A methodology to evaluate object oriented software systems using change requirement traceability based on impact analysis

International Journal of Software Engineering & Applications (IJSEA), Vol.5, No.3, May 2014
DOI : 10.5121/ijsea.2014.5304 47
A METHODOLOGY TO EVALUATE OBJECT-
ORIENTED SOFTWARE SYSTEMS USING
CHANGE REQUIREMENT TRACEABILITY
BASED ON IMPACT ANALYSIS
Sunil T. D. and Dr. M. Z. Kurian
Department of Electronics and Communication Engineering,
Sri Siddhartha Institute of Technology, Tumkur, Karnataka, India
ABSTRACT
It is a well known fact that software maintenance plays a major role and finds importance in software
development life cycle. As object-oriented programming has become the standard, it is very important to
understand the problems of maintaining object-oriented software systems. This paper aims at evaluating
object-oriented software system through change requirement traceability – based impact analysis
methodology for non functional requirements using functional requirements. The major issues have been
related to change impact algorithms and inheritance of functionality.
KEYWORDS
Change Requirement Traceability, Impact Analysis, Object-Oriented Software Systems, Software
Maintenance, Change Impact algorithms, inheritance of functionality.
1. INTRODUCTION
There are several standards for traceability, such as ISO15504 and CMMI, Over the past decades,
several techniques were developed for tracing requirements. Most of the traditional techniques
like Trace-based Impact Analysis Methodology (TIAM), which is based on utilizing the trace
information and Work Product Model (WoRM) , which is to define requirement change impact
metric for determining severity in change requirements. The above methodology has predictive
value for finding classes of similar changes. TIAM which is intended for planning rather than
ensuring that changes are thoroughly implemented. TIAM potentially could be used to evaluate
the risk of volatile requirements. In case of design changes, there are cognitive consequences of
the object oriented approach. Novice designers have been found to have problems with class
creation and articulating the declarative and procedural aspects of the solution. Accordingly, here
it is to introduce traceability patterns or methods as a solution to requirement-component that can
be applied to both traditional and modern development processes. This approach has achieved as
a result of the conformance of the structure of the source code to the traceability patterns or
methods. In the software life cycle, software undergoes changes at all stages. A software product
is successful if a software changes are identified or managed from all the phases of software life
cycle, like requirement specification phase, design phase, implementation phase and maintenance
phase.

48
To obtain a software product it should be clear to have a well established threshold and it must
get higher and higher as development proceeds or no product ever appears. Software maintenance
consumes approximately forty percent of the software expenditure, since it is a non-trivial phase
in software development lifecycle capturing traceability link between code and element in
artifacts can be helpful in many tasks. Program comprehension, maintenance, requirement
tracing, impact analysis and reuse of existing software. Many number of traceability patterns or
methods were introduced to trace back elements from source code in reverse engineering.
Traceability matrix, keywords, aspect weaving, information retrieval, scenario-based, event
based, process centered, design pattern, goal centric are few examples of traceability methods.
The demand to reengineer legacy system has increased significantly with the shift toward web-
based user interface. The traceability patterns or methods are used for many reasons, such as
managing evolutionary software changes, impact analysis, software architecture. The object-
oriented paradigms such as classes and its relationship namely association, aggregations,
dependencies, multiplicity have been conducted by many researcher. The objective of this paper
is to create and provide round-trip engineering capability during traceability process.
Organisation: The literature survey about the related topic is dealt in section 2. Section 3 deals
with the types of Traceability models. Impact analysis based on Change requirement traceability
is discussed in section 4. The research results are presented in section 5. The paper is concluded
mentioning the conclusive remarks in section 6.
2.LITERATURE SURVEY
There are a number of phases in the life of a software product. The waterfall model, as described
by Ghezzi et al., [1], has five major phases. They are requirements analysis and specifications,
coding and module testing, integration testing, system testing and maintenance. This research is
concerned only with the final aspect of the final phase, maintenance. The maintenance phase is
the longest phase of the life cycle. Maintaining software becomes more difficult as time
progresses and the system evolves. Chandra Shrivastava et al., [2] stated that the algorithms
calculate the transitive closure of each of the potentially effected classes and methods. It will be
possible to greatly improve upon the information provided by the algorithms in recognition of
low-level design patterns, effects of data type changes, and effects of addition and deletion of
classes can be drawn from the LLSA model of an object-oriented system. Chen. X., Tsai et al.,[3]
presented an integrated environment for C++ program maintenance which describes three new
dependence graphs specific to object-oriented software systems: message, class and declaration
dependence in a model called C++ DG. Additionally, several new slicing techniques are
presented. The use of the new dependencies and slicing on code maintenance is described. The
dependencies are described, specifically as to the ripple effect and regression testing. The
application of the discovered dependencies and program slicing leads to recursive analysis of the
ripple effect caused by code modification. As the effects are located, classes and methods affected
can be “marked” for testing or re-execution in the testing phase.
Li.,L et al., [4] explained four algorithms that measure the effect of proposed changes to object-
oriented systems. The ripple effect is calculated by application of algorithms that
1. calculate the change effects inside of a class
2. calculate the change effects among clients
3. calculate the change effects among subclasses
4. measure the total effect by driving the algorithms in 1,2 and 3

