even delete the information or data
about the symptom. The symptom information also is listed from this page.
240
Figure 5.15 .Admin/KE page for symptom
5.
By using the symptom page, the relationship between the symptom and defect can
be managed in this page.
Figure 5.16 .Admin/KE page for symptom-defect
241
6.
By using this page, admin can add, update or even delete the information or data
about the defects. The defects information also is listed from this page.
Figure 5.17 .Admin/KE page for defects
7.
By using the defect page, the relationship among the defect and the symptom, the
grade, the causes, the process, the sub process and the recommendation can be
managed in this page.
Figure 5.18 .Admin/KE page for defects-causes
242
8.
By using this page, admin can add, update or even delete the information or data
about the causes of defects. The causes information also are listed from this page
Figure 5.19 .Admin/KE page for causes
9.
By using the cause page, the relationship between the causes and defect can be
managed in this page.
Figure 5.20 Admin/KE page for causes-defect
243
10.
By using this page, admin can add, update or even delete the information or data
about the process. The process information also is listed from this page.
Figure 5.21 .Admin/KE page for process
11.
By using the process page, the relationship between the process and defect can be
managed in this page.
Figure 5.22 .Admin/KE page for process-defect
244
12.
By using this page, admin can add, update or even delete the information or data
about the sub process. The process information also is listed from this page
Figure 5.23 .Admin/KE page for sub process
13.
By using the process page, the relationship between the sub process and defect can
be managed in this page.
Figure 5.24 .Admin/KE page for sub process-defect
245
14.
By using this page, admin can add, update or even delete the information or data
about the recommendation. The recommendation also is listed from this page.
Figure 5.25 .Admin/KE page for the recommendation
15.
By using the process page, the relationship between the recommendation and
defect can be managed in this page.
Figure 5.26 .Admin/KE page for the recommendation-defect
246
16.
By using this page, admin can add, update or even delete the information or data
about the question. The question also is listed from this page.
Figure 5.27 .Admin/KE page for question
17.
By using the question page, the relationship between the question and answer can
be managed in this page.
Figure 5.28 .Admin/KE page for question-answer
247
18.
By using this page, admin can add, update or even delete the information or data
about the answer. The answer also is listed from this page.
Figure 5.29 .Admin/KE page for answer
19.
By using the answer page, the relationship between the question and answer can be
managed in this page.
Figure 5.30 .Admin/KE page for answer-question
248
20.
By using this page, admin and KE can manipulate, add, delete and update the
information or data about the rule base. The rule base information that has been
built in DSMPP system is listed from this page.
Figure 5.31 .Admin/KE page for rule base
21.
By using this page, admin and KE can list all information about the knowledge
base regards to each category. The page can be seen in the figure below:
Figure 5.32 .Admin/KE page for listAll
249
22.
By using this page, all users can get the company information, such as the
organization information, the founder, the office, the resident and the branches.
See this figure.
Figure 5.33 .Company Info Page
23.
By using this page, all users can diagnose the defect found from the tiles
production, the diagnostics was begun by inputting the symptom of the tiles defect,
and finally the system will provides the solution. See the figure below:
Figure 5.34 .Diagnose Page
24.
By using this page, the DSMPP system provides the solution of diagnostics
process for the tile defect, including the type of defect found, the process that
250
causes the problems, the sub process, the recommendation and the tile grade
classification.
Figure 5.35 .Result Page
25.
By using this page, all users can get the information about the causes, the
symptom, the recommendation, the process and the sub process regards to the
defect chosen. See figure below for the detail information.
Figure 5.36 .Diagnose-causes Page