Content uploaded by Sayed Mohsin Reza
Author content
All content in this area was uploaded by Sayed Mohsin Reza on Dec 31, 2020
Content may be subject to copyright.
Activity based New Technique of Effort & Cost Estimation
using Functional Measurement Type for Web Application
by
Sayed Mohsin Reza (Roll: 150102)
A Thesis submitted in partial
fulfillment of the requirements for the degree of
Master of Science(M.Sc.) in Information Technology
Supervisor: Dr. M. Shamim Kaiser
Institute of Information Technology
Jahangirnagar University
Savar, Dhaka-1342
October 2016
c
Sayed Mohsin Reza 2016
All Rights Reserved
To my Beloved Parents
ii
DECLARATION
I hereby declare that this thesis is based on the results found by myself. Materials
of work found by other are mentioned by References. This thesis, neither in whole
nor in part, has been previously submitted for any degree.
Sayed Mohsin Reza
Roll: 150102
Reg: 32029
Session: 2014-15
iii
CERTIFICATE
This is to certify that the Thesis entitled Activity based New Technique of
Effort & Cost Estimation using Functional Measurement Type for Web
Application has been prepared and submitted by Sayed Mohsin Reza in partial
fulfilment of the requirements for the degree of Master of Science(M.Sc.) in Informa-
tion Technology. He embodies original work under my supervision to the best of my
knowledge.
(Signature in full of The Supervisor)
Dr. M Shamim Kaiser
Associate Professor,
Institute of Information Technology,
Jahangirnagar University.
Accepted and approved in partial fulfillment of the requirements for the degree of
Master of Science in Information Technology.
Md. Fazlul Karim Patwary K M Akkas Ali
Chairman Member
Jesmin Akhter Dr. Muhammad Mahbub Alam
Member Member (External)
iv
ACKNOWLEDGEMENTS
First of all I would like to thank the Almighty for giving me the opportunity to
complete this work successfully. My acknowledgement is meant to express my sin-
cere gratitude to all those people who have been associated with this thesis and have
helped me with it and by sharing their opinions and experiences through which i re-
ceived the required information crucial for my thesis. I am thankful to my parents for
their relentless support. Most importantly i am tremendously grateful to Shamim
Al Mamun,co-supervisor and Assistant Professor, Institute of Information Tech-
nology, Jahangirnagar University who took time out of his hectic schedule to guide
me at the beginning of this work and provide me with all the necessary materials and
sufficient knowledge that was the major requirement. Finally, I express my thanks to
the honourable Supervisor Dr. M. Shamim Kaiser for giving me the opportunity
to learn the thesis area in a practical approach.
Many thanks to the faculty members of IIT, Md. Atiqur Rahman, Project Man-
ager, Eicra Soft Limited for helping me by providing feedbacks, monitoring and tech-
nical support. Thanks to Md. Mahfujur Rahman, Md. Hasnat Parvez, Iqbal Chowd-
hury, Md. Saeed Siddik, Shihab Uddin, Rashed Ibna Anowar and Abidul Islam for
their partial cooperation.
This research is partially supported by National Science & Technology (NST)
Fellowship provided by Ministry of Science and Technology Government of
the People’s Republic of Bangladesh.
v
ABSTRACT
Software Effort Estimation helps Project Leaders to circulate assets, control spend-
ing plans, motivation and create advanced works on, prompting tasks finished on time
and within budget related arrangement.
But the challenge is when a project leaders try to find out the software effort
based on some criteria, reasonable events and caused can be missed; while hopeful
expectations can be influenced to some asset losing. Because of fast change in in-
novation, usage of complex Software frameworks at less expense and the inclination
to keep up better quality software are a portion of the significant difficulties for the
Software organizations.
The limitation as well as the toughest works is effort estimation, in the field of
software engineering. It is the estimation of total effort required in developing soft-
ware. In terms of Software development, these issues are essential and extremely
difficult in Software Development with short timetables.
Since Software Development activities are constantly changed in nature, earlier
tasks may not cover all parts of new software development. The thesis motivation is
come from the Specialists in software engineering who have proposed different tech-
niques for effort & cost estimation. From this motivation, This thesis paper at first
tries to give an understanding into the different models and systems utilized as a part
of evaluating effort of the project. It additionally concentrates on a portion of the
pertinent reasons that cause incorrect estimation. Early Software Estimation models
depend on Regression Tree or Numerical Inferences.
In this thesis paper, The development process is also particularly followed by the
Expert based Estimation and Algorithmic based Estimation as the flow is controlled
by the System Administrator. This Thesis paper objective is to propose a way to
deal with build up the correctness of Software Effort Estimation utilizing Dataset of
a Functional Measurements and Algorithmic Guidelines. It also shows a model to
foresee the Parametric effort estimation. The advantages and disadvantages of the
current effort estimation process have been highlighted in this thesis paper. There
is as such not any single strategy which can be viewed as the best technique so in
this paper it is proposed that a mix of the strategies ought to be utilized to get an
estimated effort estimation.
vi
TABLE OF CONTENTS
DEDICATION .................................. ii
DECLARATION ................................. iii
CERTIFICATE .................................. iv
ACKNOWLEDGEMENTS .......................... v
ABSTRACT ................................... vi
LIST OF FIGURES ............................... xi
LIST OF TABLES ................................ xiii
LIST OF ABBREVIATIONS ......................... xiv
CHAPTER
I. Introduction .............................. 1
1.1 Introduction ........................... 1
1.2 Motivation ............................ 2
1.3 ResearchQuestion........................ 2
1.4 Research Contribution . . . . . . . . . . . . . . . . . . . . . . 3
1.5 Socio-Economic Importance . . . . . . . . . . . . . . . . . . . 3
1.6 Organization of Thesis . . . . . . . . . . . . . . . . . . . . . . 4
II. Literature Review ........................... 5
2.1 Introduction to Effort Estimation . . . . . . . . . . . . . . . . 5
2.2 History .............................. 5
2.3 Estimation Approaches . . . . . . . . . . . . . . . . . . . . . 6
2.4 Assessing the Accuracy of Estimates . . . . . . . . . . . . . . 6
2.5 Project Management Software . . . . . . . . . . . . . . . . . 7
III. Background & Nature of Software Application ......... 8
vii
3.1 Introduction ........................... 8
3.2 Difference about Web and Software Projects . . . . . . . . . 8
3.3 Introduction to Web Effort Estimation . . . . . . . . . . . . . 9
3.4 Accuracy of an Effort Model . . . . . . . . . . . . . . . . . . 10
3.5 Sizing Web Applications . . . . . . . . . . . . . . . . . . . . 11
IV. Review on Cost Estimation Techniques for Web Projects . . 13
4.1 Web Effort Estimation Using Regression Analysis . . . . . . . 13
4.2 Web Effort Estimation Using Case-Based Reasoning . . . . . 14
4.3 Web Effort Estimation using Classification Trees . . . . . . . 15
4.4 How to Improve Your Company Effort Estimation Practices . 15
V. Survey .................................. 17
5.1 Introduction ........................... 17
5.2 Questionnaires .......................... 18
5.2.1 Questionnaire construction issues . . . . . . . . . . 18
5.2.2 Types of questions . . . . . . . . . . . . . . . . . . . 20
5.2.3 Question sequence . . . . . . . . . . . . . . . . . . . 21
5.3 Final Effort Survey Questionnaires . . . . . . . . . . . . . . . 21
5.4 Surveysampling ......................... 23
5.4.1 Probability sampling . . . . . . . . . . . . . . . . . 23
5.4.2 Bias in probability sampling . . . . . . . . . . . . . 23
5.4.3 Non-probability sampling . . . . . . . . . . . . . . . 24
5.5 Survey data collection . . . . . . . . . . . . . . . . . . . . . . 24
5.5.1 Modes of data collection . . . . . . . . . . . . . . . 25
5.5.2 Advantages of online surveys . . . . . . . . . . . . . 25
5.6 Final Survey Results . . . . . . . . . . . . . . . . . . . . . . . 26
VI. System Process Diagram ....................... 28
6.1 System Sequence Diagram (SSD) . . . . . . . . . . . . . . . . 28
6.2 ActivityDiagram......................... 28
6.3 UseCaseDiagram ........................ 29
6.4 Flowchart............................. 29
VII. Research Methods ........................... 33
7.1 Introduction ........................... 33
7.1.1 Non Algorithmic Based Estimation Methods . . . . 34
7.1.2 Algorithmic Method . . . . . . . . . . . . . . . . . . 34
7.2 ProposedSystem......................... 35
7.3 Systemoverview ......................... 36
viii
7.3.1 Numerous deployment options . . . . . . . . . . . . 36
7.4 Methodology ........................... 36
VIII. Effort Calculation Analysis ..................... 38
8.1 Effort Calculation Method Development . . . . . . . . . . . . 38
8.1.1 Workflow ....................... 38
8.1.2 DataSchema...................... 39
8.1.3 Why Linear Properties . . . . . . . . . . . . . . . . 40
8.1.4 Why Logarithm Properties used in the Program . . 41
8.2 EffortFactor ........................... 41
8.2.1 Effort Factor Tree Dependency . . . . . . . . . . . . 41
8.3 Details of Effort Factor . . . . . . . . . . . . . . . . . . . . . 41
IX. Implementation ............................ 45
9.1 Requirements........................... 45
9.2 System User Section . . . . . . . . . . . . . . . . . . . . . . . 45
9.2.1 Login Portal for System User . . . . . . . . . . . . . 45
9.2.2 Repositories View by Project Manager . . . . . . . 46
9.2.3 Task List of Individual User . . . . . . . . . . . . . 47
9.3 System Employee Portal . . . . . . . . . . . . . . . . . . . . . 47
9.3.1 Employees List in the System . . . . . . . . . . . . 47
9.3.2 Add Employee for the system . . . . . . . . . . . . 48
9.4 Project-Task Related Section . . . . . . . . . . . . . . . . . . 48
9.4.1 Project Details . . . . . . . . . . . . . . . . . . . . . 48
9.4.2 Add a New Project . . . . . . . . . . . . . . . . . . 49
9.4.3 TaskDetails ...................... 50
9.4.4 AddIssue ....................... 50
9.5 OtherSection........................... 51
9.5.1 AboutUI........................ 51
9.5.2 FAQ .......................... 52
X. Results and Discussion ........................ 53
10.1 Evaluation Criteria . . . . . . . . . . . . . . . . . . . . . . . . 53
10.2Results .............................. 54
10.2.1 Effort Calculation . . . . . . . . . . . . . . . . . . . 54
10.2.2 Cost Calculation . . . . . . . . . . . . . . . . . . . . 56
10.2.3 Final Results . . . . . . . . . . . . . . . . . . . . . . 56
XI. Conclusion & Future Work ..................... 57
11.1Conclusion ............................ 57
11.2FutureWork ........................... 58
ix
References ..................................... 59
A. List of Included Papers ........................ 62
1.1 List of Journals Organization . . . . . . . . . . . . . . . . . . 62
1.2 Journal Papers on Software Effort Estimation . . . . . . . . . 63
B. Questionnaire ............................. 65
2.1 Questionnaire list was used in the survey . . . . . . . . . . . 65
x
LIST OF FIGURES
Figure
2.1 Effort Estimation Approaches . . . . . . . . . . . . . . . . . . . . . 6
3.1 Effort Estimation Accuracy Process . . . . . . . . . . . . . . . . . 11
6.1 SequenceDiagram............................ 29
6.2 ActivityDiagram ............................ 30
6.3 UseCaseDiagram............................ 31
6.4 Flowchart ................................ 32
8.1 Effort Factor Tree Dependency . . . . . . . . . . . . . . . . . . . . 42
9.1 Login Portal for System User . . . . . . . . . . . . . . . . . . . . . 46
9.2 Repositories View by Project Manager . . . . . . . . . . . . . . . . 46
9.3 Task List of Individual User . . . . . . . . . . . . . . . . . . . . . . 47
9.4 System Employees List . . . . . . . . . . . . . . . . . . . . . . . . . 47
9.5 Add Employee UI for the system . . . . . . . . . . . . . . . . . . . 48
9.6 ProjectDetails ............................. 49
9.7 AddaNewProject ........................... 49
9.8 TaskDetails............................... 50
9.9 AddIssue ................................ 51
9.10 About .................................. 51
xi
9.11 FAQonPROCOST........................... 52
xii
LIST OF TABLES
Table
2.1 List of Project Management Software . . . . . . . . . . . . . . . . . 7
3.1 EstimationWay............................. 10
7.1 Comparison with Existing System . . . . . . . . . . . . . . . . . . . 34
7.2 Proposed System Process . . . . . . . . . . . . . . . . . . . . . . . . 35
7.3 FeatureComparison........................... 37
8.1 Status of the Task (ST) . . . . . . . . . . . . . . . . . . . . . . . . . 42
8.2 Job Types / Job Position (JP) . . . . . . . . . . . . . . . . . . . . . 43
8.3 Project Basis Roles (PBR) . . . . . . . . . . . . . . . . . . . . . . . 43
8.4 Assigned Role of a task (AR) . . . . . . . . . . . . . . . . . . . . . 43
8.5 Duration Type of the task (DTT) . . . . . . . . . . . . . . . . . . . 44
8.6 Dependency of another Task (DAT) . . . . . . . . . . . . . . . . . . 44
10.1 Difference in Actual & Predicted Effort . . . . . . . . . . . . . . . . 55
10.2 CostCalculator ............................. 56
xiii
LIST OF ABBREVIATIONS
4GL Fourth-Generation Programming Language
AD Active Directory
ASM Algorithmic State Machine
BI Business Intelligence
BRM Business Reference Model
CAD Computer-Aided Design
CSS Cascading Style Sheets
DBA Database Administrator
DBMS Database Management System
DML Data Manipulation Language
EOF End of File
FOSS Free and Open Source Software
FS File System
GPL General Public License
GUI Graphical User Interface
HTML Hypertext Markup Language
JS JavaScript
LAMP Linux Apache MySQL PHP
MVC Model-View-Controller
OLAP Online Analytical Processing
OOP Object-Oriented Programming
xiv
OSS Open-Source Software
QoS Quality of Service
regexp Regular Expression
SaaS Software as a Service
SCM Source Code Management
SDK Software Development Kit
SPM Software project management
SQL Structured Query Language
SSD Software Specification Document
UML Unified Modeling Language
URL Uniform Resource Locator
usr user
var variable
W3C World Wide Web Consortium
XAML eXtensible Application Markup Language
xv
CHAPTER I
Introduction
1.1 Introduction
The topic of this Masters thesis is effort estimation in software development in
the context of Agile self-organizing teams using functional measurement type. The
focus of the thesis is on the level of individual tasks within the teams. Different effort
estimation methods are inspected in the literature review to determine what kind of
effort estimation methods there are, what to take into consideration and when they
should be used to come up with the most accurate estimates possible.