49
The author also presented the details of how different types of changes affect the system. Changes
are broadly categorized as method or member change, and then refined to more detail such as
adding a member or changing an attribute. Gallangher, K [5] described about program slicing to
select a point in an ANSI C program for observation. The method looks at program variables and
essentially models dependencies that exist among variables via assignment statements and
parameter passing. The method is a visualization of the data collected by the Surgeon’s Assistant
and is called the Decomposition Slice Display System. According to Hutchins et al., [6] Visual
Impact Analysis has improved the recognition of further dependencies such as interference.
Bohner.S.A [7] presented that software engineering practice evolves to respond to demands for
distributed applications on heterogeneous platforms; software change is increasingly influenced
by middleware and components. Interoperability dependency relationships now point to more
relevant impacts of software change and necessarily drive the analysis. Software changes to
software systems that incorporate middleware components like Web services expose these
systems and the organizations they serve to unforeseen ripple effects that frequently result in
failures. Current software change impact analysis models have not adequately addressed this
trend. Moreover, as software systems grow in size and complexity, the dependency webs of
information extend beyond most software engineers’ ability to comprehend them. This paper
examines preliminary research for extending current software change impact analysis to
incorporate interoperability dependency relationships for addressing distributed applications and
explores three dimensional (3D) visualization techniques for more effective navigation of
software changes. Pressman [8] explained that as software system becomes larger and more
complex, numerous corrections, extensions and adaptations tend to be more chaotic and
unmanageable. The traditional way of addressing the maintenance task individually is no longer
practical. It needs a special management system, called the Software Configuration Management
(SCM) that covers the procedures, rules, policies and methods to handle the software evolution
(IEEE, 1998b). SCM has been identified as a major part of a well defined software development
and maintenance task. SCM deals with controlling the evolution of complex software systems
that supports version controls and administrative aspects such as to handle change requests, and to
perform changes in a controlled manner by introducing well-defined processes. Suhaimi Bin
Ibrahim [9] illustrates that most of the Computer Aided Software Engineering (CASE) tools and
applications focuses on the high level software and yet are directly applicable to software
development rather than maintenance. While the low level software, e.g. code is given less
priority and very often left to users to decide. This makes the software change impact analysis
extremely difficult to manage at both levels. Secondly, there exists some research works on
change impact analysis but the majority confine their solution at the limited space i.e. code,
although more evolvable software can be achieved at the meta model level. No proper visibility is
being made by the ripple effects of a proposed change across different levels of work product. If
this can be achieved, a more concrete estimation can be predicted that can support change
decision, cost estimation and schedule plan. M.Z.Kurian et al., [10] explained a comparative
software maintenance methodology to assist in Object Oriented systems was carried out with
main intention regarding to impact analysis and ripple effect to retesting of affected and changed
components. This reduces the cost of testing and assists in identifying change impact in object-
oriented maintenance. Since, it does not emphasize on the change requirement analysis and
tracing object oriented software system it is to look forward with other methods. Ali R. Sharafat
et al., [11] proposed an estimation of change-proneness of parts of a software system is an active
topic in the area of software engineering. Such estimates can be used to predict changes to
different classes of a system from one release to the next. They can also be used to estimate and
possibly reduce the effort required during the development and maintenance phase by balancing
the amount of developers’ time assigned to each part of a software system. This is a novel
approach to predict changes in an object-oriented software system. The rationale behind this
approach is that in a well-designed software system, feature enhancement or corrective
maintenance should affect a limited amount of existing code. The goal is to quantify this aspect of

