Module 5:Software Quality
• Objective measurement is needed to determine whether a system meets
quality requirements.
• For software developers, assessing quality only after completion is too
late.
• Quality should be evaluated during development and aligned with the
development methods used.
• Customers may focus not only on product quality but also on whether
best development practices were followed.
Importance of Quality in Project Planning
• Quality affects all stages of project planning and execution.
• It is emphasized particularly within the Step Wise project
framework.
The place of software quality in project management
Step 1 – Identify Project Scope and Objectives
• Define project goals.
• Some objectives relate specifically to quality attributes of the application.
• Helps ensure early clarity on what level of quality is expected.
Step 2 – Identify Project Infrastructure
• Identify standards, tools, and procedures required for the project.
• Many infrastructure decisions directly affect quality standards (e.g., installation standards).
Step 3 – Analyze Project Characteristics
• Analyze the application to see whether it has special quality requirements.
• Safety-critical applications require additional quality assurance activities.
• Example: N-version development
• Multiple teams develop versions of the same software.
• Outputs are run in parallel and cross-checked for discrepancies.
Used when extremely high reliability is required.
Step 4: Identify the products and activities of the project
It is at this point that the entry, exit and process requirements are identified for each activity. This is described
later in this chapter.
Step 8: Review and publicize plan
At this stage the overall quality aspects of the project plan are reviewed.
Importance of Software Quality
• Quality is important in all products and services.
• Software has special characteristics that make quality even more
critical:
• Increasing criticality of software: Users depend heavily on software;
reliability is essential.
• Intangibility of software: Hard to judge progress or completion;
quality checks must be planned.
• Accumulating errors during development: Errors propagate if not
detected early.
• Changing nature of software: Software evolves; continual change
increases risk and complexity.
• Effective quality management is essential to project success.
Challenges in Software Quality
• Errors may remain hidden until late in the project.
• The number of dependencies increases as the system grows, making fault
control harder.
• Organizations rely more on software → higher expectations for reliability
and safety.
Defining Software Quality
• A system has:
• Functional requirements (what it must do),
• Quality requirements (performance, usability, reliability),
• Resource requirements (cost, schedule).
• Defining quality is essential because:
• It forces clarity on what “quality” means for the system.
• It allows measurable criteria to be established.
Why Measure Quality?
• To judge objectively whether the system meets requirements.
• Measures should:
• Relate to the maximum number of units to be fault-free.
• Include faults per thousand lines of code, etc.
• Measurement may be direct or indirect:
• Direct → count faults, defects.
• Indirect → user satisfaction, time taken for tasks, productivity indicators.
Quality Measures in Practice
• Quality measures help communicate what is required.
• They answer: “How do we know when we have been successful?”
• Measures should:
• Be taken early (during design + development).
• Not encourage hiding errors.
• Help identify improvements.
When a Quality Characteristic Is Needed
If a project requires a specific quality characteristic:
• A clear system quality requirement should be drafted:
• Definition of required characteristic
• Measurement method
• Target value
• Actions if target not met
Examples of Quality Requirements
• Fault tolerance
• Minimal acceptable defects
• Performance thresholds
• Acceptable trade-offs with other characteristics
Software Quality Models
Need for Quality Models
• Quantitative assessment of software quality is essential.
• Helps verify whether software meets quality requirements.
• Provides measurable characteristics for evaluation.
• ISO 9126 and various models (Garvin, McCall, Dromey, Boehm) used
widely.
Garvin’s Quality Dimensions
David Garvin proposed 8 quality dimensions focusing on user-
perceived quality.
• Performance: How well the software performs functions.
• Features: Additional capabilities supporting functionality.
• Reliability: Probability of satisfactory operation over time.
• Conformance: Degree of meeting requirements/specifications.
• Durability: Expected product life.
• Serviceability: Ease of maintenance and repair.
• Aesthetics: Look and feel of the product.
• Perceived Quality: User’s opinion of overall quality.
McCall’s and Dromey’s Models
McCall’s Quality Factors
• Organized into 3 categories: Product Operation, Product Revision, Product Transition.
Key Factors:
• Correctness
• Reliability
• Efficiency
• Integrity
• Usability
• Maintainability
• Flexibility
• Testability
• Portability
• Reusability
• Interoperability
Dromey’s Model
• Emphasizes relationship between product quality and high-level quality attributes.
• Quality depends on properties of internal components.
• Factors influence software reliability, functionality, and maintainability.
Boehm’s Quality Model
Boehm defined quality based on three high-level factors important to users:
1. As-is Utility
How well the software can be used easily, reliably, efficiently.
2. Maintainability
How easy it is to understand, modify, and retest the software.
3. Portability
• How easily the software can adapt to a new or changed environment.
Boehm’s model covers:
Broader set of product attributes than McCall
High emphasis on software maintainability
Hierarchical structure of characteristics and sub-attributes