IT

SV

Dept Of ISE Bengaluru - Navya Mukund ( SVIT )

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Module-05

Chapter:-1 - Software Quality

Introduction to System Quality

Concept of Quality:

• Generally agreed to be beneficial.

• Often vaguely defined in practice.
• Requires precise definition of required qualities.

Objective Assessment:

•
IT
Judging if a system meets quality requirements needs measurement.
Critical for package selection, e.g., Brigette at Brightmouth College.

Development Perspective:
SV
• Waiting until the system is complete to measure quality is too late.
• During development, it's important to:
o Assess the likely quality of the final system.
o Ensure development methods will produce the desired quality.

Emphasis on Development Methods:

• Potential customers, like Amanda at IOE, might focus on:

o Checking if suppliers use the best development methods.
o Ensuring these methods will lead to the required quality in the final system.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 1

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

The place of software quality in project planning

Quality Concerns:

• Present at all stages of project planning and execution.

• Key points in the Step Wise framework where quality is particularly emphasized:

A. Step 1: Identify Project Scope and Objectives

o Objectives may include qualities of the application to be delivered.

B. Step 2: Identify Project Infrastructure

IT
Activity 2.2 involves identifying installation standards and procedures,
often related to quality.

C. Step 3: Analyze Project Characteristics

o Activity 3.2 involves analyzing other project characteristics, including

quality-based ones.
SV
o Example: Safety-critical applications might require additional activities
such as n-version development, where multiple teams develop versions of
the same software to cross-check outputs.

D. Step 4: Identify the Products and Activities of the Project

o Identify entry, exit, and process requirements for each activity.

E. Step 5: Review and Publicize Plan

o Review the overall quality aspects of the project plan at this stage.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 2

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

IT
SV
Importance of software quality

Special Characteristics of Software Affecting Quality Management

General Expectation:

• Quality is a concern for all producers of goods and services.

Special Demands in Software:

1. Increasing Criticality of Software

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 3

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Final customers and users are increasingly concerned about software

quality, particularly reliability.
o Greater reliance on computer systems and use in safety-critical applications
(e.g., aircraft control).
2. Intangibility of Software
o Difficulty in verifying the satisfactory completion of project tasks.
o Tangibility is achieved by requiring developers to produce "deliverables"
that can be examined for quality.
3. Accumulating Errors During Software Development
o

IT
Errors in earlier steps can propagate and accumulate in later steps.
Errors found later in the project are more expensive to fix.
The unknown number of errors makes the debugging phase difficult to
control.

Defining Software Quality

System Requirements:
SV
• Functional Requirements: Define what the system is to do.
• Resource Requirements: Specify allowable costs.
• Quality Requirements: State how well the system is to operate.

External and Internal Qualities:

• External Qualities: Reflect the user's view, such as usability.

• Internal Factors: Known to developers, such as well-structured code, which may
enhance reliability.

Measuring Quality:

• Necessity of Measurement: To judge if a system meets quality requirements, its

qualities must be measurable.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 4

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Good Measure: Relates the number of units to the maximum possible (e.g., faults
per thousand lines of code).
• Clarification Through Measurement: Helps to define and communicate what
quality really means, effectively answering "how do we know when we have been
successful?"

Direct vs. Indirect Measures:

• Direct Measurement: Measures the quality itself (e.g., faults per thousand lines
of code).
•

•
IT
Indirect Measurement: Measures an indicator of the quality (e.g., number of user
inquiries at a help desk as an indicator of usability).

Setting Targets:

Impact on Project Team: Quality measurements set targets for team members.
Meaningful Improvement: Ensure that improvements in measured quality are
meaningful.
SV
o Example: Counting errors found in program inspections may not be
meaningful if errors are allowed to pass to the inspection stage rather than
being eradicated earlier.

Drafting a Quality Specification for Software Products

When there's concern about a specific quality characteristic in a software product, a

quality specification should include the following details:

1. Definition/Description
o Definition: Clear definition of the quality characteristic.
o Description: Detailed description of what the quality characteristic entails.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 5

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

2. Scale
o Unit of Measurement: The unit used to measure the quality characteristic
(e.g., faults per thousand lines of code).
3. Test
o Practical Test: The method or process used to test the extent to which the
quality attribute exists.
4. Minimally Acceptable
o Worst Acceptable Value: The lowest acceptable value, below which the
product would be rejected.
5. Target Range

o
o

IT
Planned Range: The range of values within which it is planned that the
quality measurement value should lie.
6. Current Value
Now: The value that applies currently to the quality characteristic.

Measurements Applicable to Quality Characteristics in Software

SV
When assessing quality characteristics in software, multiple measurements may be
applicable. For example, in the case of reliability, measurements could include:

1. Availability:
o Definition: Percentage of a particular time interval that a system is usable.
o Scale: Percentage (%).
o Test: Measure the system's uptime versus downtime over a specified
period.
o Minimally Acceptable: Typically high availability is desirable; specifics
depend on system requirements.
o Target Range: E.g., 99.9% uptime.
2. Mean Time Between Failures (MTBF):
o Definition: Total service time divided by the number of failures.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 6

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Scale: Time (e.g., hours, days).

o Test: Calculate the average time elapsed between system failures.
o Minimally Acceptable: Longer MTBF indicates higher reliability;
minimum varies by system criticality.
o Target Range: E.g., MTBF of 10,000 hours.
3. Failure on Demand:
o Definition: Probability that a system will not be available when required,
or probability that a transaction will fail.
o Scale: Probability (0 to 1).
o

o
o

IT
Test: Evaluate the system's response to demand or transaction processing.
Minimally Acceptable: Lower probability of failure is desired; varies by
system criticality.
Target Range: E.g., Failure on demand probability of less than 0.01.
4. Support Activity:
Definition: Number of fault reports generated and processed.
Scale: Count (number of reports).
SV
o Test: Track and analyze the volume and resolution time of fault reports.
o Minimally Acceptable: Lower number of fault reports indicates better
reliability.
o Target Range: E.g., Less than 10 fault reports per month.

Maintainability and Related Qualities:

• Maintainability: How quickly a fault can be corrected once detected.

• Changeability: Ease with which software can be modified.
• Analysability: Ease with which causes of failure can be identified, contributing to
maintainability.

These measurements help quantify and assess the reliability and maintainability of
software systems, ensuring they meet desired quality standards.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 7

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

ISO 9126 is a significant standard in defining software quality attributes and providing a
framework for assessing them. Here are the key aspects and characteristics defined by
ISO 9126:

ISO 9126 Software Quality Characteristics

1. Functionality:
o Definition: The functions that a software product provides to satisfy user
needs.
o Sub-characteristics: Suitability, accuracy, interoperability, security,

2. Reliability:
o

3. Usability:
o
IT
compliance.

Definition: The capability of the software to maintain its level of

performance under stated conditions.
Sub-characteristics: Maturity, fault tolerance, recoverability.

Definition: The effort needed to use the software.

SV
o Sub-characteristics: Understandability, learnability, operability,
attractiveness.
4. Efficiency:
o Definition: The ability to use resources in relation to the amount of work
done.
o Sub-characteristics: Time behavior, resource utilization.
5. Maintainability:
o Definition: The effort needed to make changes to the software.
o Sub-characteristics: Analyzability, modifiability, testability.
6. Portability:
o Definition: The ability of the software to be transferred from one
environment to another.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 8

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Sub-characteristics: Adaptability, install ability, co-existence.

Quality in Use (Additional Aspect)

• Definition: Focuses on how well the software supports specific user goals in a
specific context of use.
• Elements: Effectiveness, productivity, safety, satisfaction.

ISO 14598

• Purpose: Describes procedures for assessing the degree to which a software

IT
product conforms to the ISO 9126 quality characteristics.
Flexibility: Can be adapted to assess software against different quality
characteristics if needed.
SV
ISO 9126 Sub-characteristics: Compliance and Interoperability

Compliance

• Definition: Refers to the degree to which the software adheres to application-

related standards or legal requirements, such as auditing standards.
• Purpose: Added as a sub-characteristic to all six primary ISO 9126 external
characteristics.
• Example: Ensuring software meets specific industry regulations or international
standards relevant to its functionality, reliability, usability, efficiency,
maintainability, and portability.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 9

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Interoperability

• Definition: Refers to the ability of the software to interact with other systems
effectively.
• Clarification: ISO 9126 uses "interoperability" instead of "compatibility" to avoid
confusion with another characteristic called "replaceability".
• Importance: Ensures seamless integration and communication between different
software systems or components.
• Sub-characteristics: Includes aspects like interface compatibility, data exchange
capabilities, and standards compliance to facilitate effective interaction.

IT
ISO 9126 Sub-characteristics: Maturity and Recoverability
SV
Maturity

• Definition: Refers to the frequency of failure due to faults in a software product

over time.
• Implication: Higher maturity suggests that the software has been extensively
tested and used, leading to the discovery and removal of more faults.
• Measurement: Typically assessed by the rate of failure incidents reported per unit
of time or usage.

Recoverability

• Definition: Refers to the capability of the software to restore the system to its
normal operation after a failure or disruption.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 10

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Clarification: ISO 9126 distinguishes recoverability from security, which focuses

on controlling access to the system.
• Importance: Ensures that the software can recover from failures or disruptions
swiftly and effectively to minimize downtime and data loss.
• Sub-characteristics: Includes aspects like fault tolerance, resilience, and recovery
time objectives (RTOs).