50
quality by assessing the probability that each class will change in a future generation. Peter
Zielczynski [12] explained an approach which is applied to software writing in an object-oriented
language to trace object oriented code into functional requirements. Here, it is addressed the
problem of establishing traceability links between the free text documentation associated with
development and maintenance cycle of a software system and its code. Further, vector space
models to compare different model and to assess the relative influence of affecting factors are not
considered.
In this paper, based on the requirement management to maintenance is considered so that change
requirement traceability analysis is done on the requirement as well as object-oriented software
systems and a round-trip traceability analysis is performed.
3. TRACEABILITY MODELS
Requirement traceability refers to the ability to describe and follow the life of a requirement, in
both a forwards and backward direction. Forward traceability is the ability to trace a requirement
to components of a design or implementation. Backward traceability is the ability to trace a
requirement to its source that is, to a person, institution, law, argument etc. Inter-requirements
traceability refers to the relationships between requirements. Inter-requirement traceability is
important for requirement analysis and to deal with requirements change and evolution Francisco
A et al., [13]. Extra-requirements traceability refers to the relationships between requirements
and other artifacts.
4. CHANGE REQUIREMENT TRACEABILITY BASED IMPACT
ANALYSIS
It is the result of the elicitation process Gotel O.C.Z et al., [14]. The tracing of a requirement can
be done in either way, to get information related to the process of elicitation, prior to its inclusion
in the requirements specification or to get information related to its use, after the requirement has
been elicited and included in the requirement. It has pre-requirements specification traceability
and post-requirement traceability specification traceability. Pre- requirement traceability refers to
those aspects of a requirements life prior to its inclusion in the requirement specification. Post
RS-traceability refers to those aspects of a requirements life that result from inclusion in the
requirement specification. Pre RS-traceability is used, when there is a change to a requirement
and when to get the requirements source or people supporting it to validate change. Post RS
traceability is used to get the design module to which a requirement was allocated or the test
procedures created to verify the requirements.
Change Requirement Traceability Based Impact Analysis is a Non-functional tracing and
Informal tracing that is, in functional tracing, those related to well establish mapping between
objects model types and mapping types which allow analysis models, design models, process
models, organizational models. The Non-functional tracing is related to the tracing of non-
functional aspects of software development. They are usually related to quality aspects and
results from relationships to non-tangible concepts. The traces that related requirements to goals,
objectives, intensities and decisions are example of non-functional tracing. Non-function tracing
are classified into four categories like reason, context, decision and technical.
The tracing of non-functional aspects of software development can be automatically performed
only using a representation of that aspect. Therefore, here it is to use some model to functionally
capture the non-functional aspects we want to trace, it may use an organizational model to relate
policies, goal and roles to requirements, or it may be used process model to relate requirements