The thesis is done for 3 largest projects. A survey was conducted to select factors
of Effort Estimation to find out how Scrum teams working on different software at 3
projects estimate effort for their tasks. As Scrum teams organize their work indepen-
dently, there can be differences between the teams effort estimation processes. Based
on the current state of the practice and the results from literature review, suggestions
for improvement are presented.
Software cost models and effort approximations support project supervisors to
distribute resources, control budgets and agenda and develop modern practices, lead-
ing to projects completed on time and within financial plan. If cost and effort are
determined suspicious in software projects, suitable occasions can be missed; whereas
expectant predictions can be affected to some resource losing. In the context of web
development, these issues are also vital and very challenging given that web projects
have short schedules and very fluidic opportunity. Since software projects are con-
tinually changed in nature, earlier projects may not necessarily cover all aspects of a
new project when used as a basis for cost estimation. Preliminary software estimation
models are constructed on regression analysis or mathematical sources.
Algorithmic techniques rely on mathematical equations to estimate software ef-
fort & cost. Constructive Cost Model (COCOMO) is a popular and widely used
Algorithmic set of models. Algorithmic methods have many advantages but at the
same time these methods are hard to learn and too much data are needed about the
current project state in these methods. Conversely Non-Algorithmic techniques are
1
easy to learn but we need to have complete information about one of the very similar
previous projects as compared to our current software project, as estimation in these
are made on the basis of historical data. So basically there is as such not any single
method which can be regarded as the best one. So a combination of the methods is
usually suggested to arrive at a better cost estimate.
1.2 Motivation
The thesis motivation is come from the Specialists in software engineering who
have proposed different techniques for effort & cost estimation. From this motivation,
This thesis paper at first tries to give an understanding into the different models and
systems utilized as a part of evaluating effort of the project. It additionally concen-
trates on a portion of the pertinent reasons that cause incorrect estimation. Early
Software Estimation models depend on Regression Tree or Numerical Inferences.
The Expert Based Estimation algorithms are applied to find the final result of
potential effort of a project is one of the most inspiring work and a difficult task.
By using PROCOST software tools, large volumes of effort data are being composed
and readily available to the actual effort analysis. The Expert based Estimation
procedures have become a well-known research instrument for actual effort analysis to
calculate effort and cost for a project. Different techniques of Estimation Algorithms
use different purpose of uses. The procedures contribute some of its own advantages
and disadvantages. In Parametric Effort Estimation, factors shows a vital role in
order to scrutinize the supervised effort. Effort Factors is a supervised calcualtion
process and its objectives are predefined.
1.3 Research Question
The research problem for the thesis is threefold. Firstly, an overview of the current
state of the art of effort estimation methods is needed. Secondly, it is necessary to
determine the current state of the practice of effort estimation in the Scrum teams
working. Finally, suggestions for improvement are made by combining the findings
from the literature review with the data of the current situation. Based on this, the
following research questions can be formulated:
RQ-1: What is the state of the art in software development effort estimation?
RQ-2 What is the current state of the practice of effort estimation within the Scrum
teams most cases?
RQ-3 How could the effort estimation practices in Software Development be im-
proved?
To find out the state of the-art in software effort estimation, a literature review is
conducted.
2
The current state of effort estimation within Scrum teams of Software Developer
is researched using a survey with a questionnaire as a data gathering method. Ques-
tionnaire is created on the basis that have investigated effort estimation. To find the
relevant literature on which to base the questionnaire, a systematic literature review
is conducted to find software development effort estimation studies that have utilized
a questionnaire as a data gathering method. The survey conducted during this thesis
is limited to teams working.
The final research question is answered by combining the findings from the previ-
ous questions by analyzing the data collected from to find the weak points and then
looking for appropriate solutions in the findings from the literature review.
1.4 Research Contribution
This research presents the process of Effort & Cost Model of a Software Devel-
opment project. In this research first of all I studied the existing methods of Effort
Algorithms and readied a survey to collect data from Software Engineers. I create a
well-defined questionnaire for the purpose of data collection from those people who
were related with Software Development. I collected information via Maian Survey v
1.1 (WEB). After collect information from then i decided with my supervisor which
factor is considered to be taken for effort calculation. Questionnaires are attached in
Appendix section.
In factor attribute selection, i used voted system. Then, I used PHP Programming
Language for developing the project. I used several algorithm to output final actual
& predicted effort. I find out that MRE of each project is quite lower which concludes
it as a good effort model. Effort Matrix Development here and showed how to use it
in our project. After finally i tries to fit it in 3 several projects to get the MRE and
MMRE.
1.5 Socio-Economic Importance
•Help supervisor to manage resources, control budgets and agenda and develop
modern practices.
•Help Project Managers to complete on time and within financial plan.
•Make better relationship between Software clients and Project Managers &
Developers.
•Financial Report generate for Clients satisfaction.
3
1.6 Organization of Thesis
Rest of the thesis is organized as follows
Chapter 2: Literature Review This chapter provides the researches which are
related to Software Effort & Cost Model. Different approaches are described
with their research issues and drawbacks. This chapter also tries to find the
various existing and emerging cost estimation techniques and their respective
advantages and disadvantages.
Chapter 3: Background & Nature of Software Application This chapter dis-
cusses a background of Software Effort & Cost Models. It also provides different
application for the measurement of exertion required to build up the application
on time and inside spending plan with the Background & Nature of Software
Application.
Chapter 4: Review on Cost Estimation Techniques for Web Projects This
Chapter reviews the selection of the estimation techniques. This chapter intro-
duces the characteristics and strength of the Functional Measurement Type and
Expert-based estimation.
Chapter 5: Survey This chapter includes survey methodology studies the sampling
of questionnaires to select effort factors.
Chapter 6: System Process Diagram This Chapter provides visual synopses of
the individual use cases. It also provides graphical representations of work
processes of step insightful exercises and activities with backing for decision,
emphasis and simultaneousness.
Chapter 7: Research Methods This chapter includes the process or methods that
help us in predicting the actual and total effort & cost that will be needed for
our software.
Chapter 8: Effort Calculation Analysis This chapter is most important chapter
of this thesis. It shows Effort Calculation Analysis. This chapter suggest Effort
calculation Program based on the Effort Factor Tree Dependency. It includes
the entire factor this research method declared before.
Chapter 9: Implementation This Chapter goes with the real life implementation.
It provides requirements for setup, screenshots of the desired program.
Chapter 10: Results and Discussion This chapter describes the result and dis-
cusses the result which are obtained by executing the dataset. Results are
analyzed and discussed in terms of most appropriate rules.
Chapter 11: Conclusion & Future Work The concluding remarks about this re-
search are presented in this chapter.
4
CHAPTER II
Literature Review
2.1 Introduction to Effort Estimation
Programming improvement exertion estimation is the way toward anticipating
the most practical measure of exertion (communicated regarding individual hours or
cash) required to create or keep up programming in view of inadequate, dubious and
loud information. Exertion evaluations might be utilized as contribution to venture
arranges, emphasis arranges, spending plans, speculation examinations, valuing pro-
cedures and offering rounds.[1]
Distributed overviews on estimation hone recommend that master estimation is
the predominant procedure while assessing programming improvement exertion.
2.2 History
The greater part of the examination has concentrated on the development of for-
mal programming exertion estimation models. The early models were commonly
in view of relapse investigation or numerically got from speculations from different
spaces.
From that point forward a high number of model building approaches have been
assessed, for example, approaches established on case-based thinking, characteriza-
tion and relapse trees, reenactment, neural systems, Bayesian insights, lexical inves-
tigation of necessity determinations, hereditary programming, direct programming,
monetary creation models, delicate processing, fluffy rationale displaying, measurable
bootstrapping, and blends of two or a greater amount of these models.[1][4]
The maybe most normal estimation strategies today are the parametric estimation
models COCOMO, SEER-SEM and SLIM. They have their premise in estimation
research directed in the 1970s and 1980s and are from that point forward upgraded
with new alignment information, with the last significant discharge being COCOMO
II in the year 2000. The estimation approaches in light of usefulness based size
measures, e.g., capacity focuses, is additionally in view of exploration directed in the
5
1970s and 1980s, however are re-aligned with changed size measures and distinctive
tallying methodologies, for example, the ”utilization case focuses” in the 1990s and
COSMIC in the 2000s.[4]
2.3 Estimation Approaches
The confirmation on contrasts in estimation exactness of various estimation method-
ologies and models recommend that there is no ”best approach” and that the relative
precision of one approach or model in contrast with another depends unequivocally
on the setting . This infers diverse associations advantage from various estimation ap-
proaches. Discoveries, abridged in, that may bolster the determination of estimation
methodology in view of the normal exactness of a methodology incorporate.[1][3]
Combina�on
Based
Es�ma�on
Formal
Es�ma�on
Model
Expert
Es�ma�on
Figure 2.1: Effort Estimation Approaches
•Expert estimation: based on judgmental processes.
•Formal estimation model: taking into account mechanical procedures, the uti-
lization of an equation got from recorded information.
•Combination-based estimation: in light of a judgmental and mechanical blend
of appraisals from various sources.