Importance of Distinction

Security: Focuses on access control and protecting the system from unauthorized

IT
access, ensuring confidentiality, integrity, and availability.
Recoverability: Focuses on system resilience and the ability to recover from
failures, ensuring continuity of operations.
SV
ISO 9126 Sub-characteristics: Learnability and Attractiveness

Learnability

Definition: Refers to the ease with which users can learn to operate the
software and accomplish basic tasks.
• Focus: Primarily on the initial phase of user interaction with the software.
• Measurement: Assessed by the time it takes for new users to become
proficient with the software, often measured in training hours or tasks
completed.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 11

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Operability

• Definition: Refers to the ease with which users can operate and navigate the
software efficiently.
• Focus: Covers the overall usability of the software during regular use and
over extended periods.
• Measurement: Assessed by the user's experience with everyday tasks and
efficiency in completing them.

Importance of Distinction

•
IT
Learnability vs. Operability: Learnability focuses on initial user
experience and how quickly users can grasp the basics of using the software.
Operability, on the other hand, emphasizes the efficiency and ease of use
over extended periods and during regular use.
Attractiveness: A recent addition under usability, attractiveness focuses on
SV
the aesthetic appeal and user interface design, particularly relevant in
software products where user engagement and satisfaction are crucial, such
as games and entertainment applications.

Application in Software Evaluation

• Learnability: Critical for software that requires quick adoption and minimal
training, ensuring users can start using the software effectively from the
outset.
• Operability: Crucial for software used intensively or for extended periods,
focusing on efficiency, ease of navigation, and user comfort.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 12

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Attractiveness: Important in consumer-facing software where user

engagement, satisfaction, and retention depend on the software's visual
appeal and user interface design.

Analysability IT
ISO 9126 Sub-characteristics: Analysability, Changeability, and Stability

Definition: Refers to the ease with which the cause of a failure in the software can
SV
be determined.
Focus: Helps in diagnosing and understanding software failures or issues quickly
and accurately.
Importance: Facilitates efficient debugging and troubleshooting during software
maintenance and support phases.

Changeability

Definition: Also known as flexibility, changeability refers to the ease with which
software can be modified or adapted to changes in requirements or environment.
• Focus: Emphasizes the software's ability to accommodate changes without
introducing errors or unexpected behaviors.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 13

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Importance: Enables software to evolve and remain relevant over time by

supporting modifications and updates effectively.

Stability

• Definition: Refers to the software's ability to undergo changes or modifications

without introducing unexpected errors or side effects.
• Focus: Ensures that modifications or updates to the software do not destabilize its
existing functionality or compromise its reliability.
• Importance: Provides confidence in the software's ability to maintain operational

Clarification of Terms

•
IT
integrity despite ongoing changes or updates.

Changeability vs. Flexibility: Changeability focuses on the software's ability to

be modified without causing issues, while flexibility emphasizes its capacity to
adapt to diverse requirements or environments.
Stability: Ensures that software modifications are managed effectively to
SV
minimize risks of unintended consequences or disruptions.

Application in Software Development

• Analysability: Essential for efficient debugging and problem-solving, ensuring

quick resolution of software issues.
• Changeability: Critical for agile software development, supporting iterative
changes and continuous improvement.
• Stability: Ensures reliability and consistency during software updates or
modifications, maintaining overall system integrity.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 14

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

ISO 9126 Quality Characteristics: Portability Compliance, Replaceability, and

Coexistence

Portability Compliance

•
IT
Definition: Refers to the adherence of the software to standards that facilitate its
transferability and usability across different platforms or environments.
Focus: Ensures that the software can run efficiently and effectively on various
hardware and software configurations without needing extensive modifications.
Importance: Facilitates broader deployment and reduces dependency on specific
SV
hardware or software configurations.

Replaceability

• Definition: Refers to the ability of the software to replace older versions or

components without causing disruptions or compatibility issues.
• Focus: Emphasizes upward compatibility, allowing new versions or components
to seamlessly integrate with existing systems or environments.
• Importance: Supports evolutionary software development by enabling updates
and enhancements without requiring extensive rework.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 15

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Coexistence

• Definition: Refers to the ability of the software to peacefully share resources and
operate alongside other software components within the same environment.
• Focus: Does not necessarily involve direct data exchange but ensures
compatibility and non-interference with other software components.
• Importance: Enables integration of the software into complex IT ecosystems
without conflicts or performance degradation.

Clarification of Terms

•
IT
Portability Compliance: Ensures software can run across different platforms by
adhering to common standards and programming languages.
Replaceability: Supports compatibility and seamless integration of new software
versions or components with existing systems.
Coexistence: Ensures harmonious operation of the software alongside other
components within the same environment.
SV
Guidelines for Use of Quality Characteristics

ISO 9126 provides structured guidelines for assessing and managing software quality
characteristics based on the specific needs and requirements of the software product. It
emphasizes the variation in importance of these characteristics depending on the type and
context of the software product being developed.

Steps Suggested After Establishing Requirements

Once the software product requirements are established, ISO 9126 suggests the following
steps:

1. Specify Quality Characteristics: Define and prioritize the relevant quality

characteristics based on the software's intended use and stakeholder requirements.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 16

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

2. Define Metrics and Measurements: Establish measurable criteria and metrics for
evaluating each quality characteristic, ensuring they align with the defined
objectives and user expectations.
3. Plan Quality Assurance Activities: Develop a comprehensive plan for quality
assurance activities, including testing, verification, and validation processes to
ensure adherence to quality standards.
4. Monitor and Improve Quality: Continuously monitor software quality
throughout the development lifecycle, identifying areas for improvement and
taking corrective actions as necessary.

IT
5. Document and Report: Document all quality-related activities, findings, and
improvements, and provide clear and transparent reports to stakeholders on
software quality status and compliance.

Importance of Quality Characteristics:

• Reliability: Critical for safety-critical systems where failure can have severe
consequences. Measures like mean time between failures (MTBF) are essential.
SV
• Efficiency: Important for real-time systems where timely responses are crucial.
Measures such as response time are key indicators.

External Quality Measurements:

• Reliability: Measure with MTBF or similar metrics.

• Efficiency: Measure with response time or time behavior metrics.

Mapping Measurements to User Satisfaction:

• For response time, user satisfaction could be mapped as follows (hypothetical

example):
o Excellent: Response time < 1 second
o Good: Response time 1-3 seconds

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 17

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Acceptable: Response time 3-5 seconds

o Poor: Response time > 5 seconds

Internal Measurements and Intermediate Products:

• Internal measurements like code execution times can help predict external
qualities like response time during software design and development.
• Predicting external qualities from internal measurements is challenging and often
requires validation in the specific environment where the software will operate.

Challenges with Mapping Internal and External Quality Characteristics:

IT
ISO 9126 acknowledges that correlating internal code metrics to external quality
characteristics like reliability can be difficult.
This challenge is addressed in a technical report rather than a full standard,
indicating ongoing research and development in this area.
SV
Based on the ISO 9126 framework and your points:

1. Measurement Indicators Across Development Stages:

o Early Stages: Qualitative indicators like checklists and expert judgments
are used to assess compliance with predefined criteria. These are subjective
and based on qualitative assessments.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 18

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Later Stages: Objective and quantitative measurements become more

prevalent as the software product nears completion. These measurements
provide concrete data about software performance and quality attributes.
2. Overall Assessment of Product Quality:
o Combining ratings for different quality characteristics into a single overall
rating for software is challenging due to:
▪ Different measurement scales and methodologies for each quality
characteristic.
▪ Sometimes, enhancing one quality characteristic (e.g., efficiency)

o
IT
may compromise another (e.g., portability).
Balancing these trade-offs can be complex and context-dependent.
3. Purpose of Quality Assessment:
Software Development: Assessment focuses on identifying weaknesses
early, guiding developers to meet quality requirements, and ensuring
continuous improvement throughout the development lifecycle.
Software Acquisition: Helps in evaluating software products from external
SV
suppliers based on predefined quality criteria.
o Independent Assessment: Aims to provide an unbiased evaluation of
software quality for stakeholders like regulators or consumers.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 19

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

It seems like you're describing a method for evaluating and comparing software products
based on their quality characteristics. Here's a summary and interpretation of your
approach:

o IT
1. Rating for User Satisfaction:
Products are evaluated based on mandatory quality levels that must be met.
Beyond these mandatory levels, user satisfaction ratings in the range of 0 to
5 are assigned for other desirable characteristics.
Objective measurements of functions are used to determine different levels
of user satisfaction, which are then mapped to numerical ratings (see Table
13.2 for an example).
SV
2. Importance Weighting:
o Each quality characteristic (e.g., usability, efficiency, maintainability) is
assigned an importance rating on a scale of 1 to 5.
o These importance ratings reflect how critical each quality characteristic is
to the overall evaluation of the software product.
3. Calculation of Overall Score:
o Weighted scores are calculated for each quality characteristic by
multiplying the quality score by its importance weight.
o The weighted scores for all characteristics are summed to obtain an overall
score for each software product.
4. Comparison and Preference Order:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 20

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Products are then ranked in order of preference based on their overall

scores. Higher scores indicate products that are more likely to satisfy user
requirements and preferences across the evaluated quality characteristics.

This method provides a structured approach to evaluating software products based on

user satisfaction ratings and importance weights for quality characteristics.

It allows stakeholders to compare and prioritize products effectively based on their

specific needs and preferences.