51
to activities and resources. It is also an informal need for trace definition. The definition of
traces and traceable objects should promote their uniform understanding. Differences and
interpretation are the causes of errors, and in the more serious cases once may end up tracing
what did not happen. To account for non-functional traces, the definition of traceable objects
should allow the use of hyper-media objects like videos, recording and images together with
mechanism for inspecting these kinds of objects. The relationship between recorded real world
observations and parts of conceptual model is called extended traceability Haumer P et al., [15]
Smith t et al., [16] Yu W.D. [17]. Sarah Maadawy et al., [18] presents a methodology to
measure software complexity for changes. It studies attributes that affect complexity of change
and the relation between requirements and each other to finally find a complexity measure the
will serve in finding a precise estimate for the change. However, it did not discuss the object-
oriented analysis and design aspects.
In this paper, the change requirement traceability based impact analysis methodology has been
discussed, which is for a non-functional and informal and extended traceability, also object-
oriented analysis and design aspects are discussed. This paper discusses the following phase’s i.e.
Phase One
A. Validating the new requirements from any of the stake holders.
B. Classification of requirement whether functional or non-functional requirement
C. Traceability matrix can help tracing the requirement
D. Review of the Requirements
E. Requirement Evaluation
F. Requirement Documentation
G. Acceptance Testing
Phase Two
A. Stability: Unstable Requirements
B. Completeness: Incomplete Requirements
C. Clarity: Unclear Requirements
D. Validity: Invalid requirements
E. Feasibility: Infeasible requirements
F. Precedent: Unprecedented Requirements
Stability:
This represents the system vulnerability to change. It has been noticed that software
maintainability degrades as changes are made to it which increases complexity of the software,
system stability will be calculated as in
S(#NORS + #NOCNR + #NOCUR + #NOCDR) / (#NORS)
Where S= Stability and
NORS = No. of original requirement in the system
NOCNR=No. of cumulative number of requirement
NOCUR = No. of cumulative number of requests updated in the system
NOCDR=No. of cumulative number of request deleted from the system.

52
Completeness
This represents completeness of the requirement
CMP=NARS –N IR
CMP=Completeness of the system
NARS=No. of Actual / original requirement in the system
NIR =Number of Incomplete Requirement in the system
Clarity
This represents clarity of the system.
CL=NARS –N IR-UCLR
CL=Clarity of the system
NARS=No. of Actual / original requirement in the system
NIR =Number of Incomplete Requirement in the system
UCLR=No. of Unclear requirements
Feasibility
This represents feasibility of the system.
FR=IFR -UCLR
FR=Feasibility requirements of the system
IFR =Number of Infeasible Requirement in the system
UCLR=No. of Unclear requirements
Precedent
This represents precedent of the system.
PR=CMP+CL+FR
PR=Precedent requirements of the system
CMP =Completeness of the system
CL=Clarity of the system
FR=Feasibility of the system
5. RESULTS
Case Study: Flight Booking System
Here it is to identify, visualize and analyze the change requirement traceability analysis on object-
oriented software system. Here, Flight Booking System case study has been taken as a
requirement. Based on the requirement level in, it is to split requirement into different
requirement types
1. Stakeholders need
2. Feature
3. Use Case
4. Supplementary Requirement
5. Test Cases

53
6. Scenarios
1.The requirements at the top level of the levels (stakeholders’ requests) are gathered using
various methods of knowledge elicitation:
• Asking Questions
• Conducting Workshops
• Hearing to stories
• Role
• Brainstorming
• Module
• Use cases
• Analysis of existing documents
• Observation, task demonstration
• Analysis of existing systems
2. A business analyst derives the second level of the levels (features) from stakeholders’ requests
by cleaning the requirements and translating them from the problem domain to the solution
domain. The features should have all the attributes of a good requirement:
• Testable
• Clear
• Correct
• Understandable
• Feasible
• Independent
• Atomic
• Necessary
• Implementation-free
• Consistent
• Non-redundant
• Complete
To fix the requirements that are missing at least one of these attributes, which can apply some of
the following transformations:
• Copy
• Split
• Clarification
• Qualification
• Combination
• Generalization
• Cancellation
• Completion
• Correction
• Unification
• Adding details
3. The third layer of the levels contains use cases and supplementary requirements. Use cases
capture functional requirements. Creation of use cases consists of the following steps:
1. Identify actors.

54
2. Identify use cases.
3. Design the initial use case model.
4. Structure the model.
5. Create use case documents.
4. Supplementary requirements capture mostly non functional requirements. They may also
capture some generic functional requirements not associated with any specific use cases.
Supplementary requirements can be classified as follows:
• Functionality
• Usability
• Reliability
• Performance
• Testability
• Design constraints
• Implementation requirements
• Interface requirements
• Physical requirements
• Documentation requirements
• Licensing and legal requirements
5. Test cases are created to test the requirements from the third level. The following steps are used
to derive test cases from use cases:
• Create scenarios.
• Identify variables for each use case step
• Identify significantly different options for each variable
• Combine options to be tested into test cases
• Assign values to variables
6. To create test cases from supplementary requirements, you can use one of the following
approaches:
• Execute selected functional test cases in different environments
• Add checks to all use cases
• Check and modify a specific use case
• Perform the exercise
• Checklist
• Analysis
• White-box testing
• Automated testing
7. Design diagrams are also derived from the requirements on the third level, especially Use
cases. Here are the possible approaches:
• Design classes that will capture required data and functionality
• Create one sequence diagram for each scenario
• Simultaneously add required methods and attributes to the classes on the class Diagrams
8. Documentation is created from various elements of the levels.