2.4 Assessing the Accuracy of Estimates
Most regular measure of the normal estimation precision is the MMRE (Mean
Magnitude of Relative Error), where the MRE of every assessment is characterized
as Similar symmetric measures:
6
•Weighted Mean of Quartiles of relative errors (WMQ)
•Mean Variation from Estimate (MVFE)
2.5 Project Management Software
Venture administration programming has the ability to arrange, compose, and
oversee asset devices and create asset gauges. Contingent upon the modernity of the
product, it can oversee estimation and arranging, planning, cost control and spending
administration, asset assignment, coordinated effort programming, correspondence,
basic leadership, quality administration and documentation or organization frame-
works. Today, various PC and program based venture administration programming
and contract administration programming arrangements exist, and are discovering
applications in practically every sort of business.[1][5]
qdPM Feng Office
eyeOS Collabtive
dotProject ProjectPier
MantisBT TheBugGenie
TaskFreak todoyu
.. ..
phpCollab SiteDove
Table 2.1: List of Project Management Software
7
CHAPTER III
Background & Nature of Software Application
3.1 Introduction
Programming improvement exertion estimation is the way toward foreseeing the
most reasonable measure of exertion required to create or keep up programming tak-
ing into account fragmented, unverifiable and loud info. Exertion evaluations might
be utilized as contribution to venture arranges, emphasis.[7]
Software applications into three distinct classifications.
1. Hypermedia application: An application portrayed by the writing of data
utilizing hubs (lumps of data), connections (relations between hubs), stays,
access structures (for route), and conveyance over the Web.
2. Software application: An application that regularly actualizes a traditional
business area and utilizations the Web’s foundation for execution.
3. Web application: joins qualities Contrasts amongst Web and traditional pro-
gramming advancement, Technology and Architecture, Quality Drivers, Data
Structuring, Design, and Maintenance, Stakeholders, Legal, Social, and Ethical
Issues
3.2 Difference about Web and Software Projects
Differences between Web and conventional software development
1. Application Characteristics and Availability
•determine web application or Traditional software
•Web applications are distributed, are cross-platform, integrate numerous
distinct components, and contain content that is structured using naviga-
tional structures with hyperlinks.
•Traditional software applications are generally monolithic and single plat-
form, and can integrate distinct components.
8
2. Technology and Architecture
•Web application: Determine whether the application is built on Java solu-
tions, HTML, JavaScript, XML,UML, databases, third-party components
and middle ware, and so forth. In terms of their architecture, two-tier or
an n-tier
•Web application: Determine whether the application is built on Java solu-
tions, HTML, JavaScript, XML,UML, databases, third-party components
and middle ware, and so forth. In terms of their architecture, two-tier or
an n-tier
3. Quality Drivers
•Web application: quality product, reliability, usability, and security.
•Traditional Software: quality product
4. Information Structuring, Design, and Maintenance
•web application: structured/unstructured, hyperlinks
•Traditional software: structured and seldom employ hyperlinks.
5. Disciplines and People Involved in Development
•Web application: software/ hypermedia/ requirements / usability /infor-
mation engineering/ graphics design/ network management
•Traditional Software: programming, database design, and project manage-
ment
6. Stakeholders
7. Legal, Social, and Ethical Issues
3.3 Introduction to Web Effort Estimation
Web applications are disseminated, are crossstage, incorporate various particular
segments, and contain content that is organized utilizing navigational structures with
hyperlinks.[7]
Software Effort Estimation empowers organizations to know previously and before
actualizing an application the measure of exertion required to build up the applica-
tion on time and inside spending plan.[8]
Fundamental objective is to comprehend the venture variables that may influence
exertion forecast to appraise the web expense of an undertaking.
9
Estimation
Way
Description
Expert
Based
estimation
Idealistic evaluations lead to
thought little of exertion with the
immediate result of ventures be-
ing over spending plan.
Algorithmic
Based esti-
mation
Most Well known Fabricate mod-
els that definitely speak to
the relationship amongst exertion
through the utilization of algo-
rithmic models.
Artificial
Intelli-
gence
Techniques
Procedural estimation overseeing
comparative activities.
Table 3.1: Estimation Way
A few components were built up to comprehend the venture variables. Some of
are given below.
Drawbacks of expert-based estimation
•Repeatability
•experience alone is not enough to identify the underlying relationship between
effort and size-cost drivers
•Optimistic estimates lead to underestimated effort with the direct consequence
of projects being over budget and late.
Algorithmic based estimation: most popular techniques in the Web and software
effort estimation. It is used to build models that precisely represent the relationship
between effort and one or more project characteristics via the use of algorithmic
models.Example: COCOMO.[6]
3.4 Accuracy of an Effort Model
Measuring the predictive accuracy of an effort estimation model m or technique t
is a four-step process.[8]
Step 1 Split the original data set into two subsets: validation and training.
Step 2 Use the remaining projects (training subset) to build an effort estimation
model m. can be used explicit model (e.g., case-based reasoning).
10
Breakdown
into
Subsets
Fabricate
an Effort
Es�ma�on
Model
Apply
Model
Acquire
Evaluated
Exer�on
Processed
a general
Appraisal
Figure 3.1: Effort Estimation Accuracy Process
Step 3 Apply model m to each project pn to pq, and obtain estimated effort.
Step 4 Once estimated effort and accuracy statistics for pn to pq have been at-
tained,aggregated accuracy statistics can be computed, which provide an overall
assessment.
A few datasets are utilized to reproduce a circumstance where a Web organization
has a subset of new tasks
1. Measuring Effort Prediction Accuracy
•Magnitude of Relative Error (MRE)
•Mean Magnitude of Relative Error (MMRE)
2. Cross-Validation: The part of an information set into preparing and acceptance
sets is otherwise called cross approval
3.5 Sizing Web Applications
One of the survey showed some Web measures taxonomy
Second survey is about Web quality model (WQM). It is structured based on three
orthogonal dimensions.
•Web features
•Web life-cycle processes
•Web quality characteristics
11
This survey measures according to a second set of criteria
•Granularity level: Whether the measures scope is a Web page or Web site
•Theoretical validation: Whether or not a measure has been validated theoreti-
cally
•Empirical validation: Whether or not a measure has been empirically validated
•Automated support: Whether or not there is a support tool that facilitates the
automatic calculation of the measure
12
CHAPTER IV
Review on Cost Estimation Techniques for Web
Projects
This section, first, introduces Web Effort Estimation using different analysis and
reasoning and then briefly, discuss how to improve a company effort.[9]
4.1 Web Effort Estimation Using Regression Analysis
Data Source Industrial Web projects, from 133 online Web forms aimed at giving
quotes for Web development projects.
Technique Regression Analysis
For regression analysis, Suggestion is provided the following steps -
Data Validation first screening of the data that have been collected, understanding
what the variables are, descriptive statistics (e.g., the mean, median etc.).
Variables and model Selection After Data Collection, Preliminary analysis and
model building process is necessary.
•a. Preliminary Analysis: create variables based on existing variables, dis-
card unnecessary variables, and modify existing variables.
•b. Model building: construct an effort estimation model based on our data
set and variables.
Model Inspection verify, at each stage of the stepwise process, the stability of the
effort model.
Extraction of Effort Equation Effort equation should be screening with variables
and preliminary analysis and verification should be done for Extraction.
Model Validation uses a cross-validation mechanism to assess the prediction accu-
racy of an effort model.
Terms Terms related with Regression Analysis is given here.
13
Cross Validation Cross-validation represents splitting the original data set into
training and validation sets.
Training Set It is used to build an effort model.
Validation sets The projects in the validation set are used to obtain effort estimates
for each of the projects in this validation set, which once measured, are compared
to their corresponding actual effort.
4.2 Web Effort Estimation Using Case-Based Reasoning
Data Source Industrial Web projects
Technique Case-based Reasoning (CBR)
Terms Terms of Case based Reasoning is given below.
Case-based Reasoning (CBR) CBR provides effort estimates for new projects by
comparing the characteristics of the current project to be estimated against a
library of historical data from completed projects with known effort (case base).
The six parameters that can have a bearing on the estimations obtained using
CBR are as follows
1. Feature subset selection: It involves determining the optimum subset of features
that yields the most accurate estimation.
2. Brute-force algorithm: searches the solution domain for all possible feature
subset combinations looking for the one that provides the best results
3. Similarity measure: measures the level of likeness between different cases
4. Scaling: represents the transformation of feature values according to a defined
rule
5. Number of analogies: represents the number of most similar cases that will be
used to generate an effort estimate
6. Analogy adaptation
7. Adaptation rules
14
4.3 Web Effort Estimation using Classification Trees
Technique Classification and Regression Trees (CART)
Terms Here I provided the terms related this option.
CART Truck is a procedure where free variables (indicators) are utilized to construct
paired trees where every leaf hub either speaks to a class to which an assessment
has a place or a worth for an evaluation. The previous circumstance happens
with grouping trees and the last happens with relapse trees
Classification Tree In the event that indicators are straight out (e.g., Yes or No),
the tree is known as an arrangement tree.
Regression Tree On the off chance that indicators are numerical, the tree is known
as a relapse tree.
4.4 How to Improve Your Company Effort Estimation Prac-
tices
Data Source Software Web Application
Technique Expert based Estimation Approach
Terms Terms of Expert Estimation Approach is given below.
Expert-based Effort Estimation It represents the process by which effort for a
new project to be developed is estimated by subjective means, and is often
based on previous experience from developing or managing similar projects.
In general a software development process produces a software life cycle
•Requirements analysis
•Design
•Implementation (or coding)
•Testing
•Installation
•Maintenance
Models Models are listed below.
•The Waterfall Model
•Extension to the Waterfall Model
•Collaborative Web Development
•Agile Web Engineering Process
15
To improve simple project management practices, we should do [10]
1. Process model identifying phases
2. Gantt chart development. It is a time-charting technique where horizontal bars
are used to represent project activities, and the horizontal axis represents time
using dates, weeks, and so forth
3. Critical path analysis
4. Continuous feedback and continuous improvement
16
CHAPTER V
Survey
A field of applied statistics of human research surveys, survey methodology studies
the sampling of individual units from a population and the associated survey data
collection techniques, such as questionnaire construction and methods for improving
the number and accuracy of responses to surveys. Survey methodology includes
instruments or procedures that ask one or more questions that may, or may not, be
answered. In my Software Effort Estimation, I use survey methodology for sampling
individual units from survey data collection technique.
5.1 Introduction
Survey methodology as a scientific field seeks to identify principles about the
sample design, data collection instruments, statistical adjustment of data, and data
processing, and final data analysis that can create systematic and random survey
errors. Survey errors are sometimes analyzed in connection with survey cost. Cost
constraints are sometimes framed as improving quality within cost constraints, or
alternatively, reducing costs for a fixed level of quality. Survey methodology is both
a scientific field and a profession, meaning that some professionals in the field focus
on survey errors empirically and others design surveys to reduce them. For survey
designers, the task involves making a large set of decisions about thousands of indi-
vidual features of a survey in order to improve it. In this case, cost reduction with
quality maintained in the survey of selection of questionnaires in selecting the factor
for Software Effort Estimation.
The most important methodological challenges of a survey methodologists include
making decisions on how to
•Identify and select potential sample members.
•Contact sampled individuals and collect data from those who are hard to reach
(or reluctant to respond)
•Evaluate and test questions.
17
•Select the mode for posing questions and collecting responses.
•Train and supervise interviewers (if they are involved).
•Check data files for accuracy and internal consistency.
•Adjust survey estimates to correct for identified errors.
5.2 Questionnaires
Questionnaires are the most commonly used tool in survey research. However,
the results of a particular survey are worthless if the questionnaire is written in-
adequately.Questionnaires should produce valid and reliable demographic variable
measures and should yield valid and reliable individual disparities that self-report
scales generate.
Questionnaires are frequently used in quantitative marketing research and social
research. They are a valuable method of collecting a wide range of information from
a large number of individuals, often referred to as respondents, it can be students,
workers or any person whom you require information from.
Adequate questionnaire construction is critical to the success of a survey. Inap-
propriate questions, incorrect ordering of questions, incorrect scaling, or bad ques-
tionnaire format can make the survey valueless, as it may not accurately reflect the
views and opinions of the participants.
Different methods can be useful for checking a questionnaire and making sure it
is accurately capturing the intended information. These include:
•consulting experts
•estimating the measurement quality of the questions. This can be done for
instance using test-retest, quasi-simplex, or mutlitrait-multimethod models.
•predicting the measurement quality of the question. This can be done using the
software Survey Quality Predictor.