IT
SV
Certainly, when conducting a quality assessment on behalf of a user community or
professional body, several considerations come into play:

1. Objective Assessment: The goal is to provide an objective evaluation of software

tools that support the working practices of the community's members. This
assessment aims to be independent of individual user environments and
preferences.
2. Variability in Results: The outcome of such assessments can vary significantly
based on the weightings assigned to each software characteristic. Different
stakeholders within the community may have varying requirements and priorities.
3. Caution in Interpretation: It's crucial to exercise caution in interpreting and
applying the results of these assessments. While they strive for objectivity, they

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 21

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

are still influenced by the criteria set and the relative importance assigned to each
quality characteristic.
4. Community Needs: Understanding the specific needs and expectations of the user
community is essential. The assessment should align closely with the community's
goals and the functionalities they require from the software tools being evaluated.
5. Transparency and Feedback: Providing transparency in the assessment process
and gathering feedback from community members can enhance the credibility and
relevance of the evaluation results. This helps ensure that the assessment
adequately reflects the needs and perspectives of the community.

software development:

1. Product Metrics:
o
IT
Product and process metrics

Understanding the differences between product metrics and process metrics is crucial in

Purpose: Measure the characteristics of the software product being

SV
developed.
o Examples:
▪ Size Metrics: Such as Lines of Code (LOC) and Function Points,
which quantify the size or complexity of the software.
▪ Effort Metrics: Like Person-Months (PM), which measure the
effort required to develop the software.
▪ Time Metrics: Such as the duration in months or other time units
needed to complete the development.
2. Process Metrics:
o Purpose: Measure the effectiveness and efficiency of the development
process itself.
o Examples:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 22

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

▪ Review Effectiveness: Measures how thorough and effective code

reviews are in finding defects.
▪ Defect Metrics: Average number of defects found per hour of
inspection, average time taken to correct defects, and average
number of failures detected during testing per line of code.
▪ Productivity Metrics: Measures the efficiency of the development
team in terms of output per unit of effort or time.
▪ Quality Metrics: Such as the number of latent defects per line of
code, which indicates the robustness of the software after

Differences:

•
IT
development.

Focus: Product metrics focus on the characteristics of the software being built
(size, effort, time), while process metrics focus on how well the development
process is performing (effectiveness, efficiency, quality).
Use: Product metrics are used to gauge the attributes of the final software product,
SV
aiding in planning, estimation, and evaluation. Process metrics help in assessing
and improving the development process itself, aiming to enhance quality,
efficiency, and productivity.
• Application: Product metrics are typically applied during and after development
phases to assess the product's progress and quality. Process metrics are applied
throughout the development lifecycle to monitor and improve the development
process continuously.

By employing both types of metrics effectively, software development teams can better
manage projects, optimize processes, and deliver high-quality software products that
meet user expectations.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 23

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Product versus process quality management

In software development, managing quality can be approached from two main

perspectives: product quality management and process quality management. Here’s a
breakdown of each approach and their key aspects:

Product Quality Management

Product quality management focuses on evaluating and ensuring the quality of the
software product itself. This approach is typically more straightforward to implement and
measure after the software has been developed.

Aspects:

IT
1. Measurement Focus: Emphasizes metrics that assess the characteristics and
attributes of the final software product, such as size (LOC, function points),
reliability (defects found per LOC), performance (response time), and usability
(user satisfaction ratings).
SV
2. Evaluation Timing: Product quality metrics are often measured and evaluated
after the software product has been completed or at significant milestones during
development.
3. Benefits:
o Provides clear benchmarks for evaluating the success of the software
development project.
o Facilitates comparisons with user requirements and industry standards.
o Helps in identifying areas for improvement in subsequent software versions
or projects.
4. Challenges:
o Predicting final product quality based on intermediate stages (like early
code modules or prototypes) can be challenging.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 24

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Metrics may not always capture the full complexity or performance of the
final integrated product.

Process Quality Management

Process quality management focuses on assessing and improving the quality of the
development processes used to create the software. This approach aims to reduce errors
and improve efficiency throughout the development lifecycle.

Aspects:

IT
1. Measurement Focus: Emphasizes metrics related to the development processes
themselves, such as defect detection rates during inspections, rework effort,
productivity (e.g., lines of code produced per hour), and adherence to defined
standards and procedures.
2. Evaluation Timing: Process quality metrics are monitored continuously
throughout the development lifecycle, from initial planning through to deployment
and maintenance.
SV
3. Benefits:
o Helps in identifying and correcting errors early in the development process,
reducing the cost and effort of rework.
o Facilitates continuous improvement of development practices, leading to
higher overall quality in software products.
o Provides insights into the effectiveness of development methodologies and
practices used by the team.
4. Challenges:
o Requires consistent monitoring and analysis of metrics throughout the
development lifecycle.
o Effectiveness of process improvements may not always translate directly
into improved product quality without careful management and integration.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 25

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Integration and Synergy

• While product and process quality management approaches have distinct focuses,
they are complementary.
• Effective software development teams often integrate both approaches to achieve
optimal results.
• By improving process quality, teams can enhance product quality metrics, leading
to more reliable, efficient, and user-friendly software products.

Quality Management systems

IT
ISO 9001:2000, now superseded by newer versions but still relevant in principle, outlines
standards for Quality Management Systems (QMS). Here’s a detailed look at its key
aspects and how it applies to software development:

ISO 9001:2000 Overview

ISO 9001:2000 is part of the ISO 9000 series, which sets forth guidelines and
SV
requirements for implementing a Quality Management System (QMS).

The focus of ISO 9001:2000 is on ensuring that organizations have effective processes in
place to consistently deliver products and services that meet customer and regulatory
requirements.

Key Elements:

1. Fundamental Features:
o Describes the basic principles of a QMS, including customer focus,
leadership, involvement of people, process approach, and continuous
improvement.
o Emphasizes the importance of a systematic approach to managing
processes and resources.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 26

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

2. Applicability to Software Development:

o ISO 9001:2000 can be applied to software development by ensuring that
the development processes are well-defined, monitored, and improved.
o It focuses on the development process itself rather than the end product
certification (unlike product certifications such as CE marking).
3. Certification Process:
o Organizations seeking ISO 9001:2000 certification undergo an audit
process conducted by an accredited certification body.
o Certification demonstrates that the organization meets the requirements of

o
IT
ISO 9001:2000 and has implemented an effective QMS.
4. Quality Management Principles:
o Customer focus: Meeting customer requirements and enhancing customer
satisfaction.
Leadership: Establishing unity of purpose and direction.
Involvement of people: Engaging the entire organization in achieving
quality objectives.
SV
o Process approach: Managing activities and resources as processes to
achieve desired outcomes.
o Continuous improvement: Continually improving QMS effectiveness.

Application to Software Development

In software development scenarios, ISO 9001:2000 helps organizations:

• Define and document processes related to software development, testing, and

maintenance.
• Establish quality objectives and metrics for monitoring and evaluating software
development processes.
• Implement corrective and preventive actions to address deviations from quality
standards.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 27

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Ensure that subcontractors and external vendors also adhere to quality standards
through effective quality assurance practices.

Criticisms and Considerations

• Perceived Value: Critics argue that ISO 9001 certification does not guarantee the
quality of the end product but rather focuses on the process.
• Cost and Complexity: Obtaining and maintaining certification can be costly and
time-consuming, which may pose challenges for smaller organizations.
• Focus on Compliance: Some organizations may become overly focused on

IT
meeting certification requirements rather than improving overall product quality.

Despite these criticisms, ISO 9001:2000 provides a structured framework that, when
implemented effectively, can help organizations improve their software development
processes and overall quality management practices.

It emphasizes continuous improvement and customer satisfaction, which are crucial

aspects in the competitive software industry.
SV
BS EN ISO 9001:2000 outlines comprehensive requirements for implementing a Quality
Management System (QMS). Here’s an overview of its key requirements and principles

Principles of BS EN ISO 9001:2000

1. Customer Focus:
o Understanding and meeting customer requirements to enhance satisfaction.
2. Leadership:
o Providing unity of purpose and direction for achieving quality objectives.
3. Involvement of People:
o Engaging employees at all levels to contribute effectively to the QMS.
4. Process Approach:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 28

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Focusing on individual processes that create products or deliver services.

o Managing these processes as a system to achieve organizational objectives.
5. Continuous Improvement:
o Continually enhancing the effectiveness of processes based on objective
measurements and analysis.
6. Factual Approach to Decision Making:
o Making decisions based on analysis of data and information.
7. Mutually Beneficial Supplier Relationships:
o Building and maintaining good relationships with suppliers to enhance

o
IT
capabilities and performance.

Activities in BS EN ISO 9001:2000 Cycle

1. Understanding Customer Needs:

Identifying and defining customer requirements and expectations.
2. Establishing Quality Policy:
o Defining a framework for quality objectives aligned with organizational
SV
goals.
3. Process Design:
o Designing processes that ensure products and services meet quality
objectives.
4. Allocation of Responsibilities:
o Assigning responsibilities for meeting quality requirements within each
process.
5. Resource Management:
o Ensuring adequate resources (human, infrastructure, etc.) are available for
effective process execution.
6. Measurement and Monitoring:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 29

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Designing methods to measure and monitor process effectiveness and

efficiency.
o Gathering data and identifying discrepancies between actual performance
and targets.
7. Analysis and Improvement:
o Analyzing causes of discrepancies and implementing corrective actions to
improve processes continually.

Detailed Requirements

1. Documentation:
o

IT
Maintaining documented objectives, procedures (in a quality manual),
plans, and records that demonstrate adherence to the QMS.
Implementing a change control system to manage and update
documentation as necessary.
2. Management Responsibility:
o Top management must actively manage the QMS and ensure that processes
SV
conform to quality objectives.
3. Resource Management:
o Ensuring adequate resources, including trained personnel and infrastructure,
are allocated to support QMS processes.
4. Production and Service Delivery:
o Planning, reviewing, and controlling production and service delivery
processes to meet customer requirements.
o Communicating effectively with customers and suppliers to ensure clarity
and alignment on requirements.
5. Measurement, Analysis, and Improvement:
o Implementing measures to monitor product conformity, QMS effectiveness,
and process improvements.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 30

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Using data and information to drive decision-making and enhance overall

organizational performance.

