customizing iso 9126 quality model for evaluation of b2 b applications
TRANSCRIPT
1/29
Customizing ISO 9126 Quality Model for
Evaluation of B2B Applications
Behshid Behkamal, Mohsen Kahani, Mohammad Kazem Akbari
Information and Software Technology
vol.51 pp.599–609, 2009.
報告者: 葉承宇 (Dean Yeh)
2/29
Abstract
• A software quality model acts as a framework for the evaluation of attributes of an application that contribute to the software quality.
• The most well-known quality models are studied, and reasons for using ISO 9126 quality model as the basis are discussed.
• The customization is done by extracting the quality factors from web applications and B2B EC applications, weighting these factors from the viewpoints of both developers and end users, and adding them to the model.
3/29
B2B Application
Customizing Quality Model
Quality Model (ISO 9126)
B2B Characteristics
Web Characteristics
Overview
4/29
Introduction
• Review of Quality Models
• B2B Electronic Commerce
• Need to customize ISO Model for B2B applications
• Our approach for customizing ISO Quality Model
• Applying the proposed model to a case study
• Evaluation & Discussion
• Conclusion & Future works
5/29
What is software quality?
http://www.cse.dcu.ie/essiscope/sm2/charact.html
6/29
McCall Model
http://www.cse.dcu.ie/essiscope/sm2/charact.html
7/29
Boehm Model
http://slashnode.wikidot.com/seng4420-lect117/30
8/29
FURPS Model
• FURPS
• Functionality
• Usability
• Reliability
• Performance
• Supportability
• +
• Design requirements
• Implementation requirements
• Interface requirements
• Physical requirements
Functional requirements
Non-functional requirements
IBM Software Group - Non-Functional Requirements
9/29
Dromey Model
Quality attributesProduct propertiesSoftware product
Implementation
CorrectnessFunctionality
Reliability
InternalMaintainability
Efficiency
ContextualReusability
Portability
Descriptive Usabilityhttps://export.writer.zoho.com/public/pakitosh/untitled1/fullpage
10/29
ISO Model
http://carballosa.blogspot.tw/2010/05/quint2-extended-iso-9126.html
11/29
Star Model
A Quality Model for Design Patterns
12/29
BBN Model
Data collection of Observation Bayesian belief network Diagram
http://zh.wikipedia.org/wiki/貝式網路
13/29
Comparison of software quality models
Quality model
StructureNumber of levels
Relationship Disadvantages Advantages
McCall Hierarchical Two Many to manyOverlapping of components
Having evaluation criteria
Boehm Hierarchical Three Many to many Lack of criteria Including factors related to hardware
FURPS Hierarchical Tow One to manyNot considering portability
Separating functional and non-functional requirements
Dromey Hierarchical Tow One to many Incomprehensiveness Applicable to different systems
ISO Hierarchical Three One to many Generality– Having evaluation criteria– Separating internal and external quality
StarNon-hierarchical
– Many to many Lack of criteria Considering different viewpoints
BBNNon-hierarchical
– Many to many Lack of criteria Having weighted quality factors
14/29
B2B Electronic Commerce
Benchmarking: An International Journal - Benchmarking business-to-business electronic commerce
15/29
Customize ISO model for B2B applications
Review of software quality
assessment methods
Expert Review
Software Metrics
Quality Model
16/29
Customizing ISO quality model
Step 1: Choosing ISO quality model as a basis.
Step 2: Identifying quality characteristics of a particular application.
Step 3: Choosing a group of software expert familiar with B2B ecommerce systems.
Step 4: Assigning weights to the quality factors and sub factors by experts.
Step 5: Developing the quality criteria.
17/29
Identifying quality characteristics of B2B applications
Web Application Quality
Quality factor Rank
Efficiency 1
Security 2
Usability 3
Traceability 4
Availability 5
Scalability 6
Functionality 7
Customizability 8
Recoverability 9
Consistency (Data) 10
The Quality of B2B Applications
Quality factor Total points Quality factor Total points
Security 5 Intractability 2
Scalability 5 Functionality 2
Efficiency 5 Customizability 2
Accessibility 5 Supportability 1
Traceability 3Open source platform
1
Reliability 3 Portability 1
Integrity 3 Generality 1
Usability 2 Changeability 1
Manageability 2 Compatibility 117
18/29
Applying the proposed model to a case study
First and second levels of proposed model with their weights
18
19/29
Analytical hierarchy process (AHP)
http://en.wikipedia.org/wiki/Analytic_hierarchy_process#Uses_and_applications
20/29
Development of quality criteria
Relationships between quality factors and quality criteria
20
21/29
Development of quality criteria
Relationships between quality factors and quality criteria
21
22/29
Applying the proposed model to a case study
Design and completion of the evaluation checklist
Calculating the values of the quality factors
Calculating the values of the model’s components
Calculating the final quality of system
Evaluation of ISACO portal
23/29
Calculating the values of the quality factorsNumerical values of model’s components for ISACO portal
23
24/29
Calculating the values of the model’s components & final quality of system
Final quality of system from both viewpoints
25/29
Evaluation and Discussion
• Comprehensiveness
• Web document architecture
• Considering different viewpoints
• Understandability
• It has all of these factors, so it has higher level of clarity and understandability in comparison with other models.
• Accuracy
• Considering the difference between the weights of values of the quality factors from the users’ and developers’ viewpoint.
26/29
Relation between quality factors and web document architecture
27/29
Quality evaluation of ISACO by ISO model
28/29
Conclusion and Future works
• In general, the contributions of this paper can be summarized into three areas:
• Presenting a method for customizing a general quality model for evaluation of a particular domain;
• Considering two main viewpoints(Developer & Viewer) in quality assessment of software products and presenting a method for weighting the quality factors.
• Due to the novelty of e-commerce systems, there are many research directions in which this work can be expanded.
• Therefore, focusing on environmental infrastructures as another dimension of the model may be an appropriate research direction.
29/29
Thanks for Your Listening
29