•pretest the questionnaire among a smaller subset of target respondents
5.2.1 Questionnaire construction issues
You have to know how (and whether) you will use the results of your research
before you start. If, for example, the results won’t influence your decision or you can’t
afford to implement the findings or the cost of the research outweighs its usefulness,
then save your time and money; don’t bother doing the research.
To develop Questionnaire, the thesis is followed by the terms.
18
Objectives The research objectives and frame of reference should be defined before-
hand, including the questionnaire’s context of time, budget, manpower, intru-
sion and privacy.
Population How (randomly or not) and from where (your sampling frame) you
select the respondents will determine whether you will be able to generalize
your findings to the larger population.
Nature The nature of the expected responses should be defined and retained for
interpretation of the responses, be it preferences (of products or services), facts,
beliefs, feelings, descriptions of past behavior, or standards of action.
Questions Pattern Unneeded questions are an expense to the researcher and an
unwelcome imposition on the respondents. All questions should contribute to
the objective(s) of the research.
Package If you ”research backwards” and determine what you want to say in the
report (i.e., Package A is more/less preferred by X% of the sample vs. Package
B, and y% compared to Package C) then even though you don’t know the exact
answers yet, you will be certain to ask all the questions you need and only the
ones you need in such a way (metrics) to write your report.
Topics The topics should fit the respondents frame of reference. Their background
may affect their interpretation of the questions. Respondents should have
enough information or expertise to answer the questions truthfully.
Scale Type The type of scale, index, or typology to be used shall be determined.
Measurment Level The level of measurement you use will determine what you can
do with and conclude from the data. If the response option is yes/no then you
will only know how many or what percent of your sample answered yes/no. You
cannot, however, conclude what the average respondent answered.
Type of Questions The types of questions (closed, multiple-choice, open) should
fit the statistical data analysis techniques available and your goals.
Outcome Questions and prepared responses to choose from should be neutral as to
intended outcome. A biased question or questionnaire encourages respondents
to answer one way rather than another. Even questions without bias may leave
respondents with expectations.
Order The order or ”natural” grouping of questions is often relevant. Prior previous
questions may bias later questions.
Wording The wording should be kept simple: no technical or specialized words.
Word Meaning The meaning should be clear. Ambiguous words, equivocal sen-
tence structures and negatives may cause misunderstanding, possibly invalidat-
ing questionnaire results. Double negatives should be reworded as positives.
19
Issue If a survey question actually contains more than one issue, the researcher will
not know which one the respondent is answering. Care should be taken to ask
one question at a time.
Category The list of possible responses should be collectively exhaustive. Respon-
dents should not find themselves with no category that fits their situation.
Mutually Exclusive The possible responses should also be mutually exclusive. Cat-
egories should not overlap. Respondents should not find themselves in more
than one category, for example in both the ”married” category and the ”single”
category there may be need for separate questions on marital status and living
situation.
Target Audience Writing style should be conversational, yet concise and accurate
and appropriate to the target audience.
Intimate Questons Many people will not answer personal or intimate questions.
For this reason, questions about age, income, marital status, etc. are generally
placed at the end of the survey. This way, even if the respondent refuses to an-
swer these ”personal” questions, he/she will have already answered the research
questions.
Presentation Presentation of the questions on the page (or computer screen) and
use of white space, colors, pictures, charts, or other graphics may affect respon-
dent’s interest or distract from the questions.
Numbering Numbering of questions may be helpful.
5.2.2 Types of questions
Questionnaires can be administered by research staff, by volunteers or self-administered
by the respondents. Clear, detailed instructions are needed in either case, matching
the needs of each audience.
5.2.2.1 Contingency question
A question that is answered only if the respondent gives a particular response to a
previous question. This avoids asking questions of people that do not apply to them
(for example, asking men if they have ever been pregnant).
5.2.2.2 Matrix questions
Identical response categories are assigned to multiple questions. The questions
are placed one under the other, forming a matrix with response categories along the
top and a list of questions down the side. This is an efficient use of page space and
respondents time.
20
5.2.2.3 Closed-ended questions
Respondents answers are limited to a fixed set of responses. Most scales are
closed-ended. Other types of closed-ended questions include:
Yes/no questions The respondent answers with a ”yes” or a ”no”.
Multiple choice The respondent has several option from which to choose.
Scaled questions Responses are graded on a continuum (example : rate the appear-
ance of the product on a scale from 1 to 10, with 10 being the most preferred
appearance). Examples of types of scales include the Likert scale, semantic
differential scale, and rank-order scale (See scale for a complete list of scaling
techniques.).
5.2.2.4 Open-ended questions
No options or predefined categories are suggested. The respondent supplies their
own answer without being constrained by a fixed set of possible responses. Examples
of types of open-ended questions include:
Completely unstructured For example, ”What is your opinion on questionnaires?”
Word association Words are presented and the respondent mentions the first word
that comes to mind.
Sentence completion Respondents complete an incomplete sentence. For example,
”The most important consideration in my decision to buy a new house is . . .”
Story completion Respondents complete an incomplete story.
Picture completion Respondents fill in an empty conversation balloon.
Thematic apperception test Respondents explain a picture or make up a story
about what they think is happening in the picture
5.2.3 Question sequence
Questions should flow logically from one to the next. The researcher must ensure
that the answer to a question is not influenced by previous questions. Questions
should flow from the more general to the more specific. Questions should flow from
the least sensitive to the most sensitive. Questions should flow from factual and
behavioral questions to attitudinal and opinion questions. Questions should flow
from unaided to aided questions. According to the three stage theory (also called the
sandwich theory), initial questions should be screening and rapport questions. Then
in the second stage you ask all the product specific questions. In the last stage you
ask demographic questions.
5.3 Final Effort Survey Questionnaires
21
Review Survey on Effort Estimation
Name
Designation
Email
Which Estimation Process is best for Software Estimation?
Expert Estimation
Algorithmic based estimation
Artificial Intelligence
In which way, will it be a support of implementation of estimation approach?
Weighted Micro Function Points
Use Case Analysis
Project Management Software [Activity Tracking]
Do you think, will it be helpful for effort estimation being used as Project Management Software as Expert Judgement?
Yes No
Do you think function point is relevant to Software Effort Estimation?
Yes
No
Do you think time duration development is neceessary as a factor of Software effort estimation?
Yes No
Do you think dependency of another task will be a issue for effort estimation?
Yes
No
Do you think Role of Sponsor can be a factor of Software Effort Estimation?
Yes
No
Do you think Activity type (Programming / Supporting) Software development can be factor of Software Estimation?
Yes
No
Do you think No of Assigned people of a task can be count as issue in Software Effort Estimation?
Yes
No
Do you think No of tasks can be count as issue in Software Effort Estimation?
Yes
No
Do you think Proxy Based Estimating can be count as a factor of Estimation Process?
Yes
No
Do you think task role(Project Manager, Developer) etc can be count as issue in Software Effort Estimation?
Yes
No
Do you think Job Position can be count as issue in Software Effort Estimation?
Yes
No
Do you think Status of a task can be issue for Software Estimation?
Yes
No
Do you think Changes of Company Policy affected Software Effort Estimation?
Yes
No
For spam prevention, enter the code you see on the left in the box on the right. Click captcha to refresh.
Powered by: Maian Survey v1.1 © 2008-2016 All Rights Reserved
5.4 Survey sampling
In statistics, survey sampling describes the process of selecting a sample of ele-
ments from a target population to conduct a survey. The term ”survey” may refer to
many different types or techniques of observation. In survey sampling it most often
involves a questionnaire used to measure the characteristics and/or attitudes of peo-
ple. Different ways of contacting members of a sample once they have been selected
is the subject of survey data collection. The purpose of sampling is to reduce the cost
and/or the amount of work that it would take to survey the entire target population.
A survey that measures the entire target population is called a census.
5.4.1 Probability sampling
In a probability sample (also called ”scientific” or ”random” sample) each member
of the target population has a known and non-zero probability of inclusion in the
sample. A survey based on a probability sample can in theory produce statistical
measurements of the target population that are:
•unbiased, the expected value of the sample mean is equal to the population
mean,
•have a measurable sampling error, which can be expressed as a confidence in-
terval, or margin of error.
5.4.2 Bias in probability sampling
Bias in surveys is undesirable, but often unavoidable. The major types of bias
that may occur in the sampling process are:
Non-response bias When individuals or households selected in the survey sample
cannot or will not complete the survey there is the potential for bias to result
from this non-response. Nonresponse bias occurs when the observed value de-
viates from the population parameter due to differences between respondents
and nonrespondents.
Response bias This is not the opposite of non-response bias, but instead relates to
a possible tendency of respondents to give inaccurate or untruthful answers for
various reasons.
Selection Bias Selection bias occurs when some units have a differing probability of
selection that is unaccounted for by the researcher. For example, some house-
holds have multiple phone numbers making them more likely to be selected in a
telephone survey than households with only one phone number. This selection
bias would be corrected by applying a survey weight equal to each household.
Self-selection bias A type of bias in which individuals voluntarily select themselves
into a group, thereby potentially biasing the response of that group.
23
Participation bias Bias that arises due to the characteristics of those who choose
to participate in a survey or poll.
Coverage bias Coverage bias can occur when population members do not appear
in the sample frame (undercoverage). Coverage bias occurs when the observed
value deviates from the population parameter due to differences between covered
and non-covered units. Telephone surveys suffer from a well known source of
coverage bias because they cannot include households without telephones.
5.4.3 Non-probability sampling
Many surveys are not based on probability samples, but rather on finding a suit-
able collection of respondents to complete the survey. Some common examples of
non-probability sampling are:
Judgement Samples A researcher decides which population members to include in
the sample based on his or her judgement. The researcher may provide some
alternative justification for the representativeness of the sample.
Snowball Samples Often used when a target population is rare. Members of the
target population recruit other members of the population for the survey.
Quota Samples The sample is designed to include a designated number of people
with certain specified characteristics. For example, 100 coffee drinkers. This
type of sampling is common in non-probability market research surveys.
Convenience Samples The sample is composed of whatever persons can be most
easily accessed to fill out the survey.
In non-probability samples the relationship between the target population and
the survey sample is immeasurable and potential bias is unknowable. Sophisticated
users of non-probability survey samples tend to view the survey as an experimental
condition, rather than a tool for population measurement, and examine the results
for internally consistent relationships.
5.5 Survey data collection
With the application of probability sampling in the 1930s, surveys became a stan-
dard tool for empirical research in social sciences, marketing, and official statistics.[1]
The methods involved in survey data collection are any of a number of ways in which
data can be collected for a statistical survey. These are methods that are used to col-
lect information from a sample of individuals in a systematic way. First there was the
change from traditional paper-and-pencil interviewing (PAPI) to computer-assisted
interviewing (CAI). Now, face-to-face surveys (CAPI), telephone surveys (CATI), and
mail surveys (CASI, CSAQ) are increasingly replaced by web surveys.
24
5.5.1 Modes of data collection
There are several ways of administering a survey. Within a survey, different meth-
ods can be used for different parts. For example, interviewer administration can be
used for general topics but self-administration for sensitive topics. The choice between
administration modes is influenced by several factors, including 1) costs, 2) coverage
of the target population, 3) flexibility of asking questions, 4) respondents willingness
to participate and 5) response accuracy. Different methods create mode effects that
change how respondents answer. The most common modes of administration are
listed under the following headings.
5.5.1.1 Mobile surveys
Mobile data collection or mobile surveys is an increasingly popular method of data
collection. Over 50% of surveys today are opened on mobile devices.[4] The survey,
form, app or collection tool is on a mobile device such as a smart phone or a tablet.
These devices offer innovative ways to gather data regardless of time and location of
the respondent.
5.5.1.2 Online surveys
Online (Internet) surveys are becoming an essential research tool for a variety of
research fields, including marketing, social and official statistics research. According
to ESOMARonline survey research accounted for 20% of global data-collection expen-
diture in 2006.[1] They offer capabilities beyond those available for any other type of
self-administered questionnaire.[8] Online consumer panels are also used extensively
for carrying out surveys but the quality is considered inferior because the panelists
are regular contributors and tend to be fatigued.