Process capability models

The evolution of quality assurance paradigms from product assurance to process

assurance marks a significant shift in how organizations ensure quality in their outputs.
Here’s an overview of some key concepts and methodologies related to process-based
quality management:

Historical Perspective

o
IT
1. Before the 1950s:
Quality assurance primarily focused on extensive testing of finished
products to identify defects.
2. Shift to Process Assurance:
Later paradigms emphasize that ensuring a good quality process leads to
good quality products.
SV
o Modern quality assurance techniques prioritize recognizing, defining,
analyzing, and improving processes.

Total Quality Management (TQM)

1. Definition:
o TQM focuses on continuous improvement of processes through
measurement and redesign.
o It advocates that organizations continuously enhance their processes to
achieve higher levels of quality.

Business Process Reengineering (BPR)

1. Objective:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 31

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o BPR aims to fundamentally redesign and improve business processes.

o It seeks to achieve radical improvements in performance metrics, such as
cost, quality, service, and speed.

Process Capability Models

1. SEI Capability Maturity Model (CMM) and CMMI:

o Developed by the Software Engineering Institute (SEI), CMM and CMMI
provide a framework for assessing and improving the maturity of processes.
o They define five maturity levels, from initial (ad hoc processes) to

IT
optimized (continuous improvement).
CMMI (Capability Maturity Model Integration) integrates various
disciplines beyond software engineering.
2. ISO 15504 (SPICE):
o ISO/IEC 15504, also known as SPICE (Software Process Improvement and
Capability dEtermination), is an international standard for assessing and
improving process capability.
SV
o It provides a framework for evaluating process maturity based on process
attributes and capabilities.
3. Six Sigma:
o Six Sigma focuses on reducing defects in processes to a level of 3.4 defects
per million opportunities (DPMO).
o It emphasizes data-driven decision-making and process improvement
methodologies like DMAIC (Define, Measure, Analyze, Improve, Control).

Importance of Process Metrics

• During Product Development:

o Process metrics are more meaningfully measured during product
development compared to product metrics.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 32

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o They help in identifying process inefficiencies, bottlenecks, and areas for

improvement early in the development lifecycle.

SEI capability maturity model (CMM)

The SEI Capability Maturity Model (CMM) is a framework developed by the Software
Engineering Institute (SEI) to assess and improve the maturity of software development
processes within organizations.

It categorizes organizations into five maturity levels based on their process capabilities
and practices:

SEI CMM Levels:

1. Level 1: Initial
o

▪
IT
Characteristics:
Chaotic and ad hoc development processes.
Lack of defined processes or management practices.
SV
▪ Relies heavily on individual heroics to complete projects.
o Outcome:
▪ Project success depends largely on the capabilities of individual
team members.
▪ High risk of project failure or delays.
2. Level 2: Repeatable
o Characteristics:
▪ Basic project management practices like planning and tracking
costs/schedules are in place.
▪ Processes are somewhat documented and understood by the team.
o Outcome:
▪ Organizations can repeat successful practices on similar projects.
▪ Improved project consistency and some level of predictability.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 33

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

3. Level 3: Defined
o Characteristics:
▪ Processes for both management and development activities are
defined and documented.
▪ Roles and responsibilities are clear across the organization.
▪ Training programs are implemented to build employee capabilities.
▪ Systematic reviews are conducted to identify and fix errors early.
o Outcome:
▪ Consistent and standardized processes across the organization.

o
▪

4. Level 4: Managed
o

Outcome:
IT
Better management of project risks and quality.

Characteristics:
▪ Processes are quantitatively managed using metrics.
Quality goals are set and measured against project outcomes.
Process metrics are used to improve project performance.
SV
▪ Focus on managing and optimizing processes to meet quality and
performance goals.
▪ Continuous monitoring and improvement of project execution.
5. Level 5: Optimizing
o Characteristics:
▪ Continuous process improvement is ingrained in the organization's
culture.
▪ Process metrics are analyzed to identify areas for improvement.
▪ Lessons learned from projects are used to refine and enhance
processes.
▪ Innovation and adoption of new technologies are actively pursued.
o Outcome:
▪ Continuous innovation and improvement in processes.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 34

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

▪ High adaptability to change and efficiency in handling new

challenges.
▪ Leading edge in technology adoption and process optimization.

Use of SEI CMM:

• Capability Evaluation: Used by contract awarding authorities (like the US DoD)

to assess potential contractors' capabilities to predict performance if awarded a
contract.
• Process Assessment: Internally used by organizations to improve their own

IT
process capabilities through assessment and recommendations for improvement.

SEI CMM has been instrumental not only in enhancing the software development
practices within organizations but also in establishing benchmarks for industry standards.

It encourages organizations to move from chaotic and unpredictable processes (Level 1)

to optimized and continuously improving processes (Level 5), thereby fostering better
quality, efficiency, and predictability in software development efforts.
SV
Key process areas

The Capability Maturity Model Integration (CMMI) is an evolutionary improvement over

its predecessor, the Capability Maturity Model (CMM). Here's an overview of CMMI
and its structure:

Evolution and Purpose of CMMI

1. Evolution from CMM to CMMI:

o CMM Background: The original Capability Maturity Model (CMM) was
developed in the late 1980s by the Software Engineering Institute (SEI) at
Carnegie Mellon University, primarily to assess and improve the software

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 35

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

development processes of organizations, particularly those contracting with

the US Department of Defense.
o Expansion and Adaptation: Over time, various specific CMMs were
developed for different domains such as software acquisition (SA-CMM),
systems engineering (SE-CMM), and people management (PCMM). These
models provided focused guidance but lacked integration and consistency.
2. Need for Integration:
o Challenges: Organizations using multiple CMMs faced issues like
overlapping practices, inconsistencies in terminology, and difficulty in

IT
integrating practices across different domains.
CMMI Solution: CMMI (Capability Maturity Model Integration) was
introduced to provide a unified framework that could be applied across
various disciplines beyond just software development, including systems
engineering, product development, and services.

Structure and Levels of CMMI

SV
1. Levels of Process Maturity:
o Like CMM, CMMI defines five maturity levels, each representing a
different stage of process maturity and capability. These levels are:
▪ Level 1: Initial (similar to CMM Level 1)
▪ Level 2: Managed (similar to CMM Level 2)
▪ Level 3: Defined (similar to CMM Level 3)
▪ Level 4: Quantitatively Managed (an extension of CMM Level 4)
▪ Level 5: Optimizing (an extension of CMM Level 5)
2. Key Process Areas (KPAs):
o Definition: Similar to CMM, each maturity level in CMMI is characterized
by a set of Key Process Areas (KPAs). These KPAs represent clusters of

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 36

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

related activities that, when performed collectively, achieve a set of goals

considered important for enhancing process capability.
o Gradual Improvement: KPAs provide a structured approach for
organizations to incrementally improve their processes as they move from
one maturity level to the next.
Integration across Domains: Unlike the specific CMMs for various
disciplines, CMMI uses a more abstract and generalized set of
terminologies that can be applied uniformly across different domains.

IT
SV
Benefits of CMMI

Broad Applicability: CMMI's abstract nature allows it to be applied not only to

software development but also to various other disciplines and industries.
Consistency and Integration: Provides a unified framework for improving
processes, reducing redundancy, and promoting consistency across organizational
practices.
• Continuous Improvement: Encourages organizations to continuously assess and
refine their processes to achieve higher levels of maturity and performance.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 37

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

ISO/IEC 15504, also known as SPICE (Software Process Improvement and Capability
dEtermination), is a standard for assessing and improving software development
processes. Here are the key aspects of ISO 15504 process assessment:

Process Reference Model

• Reference Model: ISO 15504 uses a process reference model as a benchmark

against which actual processes are evaluated. The default reference model is often
ISO 12207, which outlines the processes in the software development life cycle
(SDLC) such as requirements analysis, architectural design, implementation,

Process Attributes

IT
testing, and maintenance.

Nine Process Attributes: ISO 15504 assesses processes based on nine attributes,
which are:
1. Process Performance (PP): Measures the achievement of process-specific
objectives.
SV
2. Performance Management (PM): Evaluates how well the process is
managed and controlled.
3. Work Product Management (WM): Assesses the management of work
products like requirements specifications, design documents, etc.
4. Process Definition (PD): Focuses on how well the process is defined and
documented.
5. Process Deployment (PR): Examines how the process is deployed within
the organization.
6. Process Measurement (PME): Evaluates the use of measurements to
manage and control the process.
7. Process Control (PC): Assesses the monitoring and control mechanisms in
place for the process.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 38

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

8. Process Innovation (PI): Measures the degree of innovation and

improvement in the process.
9. Process Optimization (PO): Focuses on optimizing the process to improve
efficiency and effectiveness.

Compatibility with CMMI

• Alignment with CMMI: ISO 15504 and CMMI share similar goals of assessing
and improving software development processes. While CMMI is more
comprehensive and applicable to a broader range of domains, ISO 15504 provides

development.

IT
a structured approach to process assessment specifically tailored to software

Benefits and Application

• Benefits: ISO 15504 helps organizations:

o

o
Evaluate their current processes against a recognized standard.
Identify strengths and weaknesses in their processes.
SV
o Implement improvements based on assessment findings.
• Application: The standard is used by organizations to conduct process
assessments either internally for improvement purposes or externally for
certification purposes.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 39

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

IT
SV
When assessors are judging the degree to which a process attribute is being fulfilled they
allocate one of the following scores:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 40

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

In the context of assessing process attributes according to ISO/IEC 15504 (SPICE),

evidence is crucial to determine the level of achievement for each attribute.