55
Algorithms for ‘Book a Flight’
Step 1: Begin Algorithm
Step 2: Enter URL
Step 3: Enter flight data search flights
Step 4: Select a flight
Step 5: System Display return flights
Step 6: System Display details of flights
Step 7: Confirm the flight
Step 8: New User Register
Step 9: Login
Step 10: Provide passenger information
Step 11: Display available seats
Step 12: Select Seats
Step 13: Enter Billing information
Step 14 : Provide confirmation number
Step 15: End algorithm
Use Cases
Figure 1.1 An ACTOR and a use case
Figure 1.2 Use case Initiated by Travellers and User

56
Structuring the Use Cases
The main purpose of structuring the model is to remove any redundancy, making the use
cases easier to understand and maintain. First, it is needed to analyze use cases and find
any parts of the flows that contain similar steps. Then it is to apply some of the three
types of relationships between use cases:
• Include
• Extend
Include Relationship
The included use case should be self-contained and cannot make any assumptions about
which use case is including it.
Figure 1.3 an include relationship between two use cases
Extend Relationship
If some part of the use case is optional or conditional, to make the model more clear,
extract it as a separate use case that is connected with an extend relationship.
Figure 1.4 an extend relationship between two use cases
Figure 1.5 A Context diagram for the Use case book a flight

57
Traceability Structure
Figure 1.6 shows traceability structure in this case study
Figure 1.6 Traceability structure for case study “Book a Flight”
• Stakeholder Requests (STRQ) will be traced to Features (FEAT) defined in the Vision
document and supplementary Requirements defined in the Supplementary Specification. There
may be a many-to-many relationship between STRQ and FEAT, but usually it is one Stakeholder
Request to many Features. Every approved Request must trace to at least one Feature or
Supplementary Requirement.
• Feature Requirements (FEAT) (defined in the Vision document) will be traced to either a Use
Case or Supplementary Requirement. Every approved feature must trace to at least one Use Case
or Supplementary Requirement. There may be many-to-many relationships
between Features and Use Cases and Supplementary Requirements.
• Use Case Requirements (UC) defined in the Use Case Specifications will be traced back to
Features.
• Supplementary Requirements (SUPL) will be traced back to Features.
Object Oriented System Design from Use Cases

58
Figure 1.7 A class diagram showing classes that implement the functionality of a basic flow of the Book a
flight use case.
Figure 1.8 Class Reservation and related classes
6. CONCLUSION
The Change requirement traceability of a case study “book a flight” requirement provides an
object oriented approach architecture till a class diagram. The object oriented analysis has been
done for the case study, “book a flight” and arrived at a class diagram by using ‘use cases’ and
arrived at an algorithm for “book a flight’ case study. The proposed analysis is highly dependent
on a very well defined software requirements specification and non-functional traceable
requirement. Further based on change in the requirements the impact on the class diagram till test
case attributes can be identified. Based on the requirement traceability which test cases must be
changed can be identified and also impact of object-oriented paradigms can be analyzed.