5.5.2 Advantages of online surveys
faster, simpler, and cheaper Web surveys are faster, simpler, and cheaper.[2] How-
ever, lower costs are not so straightforward in practice, as they are strongly
interconnected to errors. Because response rate comparisons to other survey
modes are usually not favourable for online surveys, efforts to achieve a higher
response rate (e.g., with traditional solicitation methods) may substantially in-
crease costs.[1]
data collection period The entire data collection period is significantly shortened,
as all data can be collected and processed in little more than a month.[2]
Interaction Interaction between the respondent and the questionnaire is more dy-
namic compared to e-mail or paper surveys.[8] Online surveys are also less in-
trusive, and they suffer less from social desirability effects.[2]
patterns Complex skip patterns can be implemented in ways that are mostly invis-
ible to the respondent
25
questions pattern Pop-up instructions can be provided for individual questions to
provide help with questions exactly where assistance is required.
Style Questions with long lists of answer choices can be used to provide immediate
coding of answers to certain questions that are usually asked in an open-ended
fashion in paper questionnaires.
situation Online surveys can be tailored to the situation (e.g., respondents may be
allowed save a partially completed form, the questionnaire may be preloaded
with already available information, etc.).
testing Online questionnaires may be improved by applying usability testing, where
usability is measured with reference to the speed with which a task can be
performed, the frequency of errors and user satisfaction with the interface.
5.6 Final Survey Results
Final Survey results is in the form of Bar Chart. The bar chart compares the
answers spending on different Questions prepared for taking the factor of Software
Effort Estimation.
It is clear that people in software engineering area spent significantly more effort
than people in the other work of area. Of the Fifteen Qouestion, Software Engineer
spent the most valuable time on this survey and finally allow to create the results in
percentage(%) format.
Best(Popular) Estimation Process in Software Effort & Cost Model is Expert Es-
timation just over 57% on the basis of 3 popular process(Artificial Intelligence(16%),
Algorithmic based estimation(26%) and Expert Estimation(57%)), which is the high-
est figure shown on the chart. In another questions, Project Management Software
[Activity Tracking] was the highest support of implementation of estimation approach
overall 3 approaches (Weighted Micro Function Points, Use Case Analysis, Project
Management Software [Activity Tracking]). The figures for spending time(Duration)
on software development were the yes mostly around(73%), at nearly 30 persons
among 41 peoples on this judgement.
However, while questioning on dependency of another task took place ”Yes” more
than ”No” around 68%,
In Contrast, Proxy Based Estimating paid out less Yes than No. The amount of
Yes is less than 32% by the survey user, around 13, is the lowest figure shown on the
chart.
26
41 people have participated in this survey.
Survey Results
Review Survey on Effort Estimation
Name
View All Responses to this Question
Designation
View All Responses to this Question
Email
View All Responses to this Question
Which Estimation Process is best for Software Estimation?
Artificial Intelligence 16.7% (7)
Algorithmic based estimation 26.2% (11)
Expert Estimation 57.1% (24)
In which way, will it be a support of implementation of estimation approach?
Weighted Micro Function Points 19.5% (8)
Use Case Analysis 29.3% (12)
Project Management Software [Activity
Tracking]
51.2% (21)
Do you think, will it be helpful for effort estimation being used as Project Management Software as Expert Judgement?
No 26.8% (11)
Yes 73.2% (30)
Do you think function point is relevant to Software Effort Estimation?
Yes 39.0% (16)
No 61.0% (25)
Do you think time duration development is neceessary as a factor of Software effort estimation?
Yes 73.2% (30)
No 26.8% (11)
Do you think dependency of another task will be a issue for effort estimation?
Yes 68.3% (28)
No 31.7% (13)
Do you think Role of Sponsor can be a factor of Software Effort Estimation?
Yes 26.8% (11)
No 73.2% (30)
Do you think Activity type (Programming / Supporting) Software development can be factor of Software Estimation?
Yes 63.4% (26)
No 36.6% (15)
Do you think No of Assigned people of a task can be count as issue in Software Effort Estimation?
Yes 70.7% (29)
No 29.3% (12)
Do you think No of tasks can be count as issue in Software Effort Estimation?
Yes 63.4% (26)
No 36.6% (15)
Do you think Proxy Based Estimating can be count as a factor of Estimation Process?
Yes 31.7% (13)
No 68.3% (28)
Do you think task role(Project Manager, Developer) etc can be count as issue in Software Effort Estimation?
Yes 78.0% (32)
No 22.0% (9)
Do you think Job Position can be count as issue in Software Effort Estimation?
No 34.1% (14)
Yes 65.9% (27)
Do you think Status of a task can be issue for Software Estimation?
Yes 63.4% (26)
No 36.6% (15)
Do you think Changes of Company Policy affected Software Effort Estimation?
Yes 43.9% (18)
No 56.1% (23)
Powered by: Maian Survey v1.1 © 2008-2016 All Rights Reserved
CHAPTER VI
System Process Diagram
6.1 System Sequence Diagram (SSD)
In Software Engineering, a system sequence diagram (SSD) is an arrangement
outline that appears, for a specific situation of an utilization case, the occasions that
outer performing artists create, their request, and conceivable between framework
occasions.
system sequence diagram (SSD) are visual synopses of the individual use cases.
6.2 Activity Diagram
Activity Diagrams are graphical representations of work processes of step insight-
ful exercises and activities with backing for decision, emphasis and simultaneousness.
In the Unified Modeling Language, Activity Diagrams are expected to demonstrate
both computational and authoritative procedures (i.e. work processes). Action charts
demonstrate the general stream of control.
28
Figure 6.1: Sequence Diagram
6.3 Use Case Diagram
An use case diagram at its least complex is a representation of a users’s associa-
tion with the framework that demonstrates the relationship between the client and
the diverse use cases in which the client is included. An utilization case graph can
distinguish the diverse sorts of clients of a framework and the distinctive use cases
and will frequently be joined by different sorts of charts also.
6.4 Flowchart
29
Figure 6.2: Activity Diagram
30
Figure 6.3: Use Case Diagram
31
Figure 6.4: Flowchart
32
CHAPTER VII
Research Methods
7.1 Introduction
The industry of Software ought to be proficient. Because of quick change in inno-
vation, execution of complex programming frameworks at less expensive expense and
the inclination to keep up better quality programming are a portion of the significant
difficulties for the product organizations.
Cost estimation includes the process or methods that help us in predicting the
actual and total cost that will be needed for our software and is considered as one
of the complex and challenging activity for the software companies. Their goal is to
develop software which is cheap and at the same time deliver good quality. Software
cost estimation is used basically by system analysts to get an approximation of the
essential resources needed by a particular software project and their schedules.
In this arena, some methods are run sequentially. Some of them are COCOMO,
Expert Judgment, Function Point, Analogy etc. Some Modules of this System: Line
of Code,Fast Prediction. Some of the methods are required much data for calculation.
Different variables that influence the expense are software engineer capacity, experi-
ence of the designer’s zone, unpredictability of the venture and dependability prereq-
uisites and so forth the data sources and yields of the product cost estimation process.
The essential cost driver is thought to be the product necessities. It is the essential
contribution to the estimation procedure.
Fetched estimation strategies are essentially of two sorts: algorithmic and non-
algorithmic. Algorithmic strategies utilize an equation to ascertain the product cost
gauge. The equation is produced from models which are made by joining related
cost variables. Likewise, the measurable technique is utilized for model development.
Non-algorithmic strategies don’t utilize a recipe to figure the product cost gauge. The
principle point of this paper is to give an audit of these current models and strategies.
33
7.1.1 Non Algorithmic Based Estimation Methods
7.1.1.1 Expert Judgment Method
Master judgment methods include counseling with programming cost estimation
master or a gathering of the specialists to utilize their experience and comprehension
of the proposed venture to touch base at an appraisal of its expense. It is the most
usable strategies for the product cost estimation. Generally organizations utilized
this technique for creating the expense of the item.
7.1.1.2 Estimating by Analogy
Evaluating by relationship implies contrasting the proposed venture with already
finished comparable undertaking where the task advancement data id known. Real
information from the finished undertakings are extrapolated to assess the proposed
venture. This technique can be utilized either at framework level or at the segment
level.
7.1.2 Algorithmic Method
The algorithmic technique is intended to give some numerical conditions to per-
form programming estimation. These numerical conditions depend on exploration
and chronicled information and use data sources, for example, Source Lines of Code
(SLOC), number of capacities to perform, and other cost drivers, for example, dialect,
outline procedure, aptitude levels, hazard appraisals, and so forth.
Comparison above all systems with PROCOST according to free addition.Such as:
Topics PROCOST COCOMO Function Point Top-down Analogy
Type (Algorith-
mic)
Y Y Y N N
Adapt to espe-
cial projects
Y N Y N N
Much data re-
quirement
Y N Y N N
Old Data Re-
quirement
Y Y N N N
Reinforces poor
practice
Y N N N Y
Generally pro-
duces large
overruns
Y N N N N
Table 7.1: Comparison with Existing System
34
Taken a toll estimation if done before the start of a task can help in deciding the
elements which can be incorporated inside the constrained assets of the undertaking.
It likewise helps in lessening dangers.
7.2 Proposed System
Identify these problems, this proposed system target is to overcome those prob-
lems. Primarily, The Proposal entitled as PROCOST can do the task of issue tracking
application. PROCOST is a free Web team collaboration tool that will provide bug
tracking, issue tracking, and project management functions. PROCOST covers all
aspects of process of predicting the most realistic amount of effort (expressed in terms
of person-hours or money) required to develop or maintain software based on incom-
plete, uncertain and noisy input from setting up some effort factor to operational
tasks such as creating issues, reviewing, editing, publishing, archiving, and indexing
of a project or repositories. PROCOST also helps to manage the people involved in
a project. It also helps employees and Project Leader.
Below, Overview of the process PROCOST uses is given in Table 7.2.
Topics Super
Admin
Project
Leader
Employee Role
As-
signer
System
User
Install and Config Y N N N N
Create Project Y Y N N N
Open and Close Sta-
tus
Y Y N N N
Add Employee Y Y Y N N
Create New Task Y Y Y Y N
Assign Role Y Y N Y N
Task Status changed Y Y Y N N
Review Task Y Y N Y N
Effort Result Report Y Y Y Y Y
Individual Task List Y Y Y Y Y
Table 7.2: Proposed System Process
PROCOST is composed in PHP and utilizations the function, Apache, and innova-
tion stack. For remote method calls (RPC), PROCOST underpins JQUERY, AJAX,
SOAP and XML-RPC. PROCOST incorporates with source control projects, for ex-
ample, Clearcase, Concurrent Versions System (CVS), Git, Mercurial, Perforce, Sub-
version, and Team Foundation Server.
35
The principle elements of PROCOST for deft programming improvement are the
usefulness to arrange advancement emphasess, the emphasis reports and the bug fol-
lowing usefulness.
7.3 System overview
PROCOST is an open source Software as a project management tool for agile
teams to manage and publish online resources and estimate effort for projects. PRO-
COST shields all aspects of available tsks management and publishing, from setting
up project to operational tasks such as submitting, reviewing, editing, publishing,
archiving, and indexing of the task issues. PROCOST also helps to manage the
project people involved in organizing a project. Agile teams can stay focused on de-
livering iterative and incremental value, as fast as possible, with customizable scrum
boards.
Additionally, PROCOST is flexible and scalable. A single installation of PRO-
COST can support the operation of multiple projetcs, and multiple years for each
projects. Each project has its own unique URL as well as its own look and feel.
PROCOST can enable a single director to manage all aspects of a project. At last,
it is an open source project and listed in GITHUB.
In other sense, Teams have access to more than a dozen out-of-the-box reports
with real-time, actionable insights into how their teams are performing sprint over
sprint in PROCOST to deliver Software Effort and Cost. Its core fuynctionality is to
get visibility across all teams and projects with Portfolio. It submerge the tasks to
forecast realistic roadmaps, manage team resources and track progress with real-time
planning.
7.3.1 Numerous deployment options
In the cloud, on your own infrastructure, or at massive scale, PROCOST Software
has its platform covered.
7.4 Methodology
1. The research approach try to introduce dataset includes basic requirements of
projects with some Functional Measurement types and complexity Factor of
the software development effort. So, using dataset for evaluating the proposed
model is based on Algorithmic model.
2. The second attempt will to create an all requirement dataset based on one of
requirement, based on model and algorithm.