Here’s how evidence might be identified and evaluated for assessing the process

IT
attributes, taking the example of requirement analysis processes:

Example of Assessing Requirement Analysis Processes

1. Process Definition (PD):

o Evidence: A section in the procedures manual that outlines the steps, roles,
and responsibilities for conducting requirements analysis.
SV
o Assessment: Assessors would review the documented procedures to ensure
they clearly define how requirements analysis is to be conducted. This
indicates that the process is defined (3.1 in Table 13.5).
2. Process Deployment (PR):
o Evidence: Control documents or records showing that the documented
requirements analysis process has been used and followed in actual
projects.
o Assessment: Assessors would look for signed-off control documents at
each step of the requirements analysis process, indicating that the defined
process is being implemented and deployed effectively (3.2 in Table 13.5).

Using ISO/IEC 15504 Attributes

• Process Performance (PP):

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 41

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Evidence: Performance metrics related to the effectiveness and efficiency

of the requirements analysis process, such as the accuracy of captured
requirements, time taken for analysis, and stakeholder satisfaction surveys.
o Assessment: Assessors would analyze the metrics to determine if the
process meets its performance objectives (e.g., accuracy, timeliness).
• Process Control (PC):
o Evidence: Procedures and mechanisms in place to monitor and control the
requirements analysis process, such as regular reviews, audits, and
corrective action reports.

•
o

o
IT
Assessment: Assessors would review the control mechanisms to ensure
they effectively monitor the process and address deviations promptly.
Process Optimization (PO):
Evidence: Records of process improvement initiatives, feedback
mechanisms from stakeholders, and innovation in requirements analysis
techniques.
Assessment: Assessors would examine how the organization identifies
SV
opportunities for process improvement and implements changes to optimize
the requirements analysis process.

Importance of Evidence

• Objective Assessment: Evidence provides objective data to support the

assessment of process attributes.
• Validation: It validates that the process attributes are not just defined on paper but
are effectively deployed and managed.
• Continuous Improvement: Identifying evidence helps in identifying areas for
improvement and optimizing processes over time.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 42

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Implementing process improvement

Implementing process improvement in UVW, especially in the context of software

development for machine tool equipment, involves addressing several key challenges
identified within the organization.

Here’s a structured approach, drawing from CMMI principles, to address these issues and
improve process maturity:

Identified Issues at UVW

o
o

IT
1. Resource Overcommitment:
Issue: Lack of proper liaison between the Head of Software Engineering
and Project Engineers leads to resource overcommitment across new
systems and maintenance tasks simultaneously.
Impact: Delays in software deliveries due to stretched resources.
2. Requirements Volatility:
o Issue: Initial testing of prototypes often reveals major new requirements.
SV
o Impact: Scope creep and changes lead to rework and delays.
3. Change Control Challenges:
o Issue: Lack of proper change control results in increased demands for
software development beyond original plans.
o Impact: Increased workload and project delays.
4. Delayed System Testing:
o Issue: Completion of system testing is delayed due to a high volume of bug
fixes.
o Impact: Delays in product release and customer shipment.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 43

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Steps for Process Improvement

1. Formal Planning and Control

• Objective: Introduce structured planning and control mechanisms to assess and

distribute workloads effectively.
• Actions:
o Implement formal project planning processes where software requirements
are mapped to planned work packages.
o Define clear milestones and deliverables, ensuring alignment with both
hardware and software development phases.

•
o

o
o
IT
Monitor project progress against plans to identify emerging issues early.
Expected Outcomes:
Improved visibility into project status and resource utilization.
Early identification of potential bottlenecks or deviations from planned
schedules.
Enable better resource allocation and management across different projects.
SV
2. Change Control Procedures

• Objective: Establish robust change control procedures to manage and prioritize

system changes effectively.
• Actions:
o Define a formal change request process with clear documentation and
approval workflows.
o Ensure communication channels between development teams, testing
groups, and project stakeholders are streamlined for change notifications.
o Implement impact assessment mechanisms to evaluate the effects of
changes on project timelines and resources.
• Expected Outcomes:
o Reduced scope creep and unplanned changes disrupting project schedules.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 44

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Enhanced control over system modifications, minimizing delays and

rework.

3. Enhanced Testing and Validation

• Objective: Improve testing and validation processes to reduce delays in system

testing and bug fixes.
• Actions:
o Strengthen collaboration between development and testing teams to ensure
comprehensive test coverage early in the development lifecycle.

o
o

o
cycles.

IT
Implement automated testing frameworks where feasible to expedite testing

Foster a culture of quality assurance and proactive bug identification

throughout the development phases.
Expected Outcomes:
Faster turnaround in identifying and resolving bugs during testing.
Timely completion of system testing phases, enabling on-time product
SV
releases.

Moving Towards Process Maturity Levels

• Level 1 to Level 2 Transition:

o Focus: Transition from ad-hoc, chaotic practices to defined processes with
formal planning and control mechanisms.
o Benefits: Improved predictability in project outcomes, better resource
management, and reduced project risks.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 45

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

IT
SV

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 46

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Six Sigma

Implementing Six Sigma Methodology at UVW

UVW, a company specializing in machine tool equipment with sophisticated control

software, can benefit significantly from implementing Six Sigma methodologies.

Here’s how UVW can adopt and benefit from Six Sigma:

Overview of Six Sigma

Six Sigma is a disciplined, data-driven approach aimed at improving process outputs by

identifying and eliminating causes of defects, thereby reducing variability in processes.

1. Define:
o

o
IT
The goal is to achieve a level of quality where the process produces no more than 3.4
defects per million opportunities.

Steps to Implement Six Sigma at UVW

Objective: Clearly define the problem areas and goals for improvement.
Action: Identify critical processes such as software development, testing,
SV
and deployment where defects and variability are impacting quality and
delivery timelines.
2. Measure:
o Objective: Quantify current process performance and establish baseline
metrics.
o Action: Use statistical methods to measure defects, cycle times, and other
relevant metrics in software development and testing phases.
3. Analyse:
o Objective: Identify root causes of defects and variability in processes.
o Action: Conduct thorough analysis using tools like root cause analysis,
process mapping, and statistical analysis to understand why defects occur
and where process variations occur.
4. Improve:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 47

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Objective: Implement solutions to address root causes and improve process

performance.
o Action: Develop and implement process improvements based on the
analysis findings. This could include standardizing processes, enhancing
communication between teams (e.g., software development and testing),
and implementing better change control procedures.
5. Control:
o Objective: Maintain the improvements and prevent regression.
o Action: Establish control mechanisms to monitor ongoing process

Focus Areas:
o

o
IT
performance. Implement measures such as control charts, regular audits,
and performance reviews to sustain improvements.

Application to UVW's Software Development

Addressing late deliveries due to resource overcommitment.

Managing requirements volatility and change control effectively.
SV
o Enhancing testing processes to reduce defects and delays in system testing
phases.
• Tools and Techniques:
o Use of DMAIC (Define, Measure, Analyse, Improve, Control) for existing
process improvements.
o Application of DMADV (Define, Measure, Analyse, Design, Verify) for
new process or product development to ensure high-quality outputs from
the outset.

Benefits of Six Sigma at UVW

• Improved Quality: Reduced defects and variability in software products.

• Enhanced Efficiency: Streamlined processes leading to faster delivery times.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 48

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Cost Savings: Reduced rework and operational costs associated with defects.

Techniques to enhance software quality

The discussion highlights several key themes in software quality improvement over time,
emphasizing shifts in practices and methodologies:

1. Increasing Visibility:
o Early practices like Gerald Weinberg's 'egoless programming' promoted
code review among programmers, enhancing visibility into each other's
work.
o

IT
Modern practices extend this visibility to include walkthroughs,
inspections, and formal reviews at various stages of development, ensuring
early detection and correction of defects.
2. Procedural Structure:
o Initially, software development lacked structured methodologies, but over
time, methodologies with defined processes for every stage (like Agile,
Waterfall, etc.) have become prevalent.
SV
o Structured programming techniques and 'clean-room' development further
enforce procedural rigor to enhance software quality.
3. Checking Intermediate Stages:
o Traditional approaches involved waiting until a complete, albeit imperfect,
version of software was ready for debugging.
o Contemporary methods emphasize checking and validating software
components early in development, reducing reliance on predicting external
quality from early design documents.
4. Inspections:
o Inspections are critical in ensuring quality at various development stages,
not just in coding but also in documentation and test case creation.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 49

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Techniques like Fagan inspections, pioneered by IBM, formalize the review

process with trained moderators leading discussions to identify defects and
improve quality.
5. Japanese Quality Techniques:
o Learnings from Japanese quality practices, such as quality circles and
continuous improvement, have influenced global software quality
standards, emphasizing rigorous inspection and feedback loops.
6. Benefits of Inspections:
o Inspections are noted for their effectiveness in eliminating superficial

IT
errors, motivating developers to write better-structured code, and fostering
team collaboration and spirit.
They also facilitate the dissemination of good programming practices and
improve overall software quality by involving stakeholders from different
stages of development.