59
References
[1] Carlo Ghezzi, Mehdi Jazayeri, Dino Mandrioli, Fundamentals of Software Engineering, Prentice Hall
Publishing, (1991).
[2] Chandra Shrivastava, D. L. Carver, "Using Low-Level Software Architecture for Software
Maintenance of Object-Oriented Systems", Proceedings of the 1995 Software Engineering Forum,
Boca Raton, FL, November pp. 31-40, (1995).
[3] Chen. X., Tsai. W., Hunag. H., Poonawala. M., Rayadurgam. S., Wang. Y., , Omega-an Integrated
Environment for C++ Program Maintenance, Proceedings of the International conference on software
Maintenance, pp.114-123, (1996).
[4] Li.,L.,Offutt,A.J., Algorithmic Analysis of the Impact of Changes to Object-oriented Software,
Proceedings of the International Conference on Software Maintenance, pp. 171-184, (1996).
[5] Gallangher, K., Visual Impact Analysis, Proceedings of the International Conference on Software
Maintenance, pp. 52-58, (1996).
[6] Hutchins,M., Gallagher,K., Improving Visual Impact Analysis, Proceedings of the International
Conference on Software Maintenance, pp.294-301, (1996).
[7] Bohner.S.A., Software change impacts–an evolving perspective, Proceedings of the International
Conference on Software maintenance, pp 263 – 272, (2002).
[8] Pressman. A Dynamic Analysis Approach Concept Location. Technical report of Software
Engineering, (2004).
[9] Suhaimi Bin Ibrahim A Document-Based Software Traceability to Support Change Impact Analysis
of Object-Oriented Software, University Teknologi Malaysia, Thesis, pp. 45-56, (2006).
[10] M.Z.Kurian and A S Manjunath Requirement traceability and impact analysis methodology to
evaluate software requirements changes, National Conference on Trends in Advanced Computing, at
DMCE,Airoli,Navi Mumbai,28-29, (2007).
[11] Ali R. Sharafat and Ladan Tahvildari, Change Prediction in Object- Oriented Software Systems: A
Probabilistic Approach, Journal of Software, Vol. 3, No. 5, pp.10-38, (2008).
[12] Peter Zielczynski, IBM, Requirements Manangement Using IBM Rational Requisite Pro, (2013).
[13] Francisco A C Pincher, Requirement traceability Technical Report, University of Brasilia, (2000)
[14] Gotel O.C.Z and Finkelstein ACW. An analysis of the requirements traceability problem.
Proceedings of ICRE94, 1st Internation conference on requirements engineering, 1994, Colorado
springs Co, IEEE CS Press (1994).
[15] Haumer P ., Pohl K., Weidenhaupt K and Jarke M . Improving reviews by extended traceability.
Proceedings of 32nd Hawaii International Conference on system sciences volume 3; January 05-08;
Maui, Hawaii, (1999).
[16] Smith t,T J READS: A requirements engineering tool. Proceedings of RE’93, International
Symposium on Requirements Engineering; January 4-6; san Diego,C.A. Los Alamitos,CA,IEEE
Computer Society,(1993).
[17] Yu W.D. Verifying software requirements – a requirement tracing methodology and its software tool
– RADIX, IEEE Journal on Selected Areas of Communication;12(2):234-240, (1994)
[18] Sarah Maadawy and Akram Salah, Measuring Change Complexity from Requirements: A proposed
methodology, IMACST Volume 3, Number ,Feburary (2012).
Authors
Asst. Prof. Sunil T D received Bachelor Degree from Bangalore University in Electronics
and Post graduate degree in Electronics from Visvesvaraya Technological University at
BMSCE, Bangalore and Pursuing Ph.D degree in Software Engineering from
VTU, Belgaum, Karnataka, India. Having 12 Years of Teaching experience in the field
of Electronics & Communication Engineering. Published several papers in peer reviewed
international journals, and several conference papers.

60
Dr M.Z.Kurian received his Bachelor Degree from Bangalore University and Post graduate
degree in Industrial Electronics from Mysore University, and Ph.D degree in Software
Engineering from Dr.MGR University, Chennai, Tamil Nadu, India. He has more
than 30 Years of Teaching experience in the field of Electronics & Communication
Engineering. Published several papers in peer reviewed international journals including
IEEE, and several conference papers.

A methodology to evaluate object oriented software systems using change requirement traceability based on impact analysis

Recommended

More Related Content

What's hot (17)

Viewers also liked (20)

Similar to A methodology to evaluate object oriented software systems using change requirement traceability based on impact analysis (20)

Recently uploaded (20)

A methodology to evaluate object oriented software systems using change requirement traceability based on impact analysis