36
Feature Comparison Cloud Server Data Center
Project and issue tracking Y Y Y
Scrum and kanban support Y Y Y
Backlog prioritization and
sprint planning
Y Y Y
Flexible workflow Y Y Y
Developer tool integrations Y Y Y
Out-of-the-box agile report-
ing
Y Y Y
Rich APIs Y Y Y
Plug-and-play add-ons Y Y Y
Active-active clustering N N Y
Disaster recovery N N Y
Table 7.3: Feature Comparison
•Estimate criteria for each requirement in dataset, then asserts it into sev-
eral equal intervals (lengths).
•Cost Factor matrix development
•Estimate a corresponding extra linguistic variable for each interval of re-
quirement of Functional Measurement Type.
•Estimate Project management software primary functions
37
CHAPTER VIII
Effort Calculation Analysis
This section, first, introduces the characteristics and strength of the Functional
Measurement Type and Expert-based estimation, briefly, then the new Project Man-
agement Process is explained.[12]
8.1 Effort Calculation Method Development
Here, This Research Method Program suggest Effort calculation Program based
on the Effort Factor Tree Dependency. It includes the entire factor this research
method declared before. Explanation is given after the Program.[14]
8.1.1 Work flow
System data acquisition, data organization and at last data analysis is shown in
this work flow diagram. The diagram is visualized in Flowchart ??
Here, a server as Apache Server and Windows/Linux Distributed File System will
worked. It acquires data from System that transmit Effort Factor information in DB
Format. System Server further processes this information. These processing raw data
will produce valuable resources for Project Leader to assess project effort.[18]
These information contains Project current level and other information that helps
user to take proper decision about his/her Project. [15][3]
If any vehicle changes its way of working then system will acquire its data and
process it the server. This continuous process will produce huge data for engaging
with project effort Data.[17]
At last, every data sync time, The model process the data to make information
about current project. If Values are changes by any user continuously then the server
take an automated decision, else if it continues its continuous process.
38
8.1.1.1 Algorithm
Here, This Research Method suggests Effort calculation Program based on the
Effort Factor Tree Dependency. It includes the entire factor this research method
declared before. The Algorithm of the proposed approach is referred to Algorithm
1.[16][27]
Algorithm 1: Effort Calculation Algorithm
Data: collected from the Activity Decision
Result: Effort calculation for directional task
initialization;
while Number of Project Task do
read current;
while Number of Assigning People do
read current;
Read Job Position of Assigned Current People;
Read Project Basis Role value of Assigned Current People;
Read Assigned Role value of Assigned Current People;
Calculate the multiply value of these values;
end
if Duration of Task is not extended then
Read Current Duration (Time);
else
Read LOG value of that Current Duration(Time);
end
if Check Dependency of current task on another task then
Read Current Dependency of Another Task value section becomes this
one;
else
Read the UNITY function of Current Dependency of Another Task
Value;
end
end
It will be designed from these equations in the following development.
8.1.1.2 Program Development
Here, the proposed approach algorithm is transferred to pseudo code. The pseudo
code of program is given in Program.
8.1.2 Data Schema
To estimate Effort, different data source (users information) are relied on. Ac-
quisition of data should be manipulated by a Data schema developed in the Apache
39
server. From that data, users are provided various types of information like Their
Job Information, Task Information etc.[20]
Here, The proposed model visualized a data schema installed on the Apache server.
This data schema will be helpful to the developer to collect several types of informa-
tion which is related to the Company information.
To be built a data schema, the main table Project Task Table, contains data
such as Employee information, Project Information information, Task information
etc. The abbreviations FK and PK are Foreign key and Primary Key respectively.
By querying the main table, developer can summarize the area information which is
under of Apache Server.[8][14]
8.1.3 Why Linear Properties
Linear Properties arise quite naturally when modeling many phenomena; these
properties are particularly useful by assuming that quantities of interest vary to
only a small extent from some background state. Linear equations do not include
exponents.[4]
This Informations Provides, Basic factors of Effort is not need any jacular prop-
erties to get very high result. Its have some linear equation properties like how much
you get how much you do. For These Reasons, in effort calculation I use Linear
Properties.[26]
f un ct io n e f f o r t c a l c u l a t i o n ( ) {
for( $ i = 1; $ i <= $NPT ; $ i ++){
for( $ j = 1; $ j <= $NAP; $ j++ ) {
$ a s s i g n e d a l u e = $J P [ j ] ∗$PBR [ j ] ∗$AR[ j ] ;
}
while( 1) {
i f ( $DT != e x tend e d )
$DTValue = $DT;
e ls e i f ($DT==exte n de d )
$DT Valu e = LOG( $DT ) ;
i f ( $DAT!= NO)
$DATValue = 1 ;
e ls e i f ( $DT== e x te n de d )
$DATValue = E f f o r t ( $DATValue ) ;
}
$EFFORT[ i ]= $ST∗[ $DTvalue∗$DTT]+ $AV+ log($DAT) ;
}
40
8.1.4 Why Logarithm Properties used in the Program
Logarithms, the indices of the ratios of Numbers to one another, being a series if
numbers in arithmetical progression, corresponding to others in geometrical progres-
sion. Bu means by which, arithmetical calculations can be made with much easier
and expedition than other calculation.
For these reason, in my thesis i use logarithm to get better result when the ser-
vice is extended. When a service is extended, it doesnt follow a Linear Properties.
Logarithm behavior is much more likely with extension of Effort services to calculate
Effort. [21]
8.2 Effort Factor
Survey Report helped to determine the factor of Effort Calculation. Now, I use
the factor for Effort Calculation.
1. Number of Project Task (NPT)
2. Duration of the task (DT)
3. Status of the Task (ST)
4. No of Assigning People (NAP)
5. Job Position (JP)
6. Assigned Role of a task (AR)
7. Duration Type of the task (DTT)
8. Dependency of another Task (DAT)
9. Project basic Role (PBR)
8.2.1 Effort Factor Tree Dependency
Effort Factor Tree Dependency allow us to detect relationship among factors. This
hierarchy shows the inter-dependency and how we calculate the Software Effort for a
project. Figure 8.1 shows the real view as a relationship among the factors.
8.3 Details of Effort Factor
Here, This Research Method take some values of Effort factors. These are prede-
fined by the company.
41
Number of
Project Task
(NPT)
Dura�on of
the Task (DT)
Status of the
Task (ST)
No of
Assigning
People (NAP)
Job Posi�on
(JP)
Project Bass
Role (PBR)
Assigned
Role (AR)
Dura�on
Type of the
Task (DTT)
Dependency
of Another
Task (DAT)
Figure 8.1: Effort Factor Tree Dependency
Status of the Task (ST) The factor status of a task is varied on time to time. This
is time varying factor of Software Effort calculation. It will help supervisor to
check time to time working activity. However, this factor is chosen based on
the user survey resulted positively by 63% people. Here, a sample of Project
status is given in Table 8.1.
Serial No. Status Type ST Value
1 Planning 6
2 Open 5
3 In Progress 4
4 Awaiting Confirmation 3
5 Done 2
6 Accepted 1
7 Cleared 0
Note: These values are predefined by the company.
Table 8.1: Status of the Task (ST)
Job Types / Job Position (JP) To calculate Effort, Job Types / Job Position is
crucial factor. In thesis survey, 65% survey user gave their vote to ”YES” to
this factor. This is a sample Table 8.2 (Example) for this factor.
Project Basis Roles (PBR) In PROCOST Survey, the factor - Project Basis Roles
(PBR) is selected based on the result. 78% people shows their positive attitude
42
Serial No. Status Type ST Value
1 CEO 6
2 CTO 5
3 Manager 4
4 Team leader, 3
5 Secretary, 2
6 Sales 1
Note: These values are predefined by the company.
Table 8.2: Job Types / Job Position (JP)
to this factor. It is confirmed that more 3 of 4 people wants this factor for
calculating the Software Effort. Here, the real sample of a Project Basis Roles
of a project is given in Table 8.3.
Serial No. Roles JP Value
1 Project manager 6
2 Developer 5
3 Frontend engineer 4
4 Product manager 3
5 Marketing responsible 2
6 Trainer 1
Note: These values are predefined by the company.
Table 8.3: Project Basis Roles (PBR)
Assigned Role of a task (AR) When a project is started, the tasks are scheduled
and given to their employer. As they have a project basis roles, they also have
assigned task roles (individual). The activity Type got positive answers from
63% people of the survey. The assigned roles of a sample project is given in
Table 8.4.
Serial No. Roles AR Value
1 Consulting 4
2 Programming 3
3 Support 2
4 Internal work 1
Note: These values are predefined by the company.
Table 8.4: Assigned Role of a task (AR)
43
Duration Type of the task (DTT) This factor is undoubtedly an important fac-
tor for effort calculation. 73% survey user say to ”YES” to give it to as a factor
of effort calculation. The early model used this type of Duration type. A sample
Table 8.5 is given basis on other estimation models.
Serial No. Type DTT Value
1 Hourly 1
2 Daily 8
3 Monthly 8*22
Note: These values are predefined by the company.
Table 8.5: Duration Type of the task (DTT)
Dependency of another Task (DAT) When this thesis tries to make good ques-
tionnaires, we do not think this questions answer will be positive. However,
68% users pay their attention to this questions and made this as an important
factor of effort calculation. Sample example is given to rewrite the concept in
Table 8.6.
Serial No. Type DAT Value
1 No 1
2 Yes(Effort calculate of that task) Value of dependant task 8
Note: These values are predefined by the company.
Table 8.6: Dependency of another Task (DAT)
44
CHAPTER IX
Implementation
9.1 Requirements
The following requirements are applicable if you choose to run PROCOST on your
own server.
•Apache or other* HTTP server
•PHP version 5.3.7 or newer, –with-mysqli –with-mcrypt.
•MySQL or MariaDB version 5 or newer
•15 MB of disk space for the PROCOST software, plus additional space for
submission files
•1 MB of database space per 100 tasks (actual amount will vary based on modules
installed, number of hostory of tasks, and other factors)
•The hosting account (web server) will also require.
•Create write access to procost/connect.php and procost/*
•MySQL privileges to: ALTER, CREATE, DELETE, DROP, INSERT, SE-
LECT, TRUNCATE, UPDATE
9.2 System User Section
9.2.1 Login Portal for System User
In this section System User can login to system for Tracking his/her work. He/she
use one account using their email and password credentials. After login, he/she will
see various option to manage his/her work and help project managers to calculate
effort. The figure 9.1 is described it most.
45
Figure 9.1: Login Portal for System User
9.2.2 Repositories View by Project Manager
After login to the system, a project manage can track the all projects/repossitories.
There will be exist progress bar with color to easily identify the task status. If project
manager thinks to email them to work quick or slow, he/she can do it from the home
page via email. If a task is need to rescheduled then project manager can assign it
to new person. It is very helpful for other team members to collaborate to a project.
The figure 9.2 is described it most.
Figure 9.2: Repositories View by Project Manager
46
9.2.3 Task List of Individual User
If Team member of a project can see the task list, by which he/she can understand
what to do and the time line also. He/she can change the status of project which is
given to him/her. Here, it is visualized in Figure 9.3
Figure 9.3: Task List of Individual User
9.3 System Employee Portal
9.3.1 Employees List in the System
In System, Admin can see the all employee list. The list will show the user name,
email, Job position. In addition it also inform active status and is he/she admin or
not. The figure 9.4 is described it most.
Figure 9.4: System Employees List
47
9.3.2 Add Employee for the system
In PROCOST System, Admin Can add new employee for the track working info.
It helps the project manager to estmate the effort and cost. Here, it is visualized in
Figure 9.5
Figure 9.5: Add Employee UI for the system
9.4 Project-Task Related Section
9.4.1 Project Details
It is easy to use for admin to configure a project. Project Team members can
create unlimited task, dynamically mention the task history, change date etc. All
information for a project will be shown in the the list here. Here some option exist
that helps to understand and manage that projects. The figure 9.6 is described it
most.