The general principles behind the Fagan method

SV
• Inspections are carried out on all major deliverables.
• All types of defect are noted – not just logic or function errors.
• Inspections can be carried out by colleagues at all levels except the very top.
• Inspections are carried out using a predefined set of steps.
• Inspection meetings do not last for more than two hours.
• The inspection is led by a moderator who has had specific training in the
technique.
• The other participants have defined roles. For example, one person will act as a
recorder and note all defects found, and another will act as reader and take the
other participants through the document under inspection.
• Checklists are used to assist the fault-finding process.
• Material is inspected at an optimal rate of about 100 lines an hour.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 50

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Statistics are maintained so that the effectiveness of the inspection process can be
monitored.

Structured programming and clean-room software development

The late 1960s marked a pivotal period in software engineering where the complexity of
software systems began to outstrip the capacity of human understanding and testing
capabilities. Here are the key developments and concepts that emerged during this time:

1. Complexity and Human Limitations:

o Software systems were becoming increasingly complex, making it

o
IT
impractical to test every possible input combination comprehensively.
Edsger Dijkstra and others argued that testing could only demonstrate the
presence of errors, not their absence, leading to uncertainty about software
correctness.
2. Structured Programming:
To manage complexity, structured programming advocated breaking down
software into manageable components.
SV
o Each component was designed to be self-contained with clear entry and exit
points, facilitating easier understanding and validation by human
programmers.
3. Clean-Room Software Development:
o Developed by Harlan Mills and others at IBM, clean-room software
development introduced a rigorous methodology to ensure software
reliability.
o It involved three separate teams:
▪ Specification Team: Gathers user requirements and usage profiles.
▪ Development Team: Implements the code without conducting
machine testing; focuses on formal verification using mathematical
techniques.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 51

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

▪ Certification Team: Conducts testing to validate the software, using

statistical models to determine acceptable failure rates.
4. Incremental Development:
o Systems were developed incrementally, ensuring that each increment was
capable of operational use by end-users.
o This approach avoided the pitfalls of iterative debugging and ad-hoc
modifications, which could compromise software reliability.
5. Verification and Validation:
o Clean-room development emphasized rigorous verification at the

IT
development stage rather than relying on extensive testing to identify and
fix errors.
The certification team's testing was thorough and continued until statistical
models showed that the software failure rates were acceptably low.

Overall, these methodologies aimed to address the challenges posed by complex software
systems by promoting structured, systematic development processes that prioritize
correctness from the outset rather than relying on post hoc testing and debugging.
SV
Clean-room software development, in particular, contributed to the evolution of quality
assurance practices in software engineering, emphasizing formal methods and rigorous
validation techniques.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 52

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Formal methods

It seems like you're discussing formal methods in software development and the concept
of software quality circles. Here's a summary of the points covered:

Formal Methods in Software Development

• Definition: Formal methods use unambiguous, mathematically based specification

languages like Z and VDM. They define preconditions (input conditions) and
postconditions (output conditions) for procedures.
• Purpose: These methods ensure precise and unambiguous specifications and

•
conditions.

IT
allow for mathematical proof of algorithm correctness based on specified

Adoption: Despite being taught widely in universities, formal methods are rarely
used in mainstream software development. Object Constraint Language (OCL) is a
newer, potentially more accessible development in this area.

Software Quality Circles (SWQC)

SV
• Purpose: SWQCs are adapted from Japanese quality practices to improve
software development processes by reducing errors.
• Structure: Consist of 4 to 10 volunteers in the same department who meet
regularly to identify, analyze, and solve work-related problems.
• Process: The group selects a problem, identifies its causes, and proposes solutions.
Management approval may be required for implementing improvements.
• Benefits: Enhances team collaboration, spreads best practices, and focuses on
continuous process improvement.

If you have any specific questions or if there's more you'd like to explore on these topics
or related areas, feel free to ask!

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 53

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

The process you're describing involves the compilation of most probable error lists,
which is a proactive approach to improving software development processes. Here’s a
breakdown of the steps involved:

Compilation of Most Probable Error Lists

1. Identification of Common Errors: Teams, such as those in quality circles,

assemble to identify the most frequent types of errors occurring in a particular
phase of software development, such as requirements specification.
2. Analysis and Documentation: The team spends time analyzing past projects or

o
IT
current issues to compile a list of these common errors. This list documents
specific types of mistakes that have been identified as recurring.
3. Proposing Solutions: For each type of error identified, the team proposes
measures to reduce or eliminate its occurrence in future projects. For example:
Example Measure: Producing test cases simultaneously with requirements
specification to ensure early validation.
Example Measure: Conducting dry runs of test cases during inspections to
SV
catch errors early in the process.
4. Development of Checklists: The proposed measures are formalized into
checklists that can be used during inspections or reviews of requirements
specifications. These checklists serve as guidelines to ensure that identified errors
are systematically checked for and addressed.
5. Implementation and Feedback: The checklists are implemented into the
software development process. Feedback mechanisms are established to evaluate
the effectiveness of these measures in reducing errors and improving overall
quality.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 54

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Benefits of Most Probable Error Lists

• Improved Quality: By proactively identifying and addressing common errors, the

quality of requirements specifications (or other phases) improves.
• Efficiency: Early detection and prevention of errors reduce the need for costly
corrections later in the development cycle.
• Standardization: Checklists standardize the inspection process, ensuring that
critical aspects are consistently reviewed.

This approach aligns well with quality circles and other continuous improvement

IT
methodologies by fostering a culture of proactive problem-solving and learning from past
experiences.

If you have more questions or need further elaboration on any aspect, feel free to ask!

Lessons Learned Report

The concept of Lessons Learned reports and Post Implementation Reviews (PIRs) are
SV
crucial for organizational learning and continuous improvement in project management.
Here’s a breakdown of these two types of reports:

• Purpose: The Lessons Learned report is prepared by the project manager

immediately after the completion of the project. Its purpose is to capture insights
and experiences gained during the project execution.
• Content: Typically, a Lessons Learned report includes:
o Successes and Failures: What worked well and what didn’t during the
project.
o Challenges Faced: Difficulties encountered and how they were overcome.
o Best Practices: Practices or approaches that proved effective.
o Lessons Learned: Key takeaways and recommendations for future
projects.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 55

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Audience: The report is usually intended for internal stakeholders involved in

similar future projects, such as project managers, team members, and
organizational leadership.
• Follow-up: One common issue is the lack of follow-up on the recommendations
provided in these reports. This can occur due to a lack of dedicated resources or
organizational structure for implementing these lessons.

Post Implementation Review (PIR)

• Purpose: A PIR takes place after a significant period of operation of the new

• IT
system (typically after it has been in use for some time). Its focus is on evaluating
the effectiveness of the implemented system rather than the project process itself.
Timing: Conducted by someone who was not directly involved in the project to
ensure neutrality and objectivity.
Content: A PIR includes:
o System Performance: How well the system meets its intended objectives
and user needs.
SV
o User Feedback: Feedback from users on system usability and
functionality.
o Improvement Recommendations: Changes or enhancements suggested to
improve system effectiveness.
• Audience: The audience typically includes stakeholders who will benefit from
insights into the system’s operational performance and areas for improvement.
• Outcome: Recommendations from a PIR often lead to changes aimed at
enhancing the effectiveness and efficiency of the system.

Importance of Both Reports

• Learning and Improvement: Both reports contribute to organizational learning

by capturing valuable insights and experiences.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 56

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Continuous Improvement: They provide a basis for making informed decisions

and improvements in future projects and system implementations.

Testing

The text discusses the planning and management of testing in software development,
highlighting the challenges of estimating the amount of testing required due to unknowns,
such as the number of bugs left in the code.

It introduces the V-process model as an extension of the waterfall model, emphasizing

the importance of validation activities at each development stage.

IT
1. Quality Judgement:
The final judgement of software quality is based on its correct execution.
2. Testing Challenges:
o Estimating the remaining testing work is difficult due to unknown bugs in
the code.
3. V-Process Model:
SV
o Introduced as an extension of the waterfall model.
o Diagrammatic representation provided in Figure 13.5.
o Stresses the necessity for validation activities matching the project creation
activities.
4. Validation Activities:
o Each development step has a matching validation process.
o Defects found can cause a loop back to the corresponding development
stage for rework.
5. Discrepancy Handling:
o Feedback should occur only when there is a discrepancy between specified
requirements and implementation.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 57

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Example: System designer specifies a calculation method; if a developer

misinterprets it, the discrepancy is caught during system testing.
6. System Testing:
o Original designers are responsible for checking that software meets the
specified requirements, discovering any misunderstandings by developers.

IT
SV
Framework for Planning:

The V-process model provides a structure for making early planning decisions
about testing.

Types and Amounts of Testing:

• Decisions can be made about the types and amounts of testing required from the
beginning of the project.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 58

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Off-the-Shelf Software:

• If software is acquired off-the-shelf, certain stages like program design and coding
are not relevant.
• Consequently, program testing would not be necessary in this scenario.

User Requirements and Acceptance Testing:

• User requirements must still be produced.

• User acceptance tests remain valid and necessary regardless of the software being
off-the-shelf.

1. Objectives:
o

2. Definitions:
o
IT
Verification versus validation

Both techniques aim to remove errors from software.

Verification: Ensures outputs of one development phase conform to the

SV
previous phase's outputs.
o Validation: Ensures fully developed software meets its requirements
specification.
3. Objectives Clarified:
o Verification Objective: Check if artifacts produced after a phase conform to
those from the previous phase (e.g., design documents conform to requirements
specifications).
o Validation Objective: Check if the fully developed and integrated software
satisfies customer requirements.
4. Techniques:
o Verification Techniques: Review, simulation, and formal verification.
o Validation Techniques: Primarily based on product testing.
5. Process Stages:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 59

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Verification: Conducted during the development process to ensure

development activities are correct.
o Validation: Conducted at the end of the development process to ensure the
final product meets customer requirements.
6. Phase Containment of Errors:
o Verification aims for phase containment of errors, which is a cost-effective
way to eliminate bugs and an important software engineering principle.
7. V-Process Model Activities:
o All activities on the right side of the V-process model are verification activities

IT
except for the system testing block, which is a validation activity.

Testing activities

The text provides an overview of test case design approaches, levels of testing, and main
testing activities in software development.

It emphasizes the differences between black-box and white-box testing, the stages of
SV
testing (unit, integration, system), and the activities involved in the testing process.

Test Case Design Approaches

1. Black-Box Testing:
o Test cases are designed using only the functional specification.
o Based on input/output behavior without knowledge of internal structure.
o Also known as functional testing or requirements-driven testing.
2. White-Box Testing:
o Test cases are designed based on the analysis of the source code.
o Requires knowledge of the internal structure.
o Also known as structural testing or structure-driven testing.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 60

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Levels of Testing

1. Unit Testing:
o Tests individual components or units of a program.
o Conducted as soon as the coding for each module is complete.
o Allows for parallel activities since modules are tested separately.
o Referred to as testing in the small.
2. Integration Testing:
o Checks for errors in interfacing between modules.
o Units are integrated step by step and tested after each integration.
o

3. System Testing:
o

IT
Referred to as testing in the large.

Tests the fully integrated system to ensure it meets requirements.

Conducted after integration testing.

Testing Activities
SV
1. Test Planning:
o Involves determining relevant test strategies and planning for any required
test bed.
o Test bed setup is crucial, especially for embedded applications.
2. Test Suite Design:
o Planned testing strategies are used to design the set of test cases (test suite).
3. Test Case Execution and Result Checking:
o Each test case is executed, and results are compared with expected
outcomes.
o Failures are noted for test reporting when there is a mismatch between
actual and expected results.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 61

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

The text describes the detailed process and activities involved in software test reporting,
debugging, error correction, defect retesting, regression testing, and test closure.

It highlights the importance of formal issue recording, the adjudication of issues, and
various testing strategies to ensure software quality.

Test Reporting

1. Issue Raising:
o Report discrepancies between expected and actual results.
2. Issue Recording:
o

o
IT
Formal recording of issues and their history.
3. Review Body Decisions:
Dismissal: Misunderstanding of requirement by tester.
Fault Identification: Developers need to correct the issue.
Incorrect Requirement: Adding a new requirement and additional
work/payment.
SV
o Off-Specification Fault: The application can operate with the error.
4. Test Failure Notification:
o Informal intimation to the development team to optimize turnaround time.

Debugging and Error Correction

1. Debugging:
o Identify error statements by analyzing failure symptoms.
o Various debugging strategies are employed.
2. Error Correction:
o Correct the code after locating the error through debugging.
3. Defect Retesting:
o Retesting corrected code to check if the defect has been successfully
addressed (resolution testing).

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 62

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

4. Regression Testing:
o Ensures unmodified functionalities still work correctly after bug fixes.
o Runs alongside resolution testing to check for new errors introduced by
changes.

Test Closure

Test Completion:
Archiving documents related to lessons learned, test results, and logs for
future reference.

IT
Time-Consuming Activity:
Debugging is noted as usually the most time-consuming activity in the
testing process.
SV

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 63

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Who performs testing?

The text describes who performs testing in organizations, the importance and benefits of
test automation, and various types of automated testing tools.

It emphasizes that while test automation can significantly reduce human effort, improve
thoroughness, and lower costs, different tools have distinct advantages and challenges.

Who Performs Testing

1. Independent Testing Groups:

o

o
o

Test Automation
IT
Provide an independent assessment of software correctness before release.
Can offer a final quality check but may lead developers to take less care in
their work.
2. Integrated Testing Roles:
Testers work alongside developers rather than as a separate group.
SV
1. Importance of Test Automation:
o Testing is the most time-consuming and laborious software development
activity.
o Automation reduces human effort, time, and improves testing
thoroughness.
o Enables sophisticated test case design techniques.
2. Benefits of Test Automation:
o More testing with a large number of test cases in a short period without
significant cost overhead.
o Automated test results are more reliable and eliminate human errors.
o Simplifies regression testing by running old test cases repeatedly.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 64

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Reduces monotony, boredom, and errors in running the same test cases
repeatedly.
o Substantial cost and time reduction in testing and maintenance phases.

Types of Automated Testing Tools

1. Capture and Playback Tools:

o Record inputs and outputs during manual execution to replay tests
automatically later.
o Useful for regression testing but can be costly to maintain when the unit

o
IT
under test changes.
2. Automated Test Script Tools:
o Use scripts to drive automated test tools, providing inputs and recording
outputs.
Debugging and verifying test scripts require significant effort, and changes
to the unit under test necessitate updating scripts.
3. Random Input Test Tools:
SV
o Generate random test values to cover the input space, aiming to crash the
unit under test.
o Easy to implement but limited in scope as it only finds defects that crash
the system.
4. Model-Based Test Tools:
o Use simplified representations (models) of the program, like state or
activity models, to generate tests.
o Adequately cover the state space described by the model.

Estimating Errors Based on Code Size

1. Historical Data:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 65

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Use historical data to estimate errors per 1000 lines of code from past
projects.
o Apply this ratio to new system development to estimate potential errors
based on the code size.

Error Seeding Technique

1. Seeding Known Errors:

o Known errors are deliberately left in the code during desk checks.
o Example: If 10 errors are seeded and after testing, 30 errors are found,

Independent Reviews
IT
including 6 seeded errors, 60% of seeded errors are detected.
This suggests that around 40% of errors are still to be detected.
Formula to estimate total errors:

1. Tom Gilb's Approach:

o Two independent reviewers or groups are asked to inspect or test the same
SV
code.
o They must work independently of each other.
o Three counts are collected for better error estimation.

Using these methods helps in obtaining a better estimation of latent errors, providing a
clearer understanding of the remaining testing effort needed to ensure software quality.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 66

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Software reliability

o
o

o
IT
1. Definition and Importance:
Software reliability denotes the trustworthiness or dependability of a
software product.
It is defined as the probability of the software working correctly over a
given period of time.
Reliability is a crucial quality attribute for software products.
SV
2. Defects and Reliability:
o A large number of defects typically indicate unreliability.
o Reducing defects generally improves reliability.
o It is challenging to create a mathematical formula to relate reliability
directly to the number of latent defects.
3. Execution Frequency and Defect Impact:
o Errors in infrequently executed parts of the software have less impact on
overall reliability.
o Studies show that 90% of a typical program's execution time is spent on
10% of its instructions.
o The specific location of a defect (core or non-core part) affects reliability.
4. Observer Dependency:
o Reliability is dependent on user behavior and usage patterns.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 67

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o A bug may affect different users differently based on how they use the
software.
5. Reliability Improvement Over Time:
o Reliability usually improves during testing and operational phases as
defects are identified and fixed.
o This improvement can be modeled mathematically using Reliability
Growth Models (RGM).
6. Reliability Growth Models (RGM):
o RGMs describe how reliability improves as failures are reported and bugs

Quality plans
IT
are corrected.
Various RGMs exist, including the Jelinski–Moranda model, Littlewood–
Verall’s model, and Goel–Okutomo’s model.
RGMs help predict when a certain reliability level will be achieved, guiding
decisions on when testing can be stopped.
SV
Purpose of Quality Plans:

• Quality plans detail how standard quality procedures and standards from an
organization's quality manual will be applied to a specific project.
• They ensure all quality-related activities and requirements are addressed.

Integration with Project Planning:

• If a thorough project planning approach, like Step Wise, is used, quality-related

activities and requirements are often already integrated into the main planning
process.
• In such cases, a separate quality plan might not be necessary.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 68

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Client Requirements:

• For software developed for external clients, the client's quality assurance staff may
require a quality plan to ensure the quality of the delivered products.
• This requirement ensures that the client’s quality standards are met.

Function of a Quality Plan:

A quality plan acts as a checklist to confirm that all quality issues have been
addressed during the planning process.

IT
Most of the content in a quality plan references other documents that detail
specific quality procedures and standards.
SV

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 69

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Chapter: - 2-Software Project Estimation

Observations on Software Project Estimation

Estimation is Not Always Accurate: Initial estimates often turn out to be inaccurate due to
unknown factors and project complexity. The inherent uncertainty of software development
makes precise estimation difficult.

Estimation Evolves Over Time: Early in the project, estimates are rough and become more
refined as more details are uncovered. A continuous re-estimation process can improve

IT
accuracy as the project progresses.

Complexity of Requirements: Misunderstanding or underestimating the complexity of

project requirements can lead to incorrect estimates. Clear and well-defined requirements are
crucial for more accurate predictions.

Human Factors: Over-optimism, bias, and miscommunication between developers, project

managers, and clients can skew estimates. Over-reliance on expert judgment can also be risky
SV
if experts are overly confident or not familiar with specific project details.

Risk and Uncertainty Management: The level of risk and uncertainty in a project
significantly affects estimation. Estimation techniques must consider potential risks, with
appropriate buffers to handle unexpected changes.

Impact of Scope Changes: Scope creep—unplanned additions to the project—can drastically

alter original estimates. Proper scope management and agreement on the project boundaries
are essential for maintaining the accuracy of estimates.

Dependency on Team and Technology: The skill level of the team, familiarity with the
technology stack, and the development process used (e.g., Agile or Waterfall) heavily
influence the estimate. An experienced team with a clear understanding of the project
typically provides better estimates.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 70

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Historical Data and Metrics: Utilizing historical project data (e.g., past performance, similar
projects) helps make better-informed estimates. Without such data, estimates tend to be less
reliable.