48
Figure 9.6: Project Details
9.4.2 Add a New Project
When Admin add new project in this system then he/she should be add task list
under that project. This Information helps project manager to estimate the whole
process of the project. Here, it is visualized in Figure 9.7
Figure 9.7: Add a New Project
49
9.4.3 Task Details
Admin can see the task list to manage that project. Several task can be listed
under a project within a time line. This is the most valuable part to calculate the
task history effort. The figure 9.8 is described it most.
Figure 9.8: Task Details
9.4.4 Add Issue
Admin can add issue under a task to manage that task. Several issues can be
listed under a task within a time line. All effort calculation is based on the issues
information. Here, it is visualized in Figure 9.9
50
Figure 9.9: Add Issue
9.5 Other Section
9.5.1 About UI
Admin can see the project about from the top menu. The figure 9.10 is described
it most.
Figure 9.10: About
51
9.5.2 FAQ
Admin can see the project FAQ from the top menu. Here, it is visualized in Figure
9.11
Figure 9.11: FAQ on PROCOST
52
CHAPTER X
Results and Discussion
10.1 Evaluation Criteria
According to [30] there are various approaches for evaluating the estimation ac-
curacy of software effort proposed model such as:-
MRE (Magnitude of relative error) First calculate the degree of estimating er-
ror in an individual estimate for each data point as project .It is defined as:-
MRE =|predictedvalue −actualvalue|
actualvalue (10.1)
RMSE (Root Mean Square Error) It is frequently used measure of differences
between values predicted by a model or estimator and the values actually ob-
served from the thing being modelled or estimated. It is just the square root of
the mean square error as shown in equation given below:-
RMSE =r1
N
n
X
I=1
(actualvalue −predictivevalue)2(10.2)
MMRE (Mean Magnitude of Relative Error) It is another measure and is the
percentage of the absolute values of the relative errors, averaged over the N
items in the ”Test” set and can be written as:-
MM RE =1
N
n
X
I=1
|predictedvalue −actualvalue|
actualvalue (10.3)
PRED (N) is the third criteria used for the comparison and this reports the average
percentage of estimates that were within N% of the actual values .It is commonly
used and is the percentage of predictions that fall within p % of the actual,
denoted as PRED (p), k is the number of projects where MRE is less than or
equal to p, and n is the number of projects.
P RED(p) = k
n(10.4)
53
10.2 Results
The dataset used in this thesis based on empirical validation came from ISBSG.[29]
The obtained dataset contains effort records for six phases are: Status of the Task
(ST), Job Types: [Job Position (JP)], Project Basis Roles (PBR), Assigned Role of
a task (AR), Duration Type of the task (DTT) and Dependency of another Task
(DAT). As a preliminary stage of data preprocessing we attempted to select the most
representative data, therefore we ignored the projects records that contain missing
values.
10.2.1 Effort Calculation
Simulations show that after analysing the existing and new cases for the projects,
performance evaluation is simplified. Thus, we calculate the results on the basis of
MRE and MMRE, this helps in comparing the results for each developed model. It
is clearly evident from Table II that predicted values of effort are very close to the
expected or actual values. Calculation of MRE and MMRE values over the complete
data set of models is portrayed in Table IV.
The Appraisal of proposed approach was performed in 3 web applications used
in Result analysis. Web application Effort & Cost Estimation was conducted us-
ing Schneiders Model, Karners model and proposed approach. Most Specialists use
MMRE to calculate the error Percentage of Software Effort & Cost Estimation.
MMRE is the mean of the Magnitude of Relative Error. It is very communal princi-
ple used to evaluate software cost estimation models[1][4][5]. Magnitude of Relative
Error (MRE) for each surveillance can be obtained as:
MREi=|AEi−P Ei|
AEi
(10.5)
Where, AE means Actual Effort,PE means Predicted Effort. MMRE can be ac-
complished be an average of the summation of MRE over N interpretations[18][22][9].
MM RE =1
N
n
X
I=1
MREi(10.6)
The total results is in the output in Table 10.1. The view of total result is summa-
rize here. To make a easy understandability, The thesis tries to cover all calculation in
one table. As our above rules, this thesis implements those rules into the calculation
to make the output as MRE and MMRE.
Just use Equation 10.5 and 10.6 and use the actual effort and estimate effort by
thee software that was built from the Thesis Algorithm. Moreover, This tasks are
taken from a dataset of a project with their timetable and tasks list. Difference is we
use the other factors value from their respected website.
For Example -
54
Openconf Peer-Review, Abstract and Conference Management Software - regis-
tered trademark of Zakon Group LLC. See more details in Openconf Website -
https://www.openconf.com/
Opencart open source online e-commerce solution - registered trademark of Open-
cart. See more details in Opencart Website - https://www.opencart.com/
Proconf open source Peer-Review, Abstract and Conference Management Software
- registered trademark of PROCONF. See more details in PROCONF Website
- https://www.proconf.org/
Tasks
Openconf Opencart proconf
Effort (In PM) E - Estimated Effort & A- Actual Effort
E A MRE E A MRE E A MRE
Project Start 5 7 0.2857 66 72 0.08333 24 32 0.25
Technology Preview 87 102 0.1470 125 172 0.2732 54 57 0.0526
Beta Version 109 100 0.09 187 205 0.0878 67 82 0.1829
Specification 67 87 0.2298 88 97 0.09278 24 20 0.2
Manual 75 90 0.1666 100 111 0.0990 20 18 0.111
Graphical User Interface 166 155 0.0709 208 215 0.0325 134 154 0.1298
Software Development 400 488 0.1803 607 809 0.2496 254 287 0.1149
Database coupling 256 281 0.0889 378 511 0.2602 176 199 0.1155
Back-End Functions 567 700 0.19 801 856 0.0642 302 287 0.0522
Software testing 157 163 0.0368 132 124 0.0645 56 56 0
Alpha Test 48 52 0.0769 42 42 0 12 23 0.4782
Beta Test 102 90 0.1333 80 77 0.0389 88 112 0.2142
Ship Product to Customer 11 12 0.0833 7 6 0.1666 4 4 0
Customization 5 10 0.5 102 126 0.1904 12 14 0.1428
others 89 108 0.1759 151 178 0.1516 34 87 0.6091
Pn
I=1 MREi2.4559 1.8553 2.6539
MMRE 16.37% 12.36% 17.69%
Table 10.1: Difference in Actual & Predicted Effort
A high estimation blunder can’t consequently be translated as a marker of low
estimation capacity. Elective, contending or supplementing, reasons incorporate min-
imal effort control of venture, high many-sided quality of advancement work, and
more conveyed usefulness than initially evaluated. A structure for enhanced utilize
and elucidation of estimation mistake estimation is incorporated into.
55
10.2.2 Cost Calculation
The vast majority of the work in the cost estimation field has focued on algorith-
mic cost displaying. In this procedure expenses are broke down utilizing numerical
recipes connecting expenses or contributions with measurements to create an expected
yield. The formulae utilized as a part of a formal model emerge from the examination
of authentic information. The exactness of the model can be enhanced by aligning
the model to your particular advancement environment, which essentially includes
altering the weightings of the measurements.
This application derives the PROCOST software engineering metric as found as
opensource application. The specific version utilized here is the ”basic” model.
YOUR BASIC PROCOST RESULTS!!
Projects Openconf Opencart proconf
Actual Effort 2445 3601 1432
Software Labor Rates($) Cost per Person-Month (Dollars)
Labor Rates in Bangladesh 320
Actual Cost (One time) 782400 1152320 458240
Table 10.2: Cost Calculator
Noted that, this cost is based on Software Labor Rates and one time project de-
velopment. In case of Project maintainable work, the effort factor s as usual but some
more maintaining factor will be added each month. Project which need maintenance
can use this model by adding more tasks and their evaluation by System Adminis-
trator and check those factor for ever month.
The labor rates is a dummy data from Bangladesh Government Site. For further
information, see https://sites.google.com/site/sayedmohsinreza/education/masters-
thesis
10.2.3 Final Results
The calculated errors using different models are shown in Table 10.1 and cost
calculator in Table 10.2. In this section, we are using statistical methods like MMRE
and Prediction for evaluating the cost estimating models. From this results, we
can conclude that no doubt this process is try to predict their values and show the
differences in some practical approaches. It will allow other new users to help to
understand the algorithm for optimization to relate the actual and predicted effort.
56
CHAPTER XI
Conclusion & Future Work
11.1 Conclusion
A fundamental issue for undertaking directors is the exact and dependable ap-
praisals of the required programming improvement Effort, particularly in the early
phases of the Software Product Development Life Cycle. Programming exertion
drivers as a rule have properties of vulnerability and dubiousness when they are
measured by human judgment.
A Software Product Effort Estimation Model using Functional Measurement Type
in Project Mangement systems can conquer these qualities of vulnerability and un-
clearness exist in programming effort drivers. Be that as it may, the determination
of the reasonable Project Effort Estimation systems model assumes an essential part
in concocting exact and solid exertion gauges.
Software Effort estimation utilizing such systems is an endeavor in the territory
of programming undertaking estimation. The target of this work is to give a strategy
to programming cost estimation that performs superior to anything different systems
on a given arrangement of experiments. This paper introduced another model for
taking care of imprecision and instability by utilizing the Expert Based Estimation
systems.
The target of this work is to give a strategy to programming effort estimation
that performs superior to anything different methods on the precision of Effort es-
timation. This work has appeared by applying Functional Measurement systems on
the algorithmic and non-algorithmic programming Effort estimation models precise
estimation is achievable. The proposed systems model indicated better programming
exertion gauges in perspective of the MMRE assessment criteria when contrasted with
other system.
The aforementioned results exhibit that applying Functional Measurement sys-
tems technique to the Software Effort estimation is an achievable way to deal with
tending to the issue of instability and dubiousness existed in programming exer-
57
tion drivers. Besides, the systems model exhibits better estimation exactness when
contrasted with such Effort estimation Model. The use of Functional Measurement
systems for different applications in the product designing field can likewise be inves-
tigated later on.
11.2 Future Work
Some of Software projects are fizzled because of the nonattendance of re-estimation
amid programming advancement which brings about colossal gap between starting
arrangement and ultimate result. Indeed, even with great assessment at first stage
the task supervisor must keep upgrade with venture advance and ought to have the
capacity to re-appraise the undertaking at a specific purpose of venture so as to re-
apportion the correct number of assets.
The target of this proposal paper was to check whether the earlier effort estima-
tion records can be utilized to foresee stage exertion with sensible exactness or not.
The acquired results uncovered that utilizing master estimation and parametric algo-
rithmic hypothesis lead to noteworthy change in stage-exertion estimation and give
venture administrator a developing picture about undertaking progress. Contrasting
our methodology and exponential relapse demonstrated that there is an extensive
potential in estimation precision.
As part of future plan, this thesis paper intend to expand this work to involve
some interesting features in each stage prediction and evaluate it on many datasets.
58
References
[1] Sayed Mohsin Reza, Md. Mahfujur Rahman, Md. Hasnat Parvez, M. Shamim
Kaiser and Shamim Al Mamun, Innovative Approach in Web Application Ef-
fort & Cost Estimation using Functional Measurement Type, 2nd International
Conference on Electrical Engineering and Information Communication Technol-
ogy (ICEEICT 2015), Jahangirnagar University, Dhaka, Bangladesh, May 2015.
[2] Sayed Mohsin Reza, Md. Mahfujur Rahman and Shamim Al Mamun New Ap-
proach for Road Networks - A Vehicle XML Device collaboration with Big Data.
1st International Conference on Electrical Engineering and Information Commu-
nication Technology (ICEEICT 2014), MIST, Dhaka, Bangladesh, April 2014.
[3] Sayed Mohsin Reza, Md. Mahfujur Rahman,Shamim Al Mamun and M Manjur
Mahmud, A New Approach of Big Data collaboration for Road Traffic Networks
considering Path Loss Analysis in context of Bangladesh. JU Journal Of Infor-
mation Technology (JUJIT), Jahangirnagar University,VOLUME 3, JUNE 2014.
[4] Abeer Hamdy, Fuzzy Logic for Enhancing the Sensitivity of COCOMO
Cost Model. Journal of Emerging Trends in Computing and Information Sci-
ences),VOL. 3, NO. 9, SEP 2012.
[5] Mithun Chandra Paul, Suman Sarkar, Md. Mahfujur Rahman, Sayed Mohsin
Reza and M. Shamim Kaiser, Low Cost and Portable Patient Monitoring System
for e-Health Services in Bangladesh. 2016 International Conference on Computer
Communication and Informatics (ICCCI-2016), Jan. 07 09, 2016, Coimbatore,
INDIA.
[6] Sayed Mohsin Reza and Khandoker Abdul Rahad, A Study on Network Security
Services with Cryptography and an Implementation of Vigenere- Multiplicative
Cipher. Proceedings of the Workshop on Information and Communication Tech-
nology,IUT, Dhaka, Bangladesh, October 2013.
[7] Vahid Khatibi, Dayang N. A. Jawawi Software Cost Estimation Methods:A Re-
view Journal of Emerging Trends in Computing and Information Sciences,VOL.2
NO. 1, 2010-11.
[8] Ali Bou Nassif and Luiz Fernando Capretz, Estimating Software Effort Based on
Use Case Point Model Using Sugeno Fuzzy Inference System, 23rd IEEE Inter-
national Conference on Tools with Artificial Intelligence, 2011.
59
[9] Erika Corona, Giulio Concas, Michele Marchesi, Giulio Barabino and Daniele
Grechi, Effort Estimation of Web Applications through Web CMF Objects, Sev-
enth International Conference on Software Process and Product Measurement
(IWSM-MENSURA), 2007.
[10] Wei Xia, Danny Ho, Luiz Fernando Capretz A Neuro-Fuzzy Model forFunction
Point Calibration. WSEAS TRANSACTIONS on INFORMATION SCIENCE &
APPLICATIONS, Issue 1, Volume 5, January 2008.
[11] A.J. Albrecht, Measuring application development productivity. Proceedings of
the Joint SHARE, GUIDE, and IBM Application Development Symposium.pp.
8392, Monterey, California,IBM Corporation, October 1979.
[12] Sandeep Kad, Vinay Chopra Fuzzy Logic based framework for Softwaren Devel-
opment Effort Estimation Research Cell: An International Journal of Engineering
Sciences,,Vol. 1, Issue July 2011,
[13] Erika Corona, Giulio Concas, Michele Marchesi, Giulio Barabino,Daniele Grechi
Effort Estimation of Web Applications through WebCMF Objects Seventh In-
ternational Conference on Software Process and Product Measurement (IWSM-
MENSURA),2007
[14] L. C. Briand and I. Wieczorek Resource Estimation in SoftwareEngineering En-
cyclopedia of Software Engineering,vol. 2, pp. 1160-1196, 2002.
[15] Lenandlar Singh Where are we? A Review of Effort and Cost Estimation Tech-
niques used for Web Applications Georgian Electronic Scientific Journal: Com-
puter Science and Telecommunications vol. 2009
[16] Emilia Mendes1, Ian Watson1, Chris Triggs2, Nile Mosley, Steve Counsell A
Comparative Study of Cost Estimation Models forWeb Hypermedia Applications.
Empirical Software Engineering June 2003.
[17] Ziauddin, Shahid Kamal, Shafiullah khan and Jamal Abdul Nasir, A Fuzzy Logic
Based Software Cost Estimation Model. International Journal of Software Engi-
neering and Its Applications,Vol. 7, No. 2,March, 2013.
[18] Vishal Sharma and Harsh Kumar Verma Optimized Fuzzy Logic Based Frame-
work for Effort Estimation in Software Development. IJCSI International Journal
of Computer Science Issues Vol. 7, Issue 2, No 2, March 2010.
[19] Preeth.R,Shivakumar.N,Balaji.N , Software Effort Estimation Using Attribute
Refinement based Adaptive Neuro Fuzzy Mode PIERS Proceedings,Czech Repub-
lic, 2007.
[20] Raul A. Gorcitz, Yesid Jarma, Promthe Spathis, Marcelo Dias de Amorim, Ryuji
Wakikawa, John Whitbeck, Vania Conan, and Serge Fdida, Vehicular Carriers for
Big Data Transfers IEEE Vehicular Networking Conference (VNC),Korea, 2012 .
60
[21] David Gundlegrd and Johan M Karlsson, Generating Road Traffic information
from Cellular Networks - New Possibilities in UMTS. 6thInternational Conference
on ITS Telecommunications,2006..
[22] Bandwidth (signal processing), Wikipedia, the free encyclopedia.
http://en.wikipedia.org/wiki/Bandwidth%28signal processing%29.
[23] Margaret Rouse, Data Science http://searccio.techtarget.com/definition/datascience,
September 2012.
[24] Robert Chansler, Hairong Kuang, Sanjay Radia, Konstantin
Shvachko and Suresh Srinivas, The Hadoop Distributed File System.
http://www.aosabook.org/en/hdfs.html,Ch. 9, pp.121-134.
[25] Statistical Data Analysis. http://reference.wolfram.com/language/
[26] Hadoop Distributed File System (HDFS) Architectural Documentation Use
Cases. http://itmvm.shidler.hawaii.edu/HDFS/ArchDocUseCases.html
[27] Mark Hall, Eibe Frank, Geoffrey Holmes, Bernhard Pfahringer, Peter Reute-
mann, Ian H. Witten, The WEKA Data Mining Software. ACM SIGKDD Explo-
rations Newsletter,Volume 11 Issue 1, June 2009.
[28] Rajender Bathla, Sukhdev Singh, Mittar Vishav & Rahul Gupta, Innovative
Scenario In Software Development Effort Estimation Based On A New Fuzzy
Logic Model. International Journal of Information Technology and Knowledge
Management,Volume 2, July-December 2010.
[29] ISBSG International Software Benchmarking standards Group, Data repository
release 10, http://www.isbsg.org, 2014.
[30] K. Molokken Ostvold, Project Estimation in the norwegian Software Industry A
Summary, 2004, Simula Research Laboratory.
61
Appendix A
List of Included Papers
1.1 List of Journals Organization
Note: (Numbers of papers identified in each paper in number format. The journals
with more than 5 papers in closed with [])
Advances in Information Systems 1
Annals of Software Engineering 3
Australian Journal of Information Systems 3
Automated Software Engineering 2
Empirical Software Engineering [12] - Rank 4
Computing and Control Engineering Journal 1
European Journal of Information Systems 1
Expert Systems 1
GEC Journal of Research 1
IEEE Proceedings Software Engineering 2
IEEE Transactions on Software Engineering 1 - Rank 1
IIE Transactions 1
Information Systems Journal 1
International Journal of Software Engineering and Knowledge Engineering
2
Journal of Computer and Software Engineering 1
Software - Practice and Experience 1
62
Software Engineering Journal 1
GEC Journal of Research 1
Software Engineering Notes 4
Software Quality Journal [9] Rank 6
Texas Instruments Technical Journal 1
Transactions of the Information Processing Society of Japan 1
1.2 Journal Papers on Software Effort Estimation
(Search completed September 2016. For a more updated list, that includes more
recently published journal papers, see https://sites.google.com/site/sayedmohsinreza/
education/masters-thesis)
1. Abran, A., I. Silva, et al. (2002). ”Field studies using functional size mea-
surement in building estimation models for software maintenance.” Journal of
Software Maintenance and Evolution: Research and practice 14(1): 31-64.
2. Aguilar-Ruiz, J. S., I. Ramos, et al. (2001). ”An evolutionary approach to esti-
mating software development projects.” Information and Software Technology
43(14): 875-882.
3. Ahn, Y., J. Suh, et al. (2003). ”The software maintenance project effort esti-
mation model based on function points.” Journal of Software Maintenance and
Evolution: Research and Practice 15(2): 71-85.
4. Angelis, L. and I. Stamelos (2000). ”A simulation tool for efficient analogy
based cost estimation.” Empirical Software Engineering 5(1): 35-68.
5. Antoniol, G., R. Fiutem, et al. (2003). ”Object-oriented function points: an
empirical validation.” Empirical Software Engineering 8(3): 225-254.
6. Barki, H., S. Rivard, et al. (2001). ”An integrative contingency model of soft-
ware project risk management.” Journal of Management Information Systems
17(4): 37-69.
7. Barry, E. J., T. Mukhopadhyay, et al. (2002). ”Software project duration
and effort: an empirical study.” Information Technology & Management 3(1-2):
113-136.
8. Benediktsson, O. and D. Dalcher (2003). ”Effort estimation in incremental
software development.” IEE Proceedings Software Engineering 150(6): 351-357.
9. Benediktsson, O., D. Dalcher, et al. (2003). ”COCOMO-Based Effort Esti-
mation for Iterative and Incremental Software Development.” Software Quality
Journal. 11(4): 265-281.
63
10. Briand, L. C. and J. Wst (2001). ”Modeling development effort in object-
oriented systems using design properties.” IEEE Transactions on Software En-
gineering 27(11): 963-986.
11. Burgess, C. J. and M. Lefley (2001). ”Can genetic programming improve soft-
ware effort estimation? A comparative evaluation.” Information and Software
Technology 43(14): 863-873.
64
Appendix B
Questionnaire
2.1 Questionnaire list was used in the survey
For more to know about the list,
see https://sites.google.com/site/sayedmohsinreza/education/masters-thesis)
Review Questionnaire Survey on Effort Estimation
•Name of the survey user
•Designation of the survey user
•Email of the survey user
•Which Estimation Process is best for Software Estimation?
•In which way, will it be a support of implementation of estimation approach?
•Do you think, will it be helpful for effort estimation being used as Project
Management Software as Expert Judgement?
•Do you think function point is relevant to Software Effort Estimation?
•Do you think time duration development is necessary as a factor of Software
effort estimation?
•Do you think dependency of another task will be a issue for effort estimation?
•Do you think Role of Sponsor can be a factor of Software Effort Estimation?
•Do you think Activity type (Programming / Supporting) Software development
can be factor of Software Estimation?
•Do you think No of Assigned people of a task can be count as issue in Software
Effort Estimation?
•Do you think No of tasks can be count as issue in Software Effort Estimation?
65
•Do you think Proxy Based Estimating can be count as a factor of Estimation
Process?
•Do you think task role(Project Manager, Developer) etc can be count as issue
in Software Effort Estimation?
•Do you think Job Position can be count as issue in Software Effort Estimation?
•Do you think Status of a task can be issue for Software Estimation?
•Do you think Changes of Company Policy affected Software Effort Estimation?
66
ABSTRACT
Activity based New Technique of Effort & Cost Estimation using Functional
Measurement Type for Web Application
by
Sayed Mohsin Reza
Supervisor: Dr. M. Shamim Kaiser
Software Effort Estimation helps Project Leaders to circulate assets, control spending
plans, motivation and create advanced works on, prompting tasks finished on time
and within budget related arrangement.
But the challenge is when a project leaders try to find out the software effort
based on some criteria, reasonable events and caused can be missed; while hopeful
expectations can be influenced to some asset losing. Because of fast change in in-
novation, usage of complex Software frameworks at less expense and the inclination
to keep up better quality software are a portion of the significant difficulties for the
Software organizations.
The limitation as well as the toughest works is effort estimation, in the field of
software engineering. It is the estimation of total effort required in developing soft-
ware. In terms of Software development, these issues are essential and extremely
difficult in Software Development with short timetables.
Since Software Development activities are constantly changed in nature, earlier
tasks may not cover all parts of new software development. The thesis motivation is
come from the Specialists in software engineering who have proposed different tech-
niques for effort & cost estimation. From this motivation, This thesis paper at first
tries to give an understanding into the different models and systems utilized as a part
of evaluating effort of the project. It additionally concentrates on a portion of the
pertinent reasons that cause incorrect estimation. Early Software Estimation models
depend on Regression Tree or Numerical Inferences.
In this thesis paper, The development process is also particularly followed by the
Expert based Estimation and Algorithmic based Estimation as the flow is controlled
by the System Administrator. This Thesis paper objective is to propose a way to
deal with build up the correctness of Software Effort Estimation utilizing Dataset of
a Functional Measurements and Algorithmic Guidelines. It also shows a model to
foresee the Parametric effort estimation. The advantages and disadvantages of the
current effort estimation process have been highlighted in this thesis paper. There
is as such not any single strategy which can be viewed as the best technique so in
this paper it is proposed that a mix of the strategies ought to be utilized to get an
estimated effort estimation.