External Factors: Factors like customer involvement, market conditions, legal/regulatory

changes, and vendor dependencies can impact estimates, causing delays or increased costs.

Challenges with Estimation Models: Standard estimation models like COCOMO or

function point analysis may not always apply perfectly to every project. Adjustments and
customizations are often necessary to suit the specific context of the project.

overruns.

IT
Underestimation Issues: There's a frequent tendency to underestimate time, cost, and effort
due to pressure to meet aggressive deadlines or budget constraints, leading to project

Communication and Collaboration: Clear communication between all stakeholders is

essential to ensure that everyone understands the assumptions, constraints, and expectations
regarding the estimate.
SV
Decomposition Techniques
Decomposition techniques break down complex projects into smaller, more manageable
parts, making it easier to estimate time, effort, and resources accurately.

Work Breakdown Structure (WBS)

o Overview: A hierarchical breakdown of the project into smaller tasks or deliverables.

o Benefits: Provides a detailed understanding of the project scope, making it easier to
assign resources and estimate time and costs.
o Use Case: Useful in both Waterfall and Agile projects to track project progress and
ensure nothing is overlooked.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 71

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Functional Decomposition

o Overview: Splits the project into functional units or modules based on the system’s
requirements.
o Benefits: Helps estimate effort based on functionality rather than processes. It’s easier
to understand and manage the system’s individual parts.
o Use Case: Often used in requirement analysis and design phases.

Product Breakdown Structure (PBS)

o Overview: Breaks down the project by its deliverables or components, focusing on the

o
estimated.

IT
physical structure of the product.
Benefits: Useful for complex systems where deliverables can be independently

Use Case: Ideal for projects focused on product delivery and hardware/software
integration.

Task Decomposition
SV
o Overview: Breaks down the project into individual tasks or activities, specifying the
work to be done.
o Benefits: Allows for precise task-level estimation, including identifying dependencies
and prerequisites.
o Use Case: Used in Agile for sprint planning or traditional projects for detailed
scheduling.

Top-Down Estimation

o Overview: Starts with an overall estimate and breaks it down into smaller components.
o Benefits: Useful in early stages of the project where details are not fully known. Gives
a high-level estimate that can be refined later.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 72

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Use Case: Often applied when historical data or expert judgment is available for
similar projects.

Bottom-Up Estimation

o Overview: Each task or component is estimated individually, and then these are rolled
up to create an overall project estimate.
o Benefits: Provides a more accurate estimate as it considers the smallest details of each
task.
o Use Case: Typically used after detailed requirements are available or in later project

o
phases.

IT
Object-Oriented Decomposition

Overview: Divides the system based on objects (i.e., entities with data and behavior),
and estimates the effort required for each object.
Benefits: Useful for object-oriented design and development, improving modularity
and reuse.
SV
o Use Case: Used in projects following an object-oriented design or for systems with
clear object-based architecture.

Module Decomposition

o Overview: The project is broken down into modules or subsystems, with each module
representing a set of closely related functionalities.
o Benefits: Helps estimate work for complex systems by focusing on individual modules,
which are easier to analyze and manage.
o Use Case: Often used in large-scale projects, especially in software that has a modular
architecture.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 73

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Process Decomposition

o Overview: Breaks down the project based on the processes or workflows involved in
the project.
o Benefits: Helps visualize and estimate effort for each process and identify
interdependencies between processes.
o Use Case: Useful in projects with clear, process-oriented workflows like business
process management or workflow systems.

Component-Based Decomposition

o IT
Overview: Focuses on identifying and estimating individual software components or
services that will be used in the system.
Benefits: Helps focus on reusable components, making estimation more accurate for
systems built on service-oriented or component-based architectures.
Use Case: Common in projects using microservices or component-based software
engineering.
SV
Benefits of Decomposition Techniques:

• Improved Accuracy: Breaking down complex projects into smaller parts allows for
more accurate estimations.
• Manageability: Decomposition helps make large projects manageable by dividing
them into smaller, understandable units.
• Transparency: Clearly defined tasks and deliverables improve visibility into the
project, helping track progress.
• Risk Identification: Easier identification of risks at a lower level, allowing for better
risk management.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 74

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Empirical Estimation Models in Software Project Estimation

Empirical estimation models are based on historical data and metrics from completed
projects. They use mathematical formulas or statistical techniques to predict the effort, time,
and cost required to complete a software project. These models are data-driven and rely on
past project experience to generate estimates.

COCOMO (Constructive Cost Model)

Overview: Developed by Barry Boehm in the 1980s, COCOMO is one of the most widely
used estimation models. It uses historical project data to estimate effort (in person-months),
time, and cost.

Three Levels:

o
IT
Basic COCOMO: Provides a rough estimate based on project size (measured in KLOC
– thousands of lines of code).
Intermediate COCOMO: Adjusts the estimate based on a set of project
SV
characteristics, including complexity, reliability, and team experience.
o Detailed COCOMO: Further refines the estimate by considering specific project
phases and modules.
• Use Case: Best suited for projects where code size is predictable. Applicable to both
new development and maintenance projects.
• Formula:

COCOMO II

Overview: An updated version of the original COCOMO model, COCOMO II is designed

for modern software development projects, including projects with object-oriented
programming, incremental development, and reuse of existing components.

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 75

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Three Submodels:
o Application Composition Model: For projects using rapid application
development (RAD) tools.
o Early Design Model: Provides estimates early in the development lifecycle
when project details are still unclear.
o Post-Architecture Model: Offers detailed estimates once the software
architecture is defined.
• Use Case: Suitable for modern software development practices like Agile, object-
oriented programming, and component-based software engineering.

IT
Function Point Analysis (FPA)

Overview: Measures the size of a software project based on its functionality from the user’s
perspective. Instead of lines of code, FPA measures the number of “function points,” which
represent various inputs, outputs, inquiries, and internal and external files.

• Steps:
1. Identify and classify system components.
SV
2. Assign a weight to each component (e.g., simple, medium, complex).
3. Calculate the total number of function points using a formula that factors in
complexity and functionality.
• Use Case: Useful when estimating the size of business applications, especially when
project requirements are well-defined but the codebase is not yet determined.
• Advantages: Technology-agnostic; focuses on functionality, making it useful across
different programming languages and environments.

Putnam Model (SLIM)

Overview: Developed by Lawrence Putnam, the Putnam Model (or SLIM – Software Life
Cycle Management) is based on the Rayleigh Curve, which describes the distribution of
manpower over the project lifecycle. It focuses on the relationship between effort, time, and
project size.
Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 76
BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

• Key Concept: Uses the idea that there is a trade-off between the time to complete a
project and the effort required. Shorter schedules require more effort, while longer
schedules require less.
• Use Case: Often used for large-scale projects where development efforts need to be
balanced with time constraints.
• Formula:

Delphi Method

• Steps: IT
Overview: The Delphi Method is a structured expert judgment technique. A group of experts
provides estimates anonymously, and then their responses are aggregated and discussed in
rounds until a consensus is reached.

1. Experts provide initial estimates.

2. Estimates are shared without revealing the identity of the experts.
SV
3. Experts revise their estimates based on the feedback from others.
4. The process repeats until a consensus is achieved.
• Use Case: Useful when historical data is unavailable or the project involves unique or
unprecedented elements. Often applied in Agile estimation.
• Advantages: Reduces the influence of bias or dominant opinions and fosters
collaboration.

Expert Judgment

Overview: This model relies on the knowledge and experience of experts to estimate project
effort, cost, and duration. It’s less formal than other models and is often used when there is
little historical data.

• Steps:

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 77

BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

o Experts review project requirements and provide estimates based on their

experience.
• Use Case: Most commonly used in the early stages of project planning when limited
information is available.
• Advantages: Flexible and easy to apply. Useful for projects that do not fit neatly into
formal estimation models.
• Challenges: Highly subjective; prone to over-optimism or underestimation.

Heuristic Models

Examples:
o

o
IT
Overview: Heuristic models use simplified rules of thumb or algorithms to generate project
estimates. These rules are often derived from past experience but may not be as formal as
COCOMO or FPA.

Estimating project effort as a fixed multiple of project size (e.g., 3 person-

months per KLOC).
Using simple ratios like development time to project size based on past projects.
SV
• Use Case: Appropriate for small or routine projects where formal models are
unnecessary or too cumbersome.

Proxy-Based Estimation (PROBE)

• Overview: Used in the Personal Software Process (PSP), PROBE relies on the idea
that future tasks can be estimated based on historical data from similar past tasks. It
uses size-based proxies (e.g., lines of code) to estimate the effort required.
• Steps:
o Identify past tasks similar to the new task.
o Use the historical data to predict the size of the new task.
o Apply a linear regression formula to estimate effort.
• Use Case: Most effective for individual developers or small teams working on well-
defined, repetitive tasks.
Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 78
BCS501 | SOFTWARE ENGINEERING & PROJECT MANAGEMENT | SEARCH CREATORS.

Parametric Estimation Models

• Overview: These models use statistical algorithms and formulas that consider various
parameters, such as team size, project complexity, and development environment, to
estimate effort and cost.
• Use Case: Suitable for projects where multiple factors affect the estimate and a more
granular approach is needed.

Benefits of Empirical Estimation Models

1. Data-Driven: Provide more reliable estimates based on historical data and proven
models.

projects.

IT
2. Repeatability: Offer consistent estimation practices that can be reused across multiple

3. Scalability: Useful for both small and large projects.

4. Risk Mitigation: Help identify potential risks early in the estimation process.
SV

Dept Of ISE Bengaluru - Navya Mukund ( SVIT ) Page